Dartmouth Engineering Research Activity

Cell-based protein purification systems offer advantages over conventional protein purification approaches which rely mostly on chromatographic methods to stepwise enrich for a desired protein of interest. We are developing approaches by which cells are engineered to produce their own affinity matrix to selectively sequester a desired recombinant protein. This allows for the expression and affinity purification of desired proteins in a single host, thereby obviating the need for external chromatographic purification.

Cellular engineering of protein expression hosts provides the ability to modify proteins in a site specific and controlled fashion—something increasingly important for the development of therapeutic proteins. We are developing methods by which cells are genetically engineered to incorporate sugars on a recombinant protein in a site-specific sequence dependent manner. Once a sugar is positioned on a given protein, conventional chemical modification such as PEGlylation can be used to further modify the protein and improve its therapeutic properties.

Plastic recycling is essential in both reducing the production of plastic waste and in reducing reliance on fossil fuels used as raw materials. However, plastics are notoriously difficult to recycle efficiently in part due to their wide variety of chemical makeups, properties, and additives. In addition, imperfections in recycling processes result in degradation of physical properties of recycled plastics, limiting the number of times a given material can be efficiently recycled. These limitations motivate the development of alternative plastic recycling technologies with improved efficiency and repeatability.

A recently discovered enzyme, PETase catalyzes the degradation of polyethylene terephthalate (PET). Enzymes capable of specific decomposition of plastics into monomers under mild conditions are an ideal solution to plastic recycling, as monomers can be reused to produce new plastics without the degradation in material properties associated with mechanical or pyrolytic recycling strategies. However, despite recent efforts to improve the catalytic efficiency of PETase, the best available variants still require improvement in order to better compete with traditional recycling technologies. This project will leverage molecular models of both steps of the catalytic mechanism of PETase to train a machine learning model that co-optimizes both steps together and directs simulations to obtain additional training data through active learning.

Connections Hypothesis Provider (CHP) service built by Dartmouth College (PI – Dr. Eugene Santos) and Tufts University (Co-PI – Joseph Gormley) in collaboration with the National Center for Advancing Translational Sciences (NCATS). CHP aims to leverage clinical data along with structured biochemical knowledge to derive a computational representation of pathway structures and molecular components to support human and machine-driven interpretation, enable pathway-based biomarker discovery, and aid in the drug development process. In its current version, CHP supports queries relating to genetic, therapeutic, and patient clinical features (e.g. tumor staging) contribution toward patient survival, as computed within the context of our test pilot: a robust breast cancer dataset from The Cancer Genome Atlas (TCGA). We are using this as a proving ground for our system’s basic operations as we work to incorporate structured pathway knowledge and pathway analysis methods into the tool.

A significant barrier to discovering contraceptives is the lack of systems for exploring the molecular features of ovarian follicles (an oocyte surrounded by somatic cells in the ovary), the drivers of ovulation, and for screening for contraceptive potential. My lab, in collaboration with a well-established network of ovarian collaborators will leverage these systems to enable contraceptive discovery. We will do this by identifying early signals of ovulation prior to tissue remodeling to identify targets that can preserve fertility. This project will involve substantial analyses and integration of transcriptomic data, with opportunities to develop rational frameworks for discovering druggable targets from these data types.

Goal: to provide an imaging tool to effectively monitor the long-term physiological effects

of deep space and automate diagnosis, enabling crew members to be proactive in the

event of injury. More specifically, we are designing an integrated US-EIT system and

demonstrate proof-of-concept of US-EIT for enhanced ultrasound imaging capabilities of

deep internal bleeding.

Roughly half of all U.S. carbon dioxide emissions are attributable to fossil fuels burned during industrial manufacturing and electricity production. Post-combustion carbon capture at the effluents of major emissions sources offers an essential tool for meeting urgent emissions reduction goals and mitigating climate catastrophe in the near term while alternatives to fossil fuels are further developed. Unfortunately, available technologies for addressing this need are expensive and inefficient: capture with organic solvents presents issues with corrosion and regeneration after capture, and membranes suffer from fouling and inconsistent performance under flue gas conditions.

One of the most promising alternative options is carbonic anhydrase (CA), an exceptionally efficient type of metalloenzyme that converts carbon dioxide and water to bicarbonate and hydrogen. Although CA vastly outperforms alternatives in terms of efficiency, carbon footprint, ease of captured carbon recovery, and regeneration after capture, it suffers from instability over long timescales in the presence of harsh conditions in flue gas columns, due to high temperature and salinity, alkaline pH, and other issues. Engineering of CA variants that remain highly stable and catalytically active under these conditions will open up a new avenue for efficient and cheap carbon capture. This project will use a machine learning model trained on molecular simulations to predict stabilizing mutations and identify regions of the protein to focus on.

Enzyme therapeutics are a potential means of addressing the emerging health care crisis resulting from drug resistant microbial pathogens. Efforts are focused on the redesign of antimicrobial proteins for enhanced bactericidal activity towards various clinically relevant targets. One facet of this work relates to complications associated with the genetic disease cystic fibrosis, and is being investigated in conjunction with the Cystic Fibrosis Foundation Research Development Program at The Geisel School of Medicine at Dartmouth.

Fermentation processes with high cell densities are important for the production of most bio-therapeutics which is conducted in highly controlled fed-batch processes. Commercial yeast- and E.coli-based fermentation processes often reach cell-densities in excess of 50g/l in fed-batch culture. Our laboratory has developed processes for the high cell density cultivation of Ralstonia eutropha allowing us to reach cell densities of over 150g/l and the expression of recombinant proteins at titers exceeding 10g/l. Much of this is achieved by implementing computer controlled feeding algorithms.

Fluorescence-guided surgery is important for the resection of some types of cancerous tumors where the tumor and normal tissue are similar in appearance and texture, and patient prognosis depends heavily on the completeness of resection. By selectively tagging tumor tissue with fluorescent dyes, it becomes possible to visually discriminate between normal and tumor tissues and improve significantly the completeness of tumor resection.

In large-scale industrial fermentations, it can be expensive to add oxygen, and to cool fermenters to mesophilic temperatures (20-45C). Using anaerobic thermophilic bacteria avoids both problems, however many genetic tools that have been originally developed for model organisms such as Saccharomyces cerevisiae and Escherichia coli do not work in these organisms. My group is working to develop several types of genetic tools necessary for domestication of thermophilic anaerobic bacteria, including:

• Ways to get foreign DNA into cells
• Tightly-controlled inducible promoters for reliable temporal control of gene expression
• Plasmid-based gene expression systems
• Chromosomal editing tools
• Ways to control the mutation rate

Glyco-engineering of proteins is being developed as a method to control the composition of glycans on glycoproteins. Such methods are of great importance in the biopharmaceutical industry because glyocoproteins constitute over 60% of all approved therapeutic proteins; and the therapeutic properties of many glycoproteins strongly depend on the composition of their glycans.

Humanization of glycosylation in yeast enables the use of yeast-based protein expression systems which offer inherent advantages over conventional mammalian cell culture. By engineering yeast-based systems to perform human-like glycosylation fully-humanized therapeutic proteins can be produced in these glyco-engineered hosts.

Immunogenicity prediction is an important means to control the safety of protein drugs, which can be recognized by the patient's immune system as foreign, inducing an adaptive immune response that can cause significant toxicity. We are developing approaches in which the immunogenicity of proteins can be assessed in vitro using protein engineering tools.

The interaction between the endocrine system and immune system remains poorly understood. The goal of this project is to use single-cell methods to better understand how hormones impact immune cell differentiation and function, focusing initially on innate immune cells. This will ultimately allow us to tailor immune responses based on hormonal contexts and sex differences, while enabling the creation of a more integrated and complete model of immune cell functions.

Polysaccharides are an essential class of functional biopolymers with wide-ranging roles, from structural support to signaling cascades and mediation of cell-cell interactions. In particular, short sugar chains known as oligosaccharides are essential in cell signaling and post-translational modification of proteins, and have applications as functional food additives with properties ranging from treatment of cancer and inflammation to promotion of gut microflora development in infants. Unfortunately, techniques for synthesizing oligosaccharides have lagged significantly behind those for other biopolymers.

Glycosynthases are engineered enzymes that present an attractive alternative to existing oligosaccharide synthesis pathways; however, these enzymes are too inefficient for industrial applications without further engineering. This project focuses on applying a novel molecular simulations approach to obtain training data for a machine learning model in order to accurately predict the effects of mutations on enzyme efficiency, helping to obtain improved variants.

We are developing algorithms and a complementary metal oxide semiconductor (CMOS) imager to perform high speed voltage imaging of brain activities.

The research lab at Dartmouth led by Professor Lee Lynd is active in research on the following topics:

• Microbial Cellulose Utilization, including fundamental and applied aspects
• Metabolic Engineering, focusing on thermophilic cellulolytic bacteria for fuel production
• Innovative Biomass Processing Technologies, including development, design, and evaluation
• Sustainable Bioenergy Futures, including resource, environment, and economic development

We approach these topics from a diversity of academic disciplines with molecular biology, microbiology, chemical/biochemical engineering providing the foundation for the first three. Consistent with the "Pasteur's Quadrant" model articulated by Donald Stokes (Brookings Institution Press, Washington, DC, 1997), we see advancing applied capability and increased fundamental understanding as having strong potential to be convergent and mutually-reinforcing, and we aspire to work in this mode.

A central theme of the Lynd group is processing cellulosic biomass in a single step without added enzymes. Such "consolidated bioprocessing" (CBP) is a potential breakthrough, and "is widely considered to be the ultimate low-cost configuration for cellulose hydrolysis and fermentation" (joint DOE/USDA Roadmap, 2007). We are focused on production of ethanol, a promising renewable fuel. The CBP strategy is however potentially applicable to a very broad range of fuels and chemicals.

Successful next generation AI/ML for real-world applications must be able to deal with incomplete, sparse and noisy data as well as unexpected or adversarial circumstances that might arise while solving real world military problems. Furthermore, successful models must be able to learn new concepts with few examples. Unfortunately, even in this age of commodity machine learning models, one still needs to train new models with a large amount of training examples to achieve the requisite performance from deep neural networks. The resulting models often lack robustness, explainability, and human-level intelligence.

One of the best models for intelligence is the one inspired by human brain itself, which can support robust and massive parallelism with ease. We are exploring two related aspects of human brain processing that will play a key role in a Third Wave AI revolution: (1) a meso-scale cortical computation that easily finds low dimensional manifolds where learning can naturally take place, and (2) recurrent module networks, as thin as one- or two-layer networks, akin to cellular automata, trained on time sequences rather than input-output pairs. We believe this process of simplification and automata implementation is a better model of the human neocortex than current state-of-the-art deep networks whose optimized architectures are often ad hoc and do not reflect biological reality. This project is being sponsored in part by DARPA.

Metabolism can be understood at many levels of aggregation from individual enzymes to the whole organism. An important intermediate level of aggregation is the metabolic pathway. Developing pathways that enable rapid production of desired compounds at high yield and titer requires a detailed understanding of both the individual components and the systems-level behavior that results from the interaction of these components, including:

• Identification of constituent enzymes in a pathway and the stoichiometry of the reactions they mediate
• Characterization of enzyme inhibition and regulation
• Development of screens and selections to improve properties of key enzymes.
• Protein engineering to increase activity, decrease inhibition, or change substrate specificity

There is a need to create versatile tools for exploring the interplay between peripheral, tissue and immune factors in tissue remodeling. To create such a platform in the context of reproductive health, this project will develop methods for culturing organoids derived from patient material, with an initial focus on organoids derived from progenitor cells isolated from mammary tissue and human breast milk (hBM). These systems will be used to better understand lactation and cancer development, and ultimately, inspire methods for engineering "true to natural" formula production.

Reinforcement learning has shown much success in games such as chess, backgammon, and Go. However, in most of these games, agents have full knowledge of the environment at all times. We describe a deep learning model that successfully maximizes its score using reinforcement learning in a game with incomplete and imperfect information. We apply our model to FlipIt (1), a two-player game in which both players, the attacker and the defender, compete for ownership of a shared resource and only receive information on the current state upon making a move. Our model is a deep neural network combined with Q-learning and is trained to maximize the defender's time of ownership of the resource. We extend FlipIt to a larger action-spaced game with the introduction of a new lower-cost move and generalize the model to multiplayer FlipIt.

(1) van Dijk, M, Juels, A, Oprea, A, Rivest, RL, FlipIt: The Game of "Stealthy Takeover." Journal of Cryptology, 26, 655–713 (2013)

To advance the science of decision-making as it pertains to how people learn to make decisions and how this process can be captured computationally, we are specifically addressing the challenge of how nonlinear decisions can be learned from data, experience, and even interactions with other decision-makers. Nonlinear thinking is a prized ability we, humans, have that is ubiquitously applied across any and all domains when the problems are challenging, and known solutions or ways of addressing the problems all fail to provide an adequate solution – e.g., All available choices are bad choices, must we settle for the least bad one? The ability to discover a new choice has been called being nonlinear, innovative, intuitive, emergent, or “outside-the-box.” It is well-documented that humans can often excel at such thinking in situations when there is a scarcity/overflow of data, significant uncertainty, and numerous contradictions in what is known or provided. However, how this can be replicated computationally for a machine has yet to be fully addressed or understood in extant research.

Novel protein expression systems are being developed based on the soil bacterium Ralstonia eutropha as an alternative to E.coli-based protein expression. This bacterial host can be grown to cell densities in excess of 150g/l in ultra high cell density culture and allow for the recovery of proteins that are prone to inclusion body formation in E.coli. Specific model proteins have been expressed at levels 100 fold higher than in E.coli thereby providing the impetus for further developing Ralstonia eutropha for the production of therapeutic proteins including monoclonal antibodies and peptides.

Optical molecular imaging is being used to provide molecular guidance in cancer surgery. Fluorescent contrast agents are in pre-clinical and clinical studies to image cancer tumors in vivo, with a dual focus, first on getting more accurate information out of the tissue, and secondly to provide better information about the specificity of the molecules as markers. Systems and algorithms for diffuse fluorescence imaging of tissue are studied, both as a stand-alone system, and as coupled to magnetic resonance imaging and computed tomography imaging. Tracer kinetic modeling is also being developed to allow quantitative imaging of molecular binding in vivo.

Maintaining immune tolerance during pregnancy is essential and evidence suggests that imbalances between T cell subsets can lead to fatal pregnancy events. This project, using existing and in-house generated single-cell datasets, will seek to better understand how peripheral immune cell features correlate with reproductive and uterine health, and in the long term, pregnancy outcomes or fertility.

Photodynamic therapy (PDT) is a newly emerging therapy for displastic tissues, such as cancer, age-related blindness, pre-malignant transformation or psoriasis. The therapy involves the administration of a photosensitizing agent, together with the application of moderate intensity light to active the molecules to produce local doses of singlet oxygen. Ongoing research topics include, developing improved dosimetry instrumentation and software, fluorescence tomography imaging to sense drug localization, and assaying unique tumor biology and treatment effects in experimental cancers.

Currently, most ethanol is produced using various strains of yeast. These organisms are very good at producing ethanol but have no native ability to consume lignocellulose. This has made it difficult to develop cost-effective yeast-based processes for lignocellulosic ethanol production. My group takes an alternative approach of starting with organisms that natively ferment lignocellulosic biomass and engineering them for efficient biofuel formation. To do this, we need to understand key aspects of the physiology of these native biomass-fermenting organisms, including:

• Which substrates they can consume, and how these substrates are used for growth, energy production, and product formation.
• Factors that limit growth and fermentation.
• Genetic adaptation to stresses associated with industrial fermentation

Protein engineering tools in immunology can be used as an unbiased and systematic means of identifying new targets, generating diagnostic and therapeutic reagents, and characterizing the basic biology of infectious disease.

We are developing a framework using advanced tools from random graph theory and spectral graph theory to carry out the quantitative analysis of the structure and dynamics of large networks—with the focus of graph merging and subgraph detection. This framework, using information theory as a demarcative tool, enables one to carry out analytic computations of observable network structures and capture the most relevant and refined quantities of real-world networks.

We are investigating how to combine recent advances in topological data analysis with recent advances in machine learning to enhance multiple hypothesis tracking system. Our goal is to improve detecting, clustering, classifying and tracking of various patterns-of-life trajectories by developing the capability to distinguish behavioral types at all scales.

Vaccine design to steer the immune response toward highly specific epitopes that are critical to pathogenic function represents a way to induce more effective antibodies. We are evaluating means to mask irrelevant, while stabilizing critical epitopes in order to direct the antibody response toward critical sites on viral and bacterial targets using yeast display technology.

Our asthma symptom monitoring solution is a wearable device that automatically detects and tracks cough, wheezing severity, lung function and inhaler use. The device provides objective information about the patient’s level of asthma control and level of compliance with the asthma plan.

Orthopedic implants are widely used to treat bone and joint disorders, such as fractures, osteoarthritis, and spinal deformities. However, conventional implant materials, such as stainless steel, titanium, and cobalt-chromium alloys, have several limitations: (i) they may cause adverse reactions and metal sensitivity due to their foreign ions and corrosion products; (ii) they may fail prematurely due to stress concentration and fatigue; (iii) they may interfere with bone remodeling and healing due to their mismatched mechanical properties with the host tissue; (iv) they can become a nidus for bacterial infection and biofilm colonization. Therefore, there is a need for novel implant materials that can overcome these challenges and improve the clinical outcomes of orthopedic surgery. Zinc is an attractive candidate for orthopedic implants because it is an essential trace element in the human body that plays a key role in bone metabolism and wound healing. Moreover, Zn is biodegradable and can be gradually resorbed by the body without leaving any permanent foreign material. The typical in vivo corrosion rates of unalloyed zinc in rats is 0.03 mm/yr, which is a useful rate for biodegradation. However, unalloyed Zn has very poor mechanical strength, which limits its application as an implant material. To address these issues, we are developing new Zn-based alloy that contains small amounts of silver, calcium, iron, magnesium, and manganese, and using a novel processing route. These alloying elements are chosen based on their beneficial effects on the biological and mechanical properties of Zn. Specifically, Ag has antibacterial activity and can reduce the risk of infection; Ca can promote bone formation and integration; Fe can enhance fracture fixation and blood compatibility; Mg can improve biocompatibility and corrosion resistance; Mn can increase ductility and strength. An adequate rate of implant degradation will allow the bone to heal properly before degrading and allowing the bone to support any loads.

Biomechanics analysis and monitoring following joint arthroplasty is valuable for achieving optimal recovery. Our laboratory has developed and implemented a novel method for monitoring continuous long term joint function using inertial measurement units (IMUs). Prospective studies are in progress to compare knee and shoulder function before and after arthroplasty. This data can be compared to a cohort of healthy individuals with no known joint arthropathy.

See more about biomechanics.

We are developing a device for continuous, non-invasive monitoring of a patient's cardiac hemodynamic status, allowing for physician intervention before the clinical symptoms of decompensation are observed, and potentially avoiding hospitalization.

Cell-based protein purification systems offer advantages over conventional protein purification approaches which rely mostly on chromatographic methods to stepwise enrich for a desired protein of interest. We are developing approaches by which cells are engineered to produce their own affinity matrix to selectively sequester a desired recombinant protein. This allows for the expression and affinity purification of desired proteins in a single host, thereby obviating the need for external chromatographic purification.

Cellular engineering of protein expression hosts provides the ability to modify proteins in a site specific and controlled fashion—something increasingly important for the development of therapeutic proteins. We are developing methods by which cells are genetically engineered to incorporate sugars on a recombinant protein in a site-specific sequence dependent manner. Once a sugar is positioned on a given protein, conventional chemical modification such as PEGlylation can be used to further modify the protein and improve its therapeutic properties.

Radiation therapy is used to treat cancer tumors by killing the tissue with high ionizing radiation doses. Until recently it has not been possible to image the radiation dose delivered to tissue, but through Cherenkov light imaging, this delivered dose can be mapped with high resolution cameras. The research group focuses on quantification of the imaging, and developing tools which allow radiation therapy to be delivered in a safer and more validated manner.

Clinical optical-electric probes are being developed for noninvasive simultaneous measurement of blood oxygenation and electrical potential changes associated with brain activity.

Combined ultrasound and electrical impedance tomography (EIT) puts 3-D ultrasound imaging together with EIT data in a co-registered volume. EIT relies on the mathematical processing of impedance data collected non-invasively from patients to reconstruct the 3-D distribution of the electrical properties of the tissues inside the patient. Combining ultrasound and EIT has the potential to greatly improve the quality and spatial resolution of the reconstructed electrical properties.

Ultrasound is a supplemental screening technique that has good sensitivity in dense breasts, is inexpensive, and is widely available, but unfortunately, it has high rates of false positives. Electrical impedance tomography (EIT) is a second attractive modality that is low-cost and has shown promise for cancer detection and in differentiating fibrocystic tissues from other tissues. Combining automated whole breast ultrasound (ABUS), which is a recent improvement to standard ultrasound, with EIT may significantly reduce the number of false positives of ABUS. If successful, the combined ABUS/EIT system could become an important screening technology for women with dense breasts.

Computational electromagnetics research is developing advanced analytical and numerical methods—such as the method of auxiliary sources, the method of moments, and pseudo spectral FDTD methods—for investigating high voltage non-linear electrostatic discharge phenomena as well as electromagnetic energy propagation in complex (Chiral and Bi-anizotropic) media.

A significant barrier to discovering contraceptives is the lack of systems for exploring the molecular features of ovarian follicles (an oocyte surrounded by somatic cells in the ovary), the drivers of ovulation, and for screening for contraceptive potential. My lab, in collaboration with a well-established network of ovarian collaborators will leverage these systems to enable contraceptive discovery. We will do this by identifying early signals of ovulation prior to tissue remodeling to identify targets that can preserve fertility. This project will involve substantial analyses and integration of transcriptomic data, with opportunities to develop rational frameworks for discovering druggable targets from these data types.

We are providing advanced integrated circuits (ICs) for sensing to Professor Ryan Halter, who is combining them with transrectal electrical impedance tomography and electrical impedance sensing biopsy devices to improve the accuracy of ultrasound-guided prostate biopsy procedures.

Dielectric properties of tissue—measured through advanced microwave imaging techniques—convey functional information useful for making clinical diagnoses. The properties reflect tissue composition of fat, bone, water, proteins, etc., and often have unique spectral characteristics. The relative proportions and dynamic aspects of these constituents can have important implications for breast cancer imaging, osteoporosis detection, brain imaging, and heat therapy monitoring.

Electrical bioimpedance measurements of tissue provide significant levels of contrast between benign and malignant pathologies due to the vastly different morphologies occurring between tissue types. Focal sensing or mapping of these properties can provide clinicians useful information regarding the extent and severity of diseases like cancer. Our group is currently developing technologies to couple bioimpedance sensors to clinical devices including: 1) intraoperative instruments for use in assessing surgical margins during tumor resection; and 2) standard biopsy needles for use in providing real-time pathological assessment of tissue.

Electrical impedance imaging for breast cancer screening is the process of imaging the electrical property (conductivity and permitivity) of tissue using electrodes located on the body surface. This project is one branch of the larger effort to develop innovative technologies for breast cancer detection.

Electrical impedance imaging for prostate cancer screening is the process of imaging non-invasively the electrical properties (conductivity and permittivity) of the prostate and its vicinity using electrodes mounted onto an intracavitary probe.

Goal: to provide an imaging tool to effectively monitor the long-term physiological effects

of deep space and automate diagnosis, enabling crew members to be proactive in the

event of injury. More specifically, we are designing an integrated US-EIT system and

demonstrate proof-of-concept of US-EIT for enhanced ultrasound imaging capabilities of

deep internal bleeding.

The most common application, so far, of electrical impedance tomography (EIT) has been pulmonary monitoring of patients. In particular, we have been pursuing EIT for cardiac output monitoring (a related application) and EIT as a surrogate for pulmonary function tests.

Throat cancers have been increasing in incidence worldwide. Despite advances in surgical and non-surgical management of these cancers, treatment continues to be associated with significant functional and cosmetic morbidity. More minimally invasive trans-oral surgical (TOS) approaches have reduced treatment morbidity and complications. However, one drawback of TOS is the difficulty in intraoperatively assessing tumor extent and locating major vascular structures. Surgical navigation with image guidance has shown improved safety and efficacy with other surgical procedures; however it is currently not feasible in TOS due to the soft tissue and airway deformation that occurs with placement of instruments needed to access the throat, thus rendering preoperative scans unusable in the intraoperative setting.

The overarching objective of our research is to develop enabling technologies that allow for surgical navigation with image guidance for TOS. Our research strategy is to acquire intraoperative imaging during TOS in order to develop models of upper aerodigestive tract deformation that reflect the intraoperative state. This, in turn, would allow for registration of preoperative images to the intraoperative state. We are well equipped to solve this problem due to the unique intraoperative CT and MR imaging resources available at the Dartmouth Center for Surgical Innovation. We have successfully developed a 3D printable polymer laryngoscopy system which, unlike standard metal laryngoscopes, is CT and MRI compatible. We have also acquired preliminary intraoperative imaging data during laryngoscopy procedures. The next steps in this research will be to further quantify and characterize tissue deformation that occurs during TOS and ultimately develop a surgical navigation platform for trans-oral procedures.

Enzyme therapeutics are a potential means of addressing the emerging health care crisis resulting from drug resistant microbial pathogens. Efforts are focused on the redesign of antimicrobial proteins for enhanced bactericidal activity towards various clinically relevant targets. One facet of this work relates to complications associated with the genetic disease cystic fibrosis, and is being investigated in conjunction with the Cystic Fibrosis Foundation Research Development Program at The Geisel School of Medicine at Dartmouth.

Fermentation processes with high cell densities are important for the production of most bio-therapeutics which is conducted in highly controlled fed-batch processes. Commercial yeast- and E.coli-based fermentation processes often reach cell-densities in excess of 50g/l in fed-batch culture. Our laboratory has developed processes for the high cell density cultivation of Ralstonia eutropha allowing us to reach cell densities of over 150g/l and the expression of recombinant proteins at titers exceeding 10g/l. Much of this is achieved by implementing computer controlled feeding algorithms.

Fluorescence-guided surgery is important for the resection of some types of cancerous tumors where the tumor and normal tissue are similar in appearance and texture, and patient prognosis depends heavily on the completeness of resection. By selectively tagging tumor tissue with fluorescent dyes, it becomes possible to visually discriminate between normal and tumor tissues and improve significantly the completeness of tumor resection.

Glyco-engineering of proteins is being developed as a method to control the composition of glycans on glycoproteins. Such methods are of great importance in the biopharmaceutical industry because glyocoproteins constitute over 60% of all approved therapeutic proteins; and the therapeutic properties of many glycoproteins strongly depend on the composition of their glycans.

Soft magnets play a vital role in efficient energy conversion in a variety of important applications and industries including wide-bandgap semiconductors, electric vehicles, aeronautics, and aerospace, particularly at high temperatures. Improving the efficiency of modern power electronics and electrical machines via advanced soft magnets has the potential to significantly contribute to global energy savings, thereby leading to a reduction of the associated carbon footprint. In this project, we are working on two novel FeCoMnAl high-entropy alloy (HEA) soft magnets, one of which is single-phase B2 (Fe₃₀Co₄₀Mn₁₅Al₁₅) and the other consists of an ordered B2-phase matrix enriched with Co/Al and uniformly distributed BCC nanoprecipitates enriched with Fe/Mn (Fe₄₀Co₃₀Mn₁₅Al₁₅). The two HEAs show similar properties, viz., a high saturation magnetization of 158-162 Am² kg^-1, a high Curie temperature of 1020-1081 K, a low coercivity of 108-114 A m^-1, a high electrical resistivity of ~230 µΩ cm, and good thermal stability. We are processing these HEAs using both a powder metallurgy route and via additive manufacturing. The magnetic properties and microstructures of the resulting materials are being examined using combination of a VSM, TEM, SEM and XRD examinations.

This project is investigating the synergy between the excellent mechanical behavior of FeMnNiAlCr high entropy alloys (HEAs) and the high solar absorptance of their native thermal oxides for high efficiency concentrated solar thermal power (CSP) systems working at >700^oC. The alloy itself would be used in high-temperature tubing to carry molten salts or supercritical carbon dioxide (sCO2), while the native oxide would act as a high-efficiency solar thermal absorber. The oxide layer is also dense and protective against oxidation (parabolic growth kinetics) at 750 °C. This new Fe-Mn based HEA system has already demonstrated a higher tensile strength and ductility than more expensive Ni-based Inconel 740 superalloys at both room temperature (with carbon doping) and 750^oC (with Ti doping), and their native thermal oxides have achieved a high optical-to-thermal conversion efficiency of ηtherm=90.8% at 750°C under 1000x solar concentration ratio. In preliminary corrosion studies, two-phase Cr-modified HEAs have sustained bromide molten salts for 14 days at 750°C with <2% weight loss. The project has close collaborations with the Oak Ridge National Laboratory (ORNL) and the Ames Laboratory in computational materials science and atom probe tomography (APT) to understand the fundamental structure-property relationship in FeMnNiAlCr HEAs.

Funded by DOE.

Humanization of glycosylation in yeast enables the use of yeast-based protein expression systems which offer inherent advantages over conventional mammalian cell culture. By engineering yeast-based systems to perform human-like glycosylation fully-humanized therapeutic proteins can be produced in these glyco-engineered hosts.

This project develops and evaluates image-based registration methods involving stereovision for registration and image updating during open spine surgeries for degenerative spondylolisthesis.

Image-guided neurosurgery gives the surgeon the ability to track instruments in reference to subsurface anatomical structures. Using clinical brain displacement data, a computational technique is being developed to model the brain deformation that typically occurs during neurosurgery. The resulting deformation predictions are then used to update the patient's preoperative magnetic resonance images seen by the surgeon during the procedure.

Assessing muscle health is important diagnostic information in the treatment of numerous peripheral nerve diseases. This project aims to incorporate spatial information from ultrasound into the inverse problem of electrical impedance tomography in order to accurately image the electrical properties of the muscle (under the skin and adipose tissue).

Immunogenicity prediction is an important means to control the safety of protein drugs, which can be recognized by the patient's immune system as foreign, inducing an adaptive immune response that can cause significant toxicity. We are developing approaches in which the immunogenicity of proteins can be assessed in vitro using protein engineering tools.

The interaction between the endocrine system and immune system remains poorly understood. The goal of this project is to use single-cell methods to better understand how hormones impact immune cell differentiation and function, focusing initially on innate immune cells. This will ultimately allow us to tailor immune responses based on hormonal contexts and sex differences, while enabling the creation of a more integrated and complete model of immune cell functions.

We are developing algorithms and a complementary metal oxide semiconductor (CMOS) imager to perform high speed voltage imaging of brain activities.

The Empower Lab creates digital therapeutics and intervention technologies, with an emphasis on inclusive design, patient experience, and health equity.

We seek students to contribute to the design, development, and evaluation of our interactive tools, which aim to positively shape behaviors and health outcomes through an empowerment approach. Our work employs human-centered methods and explores a variety of form factors and interaction paradigms (e.g., AR/VR/XR, games, toys, tangibles, musical interfaces, narratives, psychotherapeutic visualization, socially assistive robots, and more).

We work closely with stakeholder populations on a range of health topics; key areas of focus include mental health, women’s health, physical mobility/pain, aging-in-place, and pediatrics/childhood development. More details about our lab and its active projects, including specific recruiting opportunities, can be found here: empowerlab.dartmouth.edu/projects

Generally speaking, our collaborative research style integrates interdisciplinary skills from both computing and engineering as well as the social sciences and humanities. Students typically focus on contributing to one or two aspects of a project’s research activities, which include iterative design (UX, prototyping), implementation (software programming, physical fabrication), data analysis (quantitative/statistical or qualitative), stakeholder engagement (conducting interviews, observational or ethnographic work), and running studies (user testing, lab experiments, online or survey studies, or field trials).

If interested in getting involved, you can fill out our application form dartgo.org/empowerlab-apply or email the lab director, Prof. Liz Murnane (emurnane@dartmouth.edu) to share your resume, relevant interests and background, and any questions you might have. We look forward to hearing from you!

Material behavior of medical grade ultra-high molecular weight polyethylene (UHMWPE) was identified as a serious concern as it limits the overall lifetime and success of a joint replacement. Although total joint arthroplasty involving UHMWPE as a bearing surface has been one of the most successful procedures of the last century, issues of wear, oxidation, and fatigue failure remain obstacles to the longevity of joint replacements.

See more about UHMWPE material behavior.

Bearing function of retrieved knee devices sent to us by orthopaedic surgeons are assessed for damage, and also quantitatively assessed for wear. Dimensions of retrievals are compared to design specifications or shorter in-vivo duration devices to calculate both articular and backside wear. Wear and wear rate are correlated with variables including polyethylene pedigree, articular bearing geometry, device fixation, and patient factors.

Current work also includes examination of a series of reverse and total shoulders to determine the incidence of abrasive and adhesive wear and determine typical locations for these wear patterns on polyethylene components.

See more about knee/shoulder implant bearing function.

Label free genome sequencing is an advancing technology to "read" the sequence in a single DNA molecule in a massively-parallel fashion. The technology combines concepts of single nucleotide addition (SNA) sequencing, near field optics, single molecule force spectroscopy, and microfluidics. This work is performed in collaboration with Professor Dmitri Vezonov at Lehigh University.

Magnetic nanoparticle imaging is being developed to meet the needs of translational research on the biodistribution of magnetic nanoparticles (MNPs). Emerging nanotechnology platforms promise to deliver new tools to detect, monitor and treat cancer. These nanotechnology platforms offer a future of personalized medicine where a nanocarrier can be targeted to specific cancer cells, carry a drug payload, be remotely activated at a specific location in the body or upon targeted binding to selected cell types, imaged in-real time, and monitored as therapy progresses. Among the many nanocarriers in development, those utilizing MNPs are ideally suited for translational research because of their long history in biomedical research and many practical applications. We have developed an MNP imaging method called nonlinear susceptibility magnitude imaging (nSMI). Our imaging system has the potential for broad use in translational MNP research because of its functional and cost-efficient design.

This project provides quantitative data analysis, physical modeling, and numerical methods support to the Dartmouth Center of Cancer Nanotechnology Excellence. The project employs physical modeling and numerical methods to understand the electromagnetic interactions that occur when magnetic nanoparticles are placed in an alternating magnetic field in biological environments, and investigates the impact that biological parameters (e.g., blood flow) have on the ability to increase tumor temperatures locally.

Magnetic resonance elastography is being developed as a technique to measure the elasticity of tissue in vivo by gently shaking the tissue in a magnetic resonance imager. The displacements measured are used to determine tissue mechanical properties which can help identify and classify breast lesions. See also Discovering at Thayer School.

Successful next generation AI/ML for real-world applications must be able to deal with incomplete, sparse and noisy data as well as unexpected or adversarial circumstances that might arise while solving real world military problems. Furthermore, successful models must be able to learn new concepts with few examples. Unfortunately, even in this age of commodity machine learning models, one still needs to train new models with a large amount of training examples to achieve the requisite performance from deep neural networks. The resulting models often lack robustness, explainability, and human-level intelligence.

One of the best models for intelligence is the one inspired by human brain itself, which can support robust and massive parallelism with ease. We are exploring two related aspects of human brain processing that will play a key role in a Third Wave AI revolution: (1) a meso-scale cortical computation that easily finds low dimensional manifolds where learning can naturally take place, and (2) recurrent module networks, as thin as one- or two-layer networks, akin to cellular automata, trained on time sequences rather than input-output pairs. We believe this process of simplification and automata implementation is a better model of the human neocortex than current state-of-the-art deep networks whose optimized architectures are often ad hoc and do not reflect biological reality. This project is being sponsored in part by DARPA.

At the core of novel wearable and implantable devices are cutting-edge sensing technologies that can be integrated with the body. In this project, we leverage MEMS, micro/nanofluidics, and other relevant technologies to develop advanced biophysical and biochemical sensors. One example of such efforts is nanoelectrokinetics-based devices for ultra-sensitive protein biomarker detection through electrokinetic enrichment of biomolecules (Lab Chip 2017, Anal. Chem. 2018, Nanoscale 2018, PNAS 2019, Angew. Chem. 2020, etc.). https://www.pnas.org/doi/abs/1...

Microwave electronics capable of fast data acquisition (approaching real-time) are being developed for brain imaging applications. Also, site-specific antenna arrays for microwave imaging are in development for heel imaging (screening for osteoporosis) and for brain imaging applications.

Microwave imaging spectroscopy (also commonly referred to as microwave computed tomography) presents the challenge of measuring the signal data necessary to produce meaningful images. Development of site-specific antenna arrays along with improved electronic signal detection technology is rapidly making this measurement feasible for breast cancer detection. A second-generation tomographic breast imaging system has been completed and the data are being used to recover permittivity and conductivity maps of the breast for evaluation by a clinician.

There is a need to create versatile tools for exploring the interplay between peripheral, tissue and immune factors in tissue remodeling. To create such a platform in the context of reproductive health, this project will develop methods for culturing organoids derived from patient material, with an initial focus on organoids derived from progenitor cells isolated from mammary tissue and human breast milk (hBM). These systems will be used to better understand lactation and cancer development, and ultimately, inspire methods for engineering "true to natural" formula production.

Near-infrared imaging (NIR) provides a way to quantify blood and water concentrations in tissue, as well as structural and functional parameters. Since normal tissue, benign tumors, and malignant tumors each carry different concentrations of both hemoglobin and water, and have different levels of oxygen demand and ultrastructural scattering, NIR spectroscopy can be combined into standard imaging systems as an effective method of to provide additional information for breast cancer detection and diagnosis. Work is ongoing to improve techniques for better image reconstruction, display and integration with magnetic resonance imaging (MRI) and computed tomography (CT) imaging.

See also Center for Imaging Medicine

Neurovascular coupling refers to the mechanisms that relate evoked neural activity to localized responses by the cerebral vasculature. Better models of this coupling are needed to improve the interpretation of neuroimaging studies and understanding of neurodegenerative disease.

New devices research and developent for total joint arthroplasty includes:

• Testing of a new bi-material bearing for a total hip arthroplasty (THA) device against conventional bearing designs to compare levels of bearing surface damage and wear;
• Development of an intraoperative method for quantifying the orientation of prosthetic components used in total knee arthroplasty (TKA) that is efficient, easy to use, cost effective, and quick with respect to total surgical time.

See more about new devices.

New materials research and development for orthopaedic implants includes:

• Evaluation of equal channel angular extrusion (ECAE)-processed ultra-high molecular weight polyethylene (UHMWPE) for joint arthroplasty and industrial applications;
• Investigation of off-label use of a resorbable calcium sulfate antibiotic carrier in single stage and two-stage procedures to determine the potential of this use to change damage patterns or wear rates of artificial joints.

See more about new materials.

Reinforcement learning has shown much success in games such as chess, backgammon, and Go. However, in most of these games, agents have full knowledge of the environment at all times. We describe a deep learning model that successfully maximizes its score using reinforcement learning in a game with incomplete and imperfect information. We apply our model to FlipIt (1), a two-player game in which both players, the attacker and the defender, compete for ownership of a shared resource and only receive information on the current state upon making a move. Our model is a deep neural network combined with Q-learning and is trained to maximize the defender's time of ownership of the resource. We extend FlipIt to a larger action-spaced game with the introduction of a new lower-cost move and generalize the model to multiplayer FlipIt.

(1) van Dijk, M, Juels, A, Oprea, A, Rivest, RL, FlipIt: The Game of "Stealthy Takeover." Journal of Cryptology, 26, 655–713 (2013)

Non-linear image reconstruction techniques is at the core of the medical imaging projects. Excitation-induced measurements from each instrument are compared with calculations from corresponding numerical models to compute updated property images of the biological target. As the images are progressively updated (or refined) in a non-linear iterative process, important features and functional information related to the objects physiological status—tumor, benign tissue, etc.—become more apparent. The computational core of the breast imaging project works synergistically with all four groups to improve our fundamental understanding of these mathematical systems to improve overall image quality and resolution. These processes have been developed for both 2D and 3D geometries in each modality and are being expanded to exploit emerging parallel computing capabilities.

Novel protein expression systems are being developed based on the soil bacterium Ralstonia eutropha as an alternative to E.coli-based protein expression. This bacterial host can be grown to cell densities in excess of 150g/l in ultra high cell density culture and allow for the recovery of proteins that are prone to inclusion body formation in E.coli. Specific model proteins have been expressed at levels 100 fold higher than in E.coli thereby providing the impetus for further developing Ralstonia eutropha for the production of therapeutic proteins including monoclonal antibodies and peptides.

The microstructual evolution of snow under a temperature gradient has been of interest for many years since this can lead to persistent weak layers, which are possible microstructural causes of avalanches. Ice crusts can form on top or within a snowpack from a variety of meteorological conditions including significant melt/freeze or freezing rain events, and once buried, they can persist throughout the entire winter season and act as an ideal sliding surface for dangerous slab avalanches in seasonal mountain snowpacks. Both of these phenomena are important because the number of fatalities from avalanches in the US has increased annually since the 1970s. Avalanches can also have substantial economic impacts due to road closures, the costs of rescue and building damage, and, with continued global warning, more avalanches are expected in Arctic regions. To understand avalanche nucleation, we are deforming two types of specimens (heterogeneously-layered snow and snow containing an ice lens) in a micro CT located in a cold room, in which the specimens are repeatedly imaged during loading. We are also performing more macroscopic deformation experiments on larger samples at both different rates and different temperatures, which are imaged using a high-speed video camera during loading. The final deformed microstructures in both cases are imaged at high resolution using a scanning electron microscope, which provides information on both the effects of crystal orientation on deformation while clearly delineating one ice crystal orientation from another. Based on the experimental observations, a multiscale computational model is being built to understand crack initiation/crack propagation as well as the deformation mechanisms in heterogenously-layered snow samples containing persistent weak layers and ice/snow interfaces.

Optical molecular imaging is being used to provide molecular guidance in cancer surgery. Fluorescent contrast agents are in pre-clinical and clinical studies to image cancer tumors in vivo, with a dual focus, first on getting more accurate information out of the tissue, and secondly to provide better information about the specificity of the molecules as markers. Systems and algorithms for diffuse fluorescence imaging of tissue are studied, both as a stand-alone system, and as coupled to magnetic resonance imaging and computed tomography imaging. Tracer kinetic modeling is also being developed to allow quantitative imaging of molecular binding in vivo.

Implant failure analysis in the Dartmouth Biomedical Engineering Center for Orthopaedics is ongoing and plays a key role in identifying failure modes and relating them to various designs and materials being used in the industry. In fact, in 2000, NIH's Consensus Development Program produced a technology assessment statement acknowledging the value of implant retrieval programs:

• Implant retrieval and analysis is of critical importance in the process of improving care of patients in need of implants.
• Attention needs to be directed toward reducing various obstacles to implant retrieval and analysis, particularly legal and economic disincentives.
• The failure to appreciate the value of implant retrieval and analysis is a serious impediment to research in devices. A focused educational program will provide the information necessary for improving the quality of future devices.

See more about implant failure analysis.

Maintaining immune tolerance during pregnancy is essential and evidence suggests that imbalances between T cell subsets can lead to fatal pregnancy events. This project, using existing and in-house generated single-cell datasets, will seek to better understand how peripheral immune cell features correlate with reproductive and uterine health, and in the long term, pregnancy outcomes or fertility.

Photodynamic therapy (PDT) is a newly emerging therapy for displastic tissues, such as cancer, age-related blindness, pre-malignant transformation or psoriasis. The therapy involves the administration of a photosensitizing agent, together with the application of moderate intensity light to active the molecules to produce local doses of singlet oxygen. Ongoing research topics include, developing improved dosimetry instrumentation and software, fluorescence tomography imaging to sense drug localization, and assaying unique tumor biology and treatment effects in experimental cancers.

Porous thermoelectric cells (TECs) are being developed for waste heat recovery. This project seeks to convert waste thermal energy directly into electricity potentially increasing overall energy efficiency by 15-20% and providing new portable electric power sources, particularly for cold regions. The project is focusing on low cost, nanostructure-engineered TEC (NETECs) materials based on earth-abundant, highly machinable metallic alloys and intermetallic compounds by engineering the grain sizes, second phase precipitation, and nanopores in transition metal intermetallic compounds. The project is currently focusing on the compound Fe2AlV. The location of quaternary atoms in the lattice is being determined by the TEM-based technique ALCHEMI.

Funded by USA-CRREL

The goal of this project is to optimize and validate, in human and animal studies, intraoperative image updating methods that assimilate intraoperative data acquired with stereovision and ultrasound with computational modeling to update preoperative MRI in order to maintain image registration accuracy throughout surgery.

Protein engineering tools in immunology can be used as an unbiased and systematic means of identifying new targets, generating diagnostic and therapeutic reagents, and characterizing the basic biology of infectious disease.

Quantitative scatter imaging makes use of the fact that normal functioning cells and cancerous cells show differences both in the components within the cells and in the structural organization of the cells within the tissue. These physical distinctions in biological structure have been shown to scatter light differently. This study develops a new approach to imaging scatter-based contrast over a wide field of view in tissue using high frequency structured light. These scattering features may provide a method to diagnostically identify abnormal tissue without the need to administer targeted compounds. This approach has the potential to generate new diagnostic screenings and new approaches for surgical guidance.

Radiation therapy is used to treat cancer tumors by killing the tissue with high ionizing radiation doses. Modern external beam radiotherapy systems deliver high dose levels to precisely marked tumor volume in less time. As a mis-administration can have potentially severe impact to the surrounding healthy tissue, more stringent and complex quality assurance measurements are required in clinics. By developing a comprehensive optical dose imaging camera system, we aim to fundamentally simplify the quality assurance process and, in turn, to further promote the culture of safety in radiotherapy. By converting the dose to visible light using scintillation phantom, we can image and reconstruct 3D dose maps in real time, enabling complete and accurate verification in a fast enough timeframe for it to be useful in every procedure.

We are developing a framework using advanced tools from random graph theory and spectral graph theory to carry out the quantitative analysis of the structure and dynamics of large networks—with the focus of graph merging and subgraph detection. This framework, using information theory as a demarcative tool, enables one to carry out analytic computations of observable network structures and capture the most relevant and refined quantities of real-world networks.

The resolution of ultrasound imaging is typically determined by the wavelength of sound. We are circumventing this limitation with a new class of phase change nanoparticle contrast agents, called laser-activated nanodetectors (LANDs). These contrast agents, which consist of a liquid perfluorocarbon core and encapsulated dye, undergo vaporization upon exposure to pulsed laser irradiation. Several milliseconds later, the LANDs recondense back to their stable liquid state. While in their gaseous state, the LANDs provide high contrast in ultrasound images. The end result is a triggerable "blinking" contrast agent. We are currently developing algorithms to harness the blinking to improve the resolution of the imaging system as well as probe the tissue properties.

Therapy monitoring is an important emerging application of imaging modalities. These and other current research topics include:

• near-infrared imaging of brain tissue;
• near-infrared spectroscopy for diagnosing peripheral vascular disease;
• electrical impedance spectroscopy for radiation therapy monitoring;
• magnetic resonance elastography for detecting brain or prostate lesions; to follow the progression of diabetic damage in the foot; and to answer basic questions of wave propagation in tissue;
• microwave imaging spectroscopy for hyperthermia therapy monitoring, brain imaging, and detection of early-stage osteoporosis;
• electrical impedance tomography for monitoring traumatic brain injury progression and therapy.

We are investigating how to combine recent advances in topological data analysis with recent advances in machine learning to enhance multiple hypothesis tracking system. Our goal is to improve detecting, clustering, classifying and tracking of various patterns-of-life trajectories by developing the capability to distinguish behavioral types at all scales.

Photoacoustic imaging is a hybrid technique which relies on a pulsed laser to generate acoustic waves within tissue. It is capable of achieving high resolution images with optical contrast tens of millimeters deep in tissue. We are working to translate this technology to the clinic with an initial focus on identifying metastatic lymph nodes in breast cancer patients. This project involves the development of the imaging system, creation of spectroscopic imaging algorithms, and combining the two for imaging in the clinic.

Unexploded ordnance (UXO) detection and discrimination approaches are being developed to solve the Department of Defense's (DoD) most pressing environmental problems: UXO cleanup and humanitarian de-mining. The program combines advanced forward and inverse EM sensing approaches with statistical signal processing methodologies to solve these complex and challenging problems. See also UXO Research Group.

Demand for high-performance permanent magnets for motors is increasing rapidly for applications such as wind turbine generators and motors in both electric and hybrid cars. Samarium-Cobalt and Neodymium-Iron-Boron Rare Earth magnets are generally used for such challenging applications. While Rare Earth magnets are the best currently available permanent magnets, they are not without problems such as being brittle, suffering from thermal shock, and experiencing corrosion. Further, over 95% of Rare Earths are produced in China and there has been substantial price volatility. Finally, Rare Earth mining has been associated with severe environmental degradation and large energy usage. NiFe, which has been identified in meteorites as the compound Tetrataenite where it transformed from the high temperature (disordered) f.c.c. phase over thousands of years, has magnetic properties comparable to that of Rare Earth magnets. This research award develops new, low-cost, environmentally-friendly NiFe-based permanent magnets, using both quantum-mechanical calculations to predict the effects of alloying with other elements coupled with experiments to verify the effects of these additional elements on the transformation kinetics and magnetic properties. It uses the novel approach of pulsed electrical heating, which has been shown to accelerate transformations. Both women and under-represented minorities will be engaged in the research. The development of novel NiFe magnets will enable production of permanent magnets to be relocated to the USA. As part of the work, the project develops a web site that offers simple virtual experiments to explain to a wide audience magnetism and the materials science of permanent magnets.

The L1₀-structured compound Nickel-Iron (NiFe) has the potential to replace Rare Earth (RE) magnets at low cost: NiFe has a magnetic anisotropy energy, k_u, of 1.3 x 10⁶ J.m^-3 and a saturation magnetization m₀M_S of 1.59 Tesla, which is comparable to that of Nd₂Fe₁₄B. In addition, it has good corrosion resistance. The challenge is that the binary L1₀ compound has a very low transformation temperature from the high-temperature f.c.c. phase of about 320^oC that forms on casting and, thus, orders very slowly at temperatures where it is stable. This project combines ab initio quantum mechanical calculations and experimental work to design new L1₀-structured NiFe magnets with ternary elemental additions. These ternary compounds potentially have a significantly higher f.c.c.-to-L1₀ transformation temperatures and higher diffusivities than binary NiFe, but have similar saturation magnetizations. Thus, the L1₀ phase can be produced at higher temperature in short, commercially-viable times utilizing electro-pulse annealing of cold-worked material, which has also been shown to dramatically accelerate recrystallization in NiFe. The TEM-based technique ALCHEMI is used to determine the atom site locations of these ternary additions in the L1₀ unit cell. This work demonstrates a practical paradigm for designing magnets and leads to new commercially-viable permanent magnet. Commercially, NiFe can be manufactured by continuous electro-pulse annealing of rolls of sheet material or of rods. NiFe is very ductile and can easily be machined into various shapes.

Vaccine design to steer the immune response toward highly specific epitopes that are critical to pathogenic function represents a way to induce more effective antibodies. We are evaluating means to mask irrelevant, while stabilizing critical epitopes in order to direct the antibody response toward critical sites on viral and bacterial targets using yeast display technology.

Capabilities in continuous monitoring of key physiological parameters of disease have never been more important than in the context of the global COVID-19 pandemic. Soft, skin-mounted electronics that incorporate high-bandwidth, miniaturized motion sensors enable digital, wireless measurements of mechanoacoustic (MA) signatures of both core vital signs (heart rate, respiratory rate, and temperature) and underexplored biomarkers (coughing count) with high fidelity and immunity to ambient noises. This paper summarizes an effort that integrates such MA sensors with a cloud data infrastructure and a set of analytics approaches based on digital filtering and convolutional neural networks for monitoring of COVID-19 infections in sick and healthy individuals in the hospital and the home. The sensors, deployed on COVID-19 patients along with healthy controls in both inpatient and home settings, record coughing frequency and intensity continuously, along with a collection of other biometrics. The methodology creates opportunities to study patterns in biometrics across individuals and among different demographic groups. https://www.pnas.org/doi/abs/1...

Fully implantable wireless systems for the recording and modulation of neural circuits that do not require physical tethers or batteries allow for studies that demand the use of unconstrained and freely behaving animals in isolation or in social groups. Moreover, feedback-control algorithms that can be executed within such devices without the need for remote computing eliminate virtual tethers and any associated latencies. In this project, we develop a wireless and battery-less technology of this type, implanted subdermally along the back of freely moving small animals, for the autonomous recording of electroencephalograms, electromyograms and body temperature, and for closed-loop neuromodulation via optogenetics and pharmacology. The device incorporates a system-on-a-chip with Bluetooth Low Energy for data transmission and a compressed deep-learning module for autonomous operation, that offers neurorecording capabilities matching those of gold-standard wired systems. We also show the use of the implant in studies of sleep–wake regulation and for the programmable closed-loop pharmacological suppression of epileptic seizures via feedback from electroencephalography. The technology can support a broader range of applications in neuroscience and in biomedical research with small animals. https://www.nature.com/article...

With support from the Association of American Universities (AAU), we are creating a system for teaching evaluation that integrates evidence-based evaluation data from three sources: student course assessments, peer observation, and self-reflection. Existing student course assessments are being augmented to focus on observable best practices, rather than students' intuitive impression of the instructor. Peer observations are conducted in accordance with a consistent protocol, which includes guided pre-briefing and debrief sessions between the observer and observee. Self-reflection allows each instructor to note and report on their progress toward the implementation of best teaching practices. The system is currently being piloted and assessed en route to broader implementation. Learn more

Acoustics and signal processing research focuses on active noise control and distributed sensing. Active noise control reduces noise in hearing protection and communication systems to reduce noise induced hearing loss and to enhance the ability to communicate. Distributed sensing research uses signal processing to focus listening in a specified direction. This research blends mechatronics—the design of mechanical and electrical systems—with high performance signal processing and control algorithms to improve communication in noisy environments.

Image sensors are found at the heart of every camera system and convert incoming light into electronic signals. We are investigating image sensors for next-generation camera systems. This work involves multi-disciplinary electronics research including photonics and optics, semiconductor devices, mixed-signal integrated circuits and VLSI, digital image signal processing, and electronics systems design. The major thrusts in our activity are as follows:

• First, we are exploring the Quanta Image Sensor (QIS). The QIS is a revolutionary change in the way we collect images in a camera that is being invented at Dartmouth. In the QIS, the goal is to count every photon that strikes the image sensor, and to provide resolution of 1 billion or more specialized photoelements (called jots) per sensor, and to read out jot bit planes hundreds or thousands of times per second resulting in terabits/sec of data. The work involves design of deep-submicron jot devices, low-noise high-speed readout electronics, and novel ways of forming images from sequential jot bit planes at both the modeling and the simulation level and the characterization of actual devices and circuits. There is close collaboration with leading industry and scientific user communities.
• Second, we are investigating the use of image sensors in medicine and the life sciences. Photon-counting X-ray image sensors are being explored with a major medical equipment company. Application of our photon-counting QIS technology to low light fluorescence lifetime imaging microscopy (FLIM) is also being explored.
• Third, we are looking at innovative design and applications of CMOS image sensors to improving photography and scientific and industrial imaging, including low light and high speed applications. There are many avenues for innovation and invention and there is a high demand for image sensor specialists in industry.

Adversary intent inferencing and adversarial modeling investigates the feasibility of developing and utilizing an adversary intent inferencing model as a core element for predictive analyses and simulations to establish emergent adversarial behavior. It is our desire to use this intelligent adversary to predict adversary intentions, explain adversary goals, and predict enemy actions in an effort to generate alternative futures critical to performing course of action (COA) analysis. Such a system will allow planners to gauge and evaluate the effectiveness of alternative plans under varying actions and reactions to friendly COAs. This can also be applied in a broad range of areas.

Agent-based systems engineering aims to successfully cross-fertilize the fields of systems engineering and artificial intelligence. Systems engineering (control, signal processing and communications) focuses primarily on physical domains that can be characterized by rich mathematical dynamics while artificial intelligence deals with human perception, decision making and action. Goals of such cross-fertilization are to explore the modeling, performance and scientific foundations of software agent systems using ideas from classical systems engineering and computer engineering.

What is the nature of intelligence? Can we make machines that are intelligent? Machines that think like human beings or think differently? Can machines think even better than humans? What are the implications? These and other questions are being investigated.

Our asthma symptom monitoring solution is a wearable device that automatically detects and tracks cough, wheezing severity, lung function and inhaler use. The device provides objective information about the patient’s level of asthma control and level of compliance with the asthma plan.

Band-engineered semiconductors and phosphor materials for efficient solid-state lighting contributes toward the development of sustainable lighting technology for the 21st century. Lighting consumes ~1 TW of power worldwide, with greenhouse gas emission equivalent to 70% of the world's car emission. Solid-state lighting using semiconductor light emitting diodes (LEDs) has been widely recognized as the next generation energy-efficient lighting technology, since LEDs feature superior efficiency, longer lifetime, and lower environmental hazard when compared with conventional lighting sources. Currently a blue LED is usually combined with phosphor coating to generate white light. There are two major challenges to this technology: (1) LED Efficiency droop at high current density required for lighting; (2) Optical energy losses due to converting blue photons to lower energy photons by phosphors. To attack the former issue we are exploring band-engineering of semiconductors near direct-indirect gap transition to achieve anti-droop properties. For the latter issue we are investigating more efficient phosphor materials especially in the yellow-red regime to enhance its conversion efficiency. We are also conducting research on long-afterglow phosphors which can store optical energy during the day and emit light at night for emergency lighting and decorative lighting.

Bayesian knowledge bases, engineering, verification, and validation focuses on the fundamental problem of probabilistic modeling of knowledge in order to represent and reason about information in a theoretically sound manner. The world is replete with issues such as incompleteness, impreciseness, and inconsistency which makes the task of capturing even everyday tasks, processes, and activities very difficult, let alone trying to capture that of decision-making by experts or other complex phenomena. Improperly modeling uncertainty leads to numerous anomalies in reasoning as well as increased computational difficulties.

In large-scale multi-agent systems like robotic fleets, distributing the computational and decision-making tasks to be performed locally by individual robots (or agents) is a popular method of reducing the complexity and communication overhead of a fleet level control policy. However, the emergent behavior of these distributed systems can be inferior to those that are controlled in a centralized manner. This project seeks to understand how agents can partially coordinate their behavior by collaborating on group control actions. Intended outcomes are control algorithms for new, collaborative paradigms for multi-agent systems and formal guarantees on the behavior of these systems.

We are developing a device for continuous, non-invasive monitoring of a patient's cardiac hemodynamic status, allowing for physician intervention before the clinical symptoms of decompensation are observed, and potentially avoiding hospitalization.

Radiation therapy is used to treat cancer tumors by killing the tissue with high ionizing radiation doses. Until recently it has not been possible to image the radiation dose delivered to tissue, but through Cherenkov light imaging, this delivered dose can be mapped with high resolution cameras. The research group focuses on quantification of the imaging, and developing tools which allow radiation therapy to be delivered in a safer and more validated manner.

Ultrasound is a supplemental screening technique that has good sensitivity in dense breasts, is inexpensive, and is widely available, but unfortunately, it has high rates of false positives. Electrical impedance tomography (EIT) is a second attractive modality that is low-cost and has shown promise for cancer detection and in differentiating fibrocystic tissues from other tissues. Combining automated whole breast ultrasound (ABUS), which is a recent improvement to standard ultrasound, with EIT may significantly reduce the number of false positives of ABUS. If successful, the combined ABUS/EIT system could become an important screening technology for women with dense breasts.

In existing attempts to model complex systems, one critical aspect that has not been clearly ad-dressed involves the underlying mechanism for integrating the numerous “pieces” and “parts” that make up the target. Combining pieces is the process of aggregation and must handle inconsistencies among the pieces. Combining parts is the process of composition in which the parts are encapsulations of information with a set of meaningful operations defined on them. Parts are functional in nature and thus are driven by function composition. Extant research has not directly addressed this resulting in mathematically ad-hoc models opaque to analysis. We propose to develop a singular, rigorous, comprehensive computation framework that is axiomatic and provides the capabilities needed to model complex systems based on a new model of complex adaptive Bayesian Knowledge Bases and a novel, powerful analytical framework capable of wholistic end-to-end quantitative analysis of performance, robustness, vulnerability, and impacts of change on our targets being modeled. Furthermore, our results will be applicable to numerous domains of public purpose from crisis and catastrophe management for natural disasters and disease outbreaks to assessing the well-being of our financial system and national infrastructures.

Computational electromagnetics research is developing advanced analytical and numerical methods—such as the method of auxiliary sources, the method of moments, and pseudo spectral FDTD methods—for investigating high voltage non-linear electrostatic discharge phenomena as well as electromagnetic energy propagation in complex (Chiral and Bi-anizotropic) media.

Connections Hypothesis Provider (CHP) service built by Dartmouth College (PI – Dr. Eugene Santos) and Tufts University (Co-PI – Joseph Gormley) in collaboration with the National Center for Advancing Translational Sciences (NCATS). CHP aims to leverage clinical data along with structured biochemical knowledge to derive a computational representation of pathway structures and molecular components to support human and machine-driven interpretation, enable pathway-based biomarker discovery, and aid in the drug development process. In its current version, CHP supports queries relating to genetic, therapeutic, and patient clinical features (e.g. tumor staging) contribution toward patient survival, as computed within the context of our test pilot: a robust breast cancer dataset from The Cancer Genome Atlas (TCGA). We are using this as a proving ground for our system’s basic operations as we work to incorporate structured pathway knowledge and pathway analysis methods into the tool.

Cooperative control of multi-robot systems focuses on modeling and control of groups of high-speed mobile robots while accommodating communication latencies and nonlinear vehicle dynamics. In distributed cooperative control, robots communicate information about their state to each other; communication latencies and error depends on the amount of information communicated and the number of robots. We are developing distributed control system modeling and design tools that seek to maximize control bandwidth for a given information set. These tools will also assist in assessing the value of information transmitted in maintaining stability and performance of group dynamics. Both potential function path planning and control and predictive control methods are being developed.

We are providing advanced integrated circuits (ICs) for sensing to Professor Ryan Halter, who is combining them with transrectal electrical impedance tomography and electrical impedance sensing biopsy devices to improve the accuracy of ultrasound-guided prostate biopsy procedures.

Deception detection aims to automatically detect and infer the intentions behind deceptive actions. Our objectives are to 1) develop a framework for categorizing and classifying errors that may be committed by an expert, since not all errors are deception; and 2) design algorithms for automatic deception detection capable of providing detailed evidential information and explanation of deception intent, plus analysis of the deception's impact. Like insider threat, deception detection can occur in any number of scenarios and domains, and insider threat and deception detection are often interrelated.

Dielectric properties of tissue—measured through advanced microwave imaging techniques—convey functional information useful for making clinical diagnoses. The properties reflect tissue composition of fat, bone, water, proteins, etc., and often have unique spectral characteristics. The relative proportions and dynamic aspects of these constituents can have important implications for breast cancer imaging, osteoporosis detection, brain imaging, and heat therapy monitoring.

DIFUSE is an NSF-funded Dartmouth project aimed at creating opportunities for undergraduates to learn and use data science in introductory STEM courses and beyond. We work with teams of undergraduates, PhD students, and faculty to develop data science "modules" to integrate into existing course curriculum. We also offer opportunities for undergraduate and PhD students to apply data science and data visualization through Internships and work in the DALI Lab.

Distributed information retrieval aims to develop a large-scale information retrieval architecture that can be effectively and efficiently deployed in distributed environments. Heterogeneous information (such as content, formats and sources) is the typical issue that needs to be identified and handled in the distributed environment. Our objective is to develop a unified architecture called I-FGM (intelligent foraging, gathering and matching) for dealing with the massive amount of information in a dynamic search space within large-scale distributed platforms. The system will proceed to explore the information space, and continuously identify and update promising candidate information. Specific metrics are also being developed for performance evaluation.

Demand for high-performance permanent magnets is increasing rapidly for applications such as wind turbine generators and motors in both electric and hybrid cars. This market is served by rare earth (RE) magnets based on Nd2Fe14B and Sm2Co17. RE magnets are not without issues; they can chip, suffer thermal shock, and can suffer grain boundary corrosion. However, their biggest problems are: price volatility; that China largely controls the RE metals market; and that the extraction of RE metals creates severe environmental degradation. L10-structured Tau-MnAl has been of interest as a permanent magnet since the early 1960s. It has a theoretical energy product, (BH)max, between that of AlNiCo magnets and RE magnets with a value (12 MGOe) comparable to that of bonded Nd2Fe14B magnets. Further, it does not suffer from the issues associated with RE magnets, and potentially has the lowest cost per MGOe of any permanent magnet. The enigma is that the theoretical (BH)max has never been achieved: mechanically-milled particulates can show high coercivity (HC) but low saturation magnetization (MS) while warm-extruded material can show high MS but low HC. Tau-MnAlis a metastable phase that transforms from the high temperature ε phase, during which anti-phase boundaries (APBs), twins, stacking faults and dislocations are created. Depending on the processing conditions, the equilibrium β and γ2phases can also form. The fundamental difficulty with improving the magnetic properties of Tau-MnAl is that there is no clear understanding on how they depend on the defect structure. The grain size can also influence the magnetic properties either directly or by affecting the β and γ2 arrangement and defect formation.

We are using equal channel angular extrusion (ECAE) to process Tau-MnAl billets over a range of temperatures. Multiple passes will be performed utilizing Routes A (extruded billet is fed back into the ECAE jig in the same orientation) and BC(extruded billet is rotated 90o clockwise about its axis between passes). We will determine the texture of the extruded billets, and the density and arrangement of APBs, twins, stacking faults, dislocations and second phases after each pass, and relate these to the magnetic properties. The local chemistry will be explored at high resolution using atom probe tomography via collaboration with Dr. Baptiste Gault, Max-Planck-Institut fur Eisenforschung, Germany. To extend the strain range and, hence, defect densities studied we will also explore using severe plastic deformation through a collaboration with Prof. Gheoghe Gurau, University of Galati, Romania. Our working hypothesis is that we need a strong, c-axis alignment and a low density of APBs, twins and stacking faults (which locally disorder the material) for a high MS, while a low density of APBs, twins and stacking faults but a high dislocation density are required for a high HC. It is thought that a fine distribution of β and γ2 phases will also contribute to a high HC through magnetic domain wall pinning.

Funded by the National Science Foundation.

Electrical impedance imaging for prostate cancer screening is the process of imaging non-invasively the electrical properties (conductivity and permittivity) of the prostate and its vicinity using electrodes mounted onto an intracavitary probe.

Goal: to provide an imaging tool to effectively monitor the long-term physiological effects

of deep space and automate diagnosis, enabling crew members to be proactive in the

event of injury. More specifically, we are designing an integrated US-EIT system and

demonstrate proof-of-concept of US-EIT for enhanced ultrasound imaging capabilities of

deep internal bleeding.

The most common application, so far, of electrical impedance tomography (EIT) has been pulmonary monitoring of patients. In particular, we have been pursuing EIT for cardiac output monitoring (a related application) and EIT as a surrogate for pulmonary function tests.

Environmental-benign Group IV (Si, Ge, Sn) nanomaterials for solar cells, thermophotovoltaic (TPV) cells, and infrared photodetectors are being used to address the issue of sustainability as, in the case of solar cells, it is highly desirable to use naturally abundant, environmentally-friendly materials. Currently Si, CdTe and CuInGaSe2 (CIGS) are dominating solar cell materials. Although thin-film CdTe and CIGS solar cells offer higher efficiency their Si rivals, the rarity of Cd, In, and Se elements makes it impossible for them to meet a meaningful portion of global energy needs. The toxicity of Cd also raises environmental concern. By comparison, Group IV materials (Si,Ge,Sn) are very attractive due to their high abundance and low toxicity. Furthermore, (Si,Ge,Sn) alloys and nanostructures have already demonstrated many interesting properties surpassing Si, such as wide range of band gap tenability and direct gap behavior. These properties can potentially lead to better performance than CdTe and CIGS cells. We are investigating material growth, defect passivation, and band-engineering of these environmental-benign Group IV nanomaterials for solar cells and thermophotovoltaic (TPV) cells (direct conversion of thermal radiation to electricity). Similarly, in mid and far infrared photodetectors the dominant HgCdTe material may also be substituted by environmental benign, cost-effective GeSn or Sn nanostructures to achieve IR sensing and night vision.

Fluorescence-guided surgery is important for the resection of some types of cancerous tumors where the tumor and normal tissue are similar in appearance and texture, and patient prognosis depends heavily on the completeness of resection. By selectively tagging tumor tissue with fluorescent dyes, it becomes possible to visually discriminate between normal and tumor tissues and improve significantly the completeness of tumor resection.

High performance search and optimization aims to develop new models and algorithms for solving challenging engineering problems in domains such as mission planning and logistics, manufacturing process optimization, composite materials production, distributed plant scheduling and management, and policy evaluation, to name a few.

This project is investigating the synergy between the excellent mechanical behavior of FeMnNiAlCr high entropy alloys (HEAs) and the high solar absorptance of their native thermal oxides for high efficiency concentrated solar thermal power (CSP) systems working at >700^oC. The alloy itself would be used in high-temperature tubing to carry molten salts or supercritical carbon dioxide (sCO2), while the native oxide would act as a high-efficiency solar thermal absorber. The oxide layer is also dense and protective against oxidation (parabolic growth kinetics) at 750 °C. This new Fe-Mn based HEA system has already demonstrated a higher tensile strength and ductility than more expensive Ni-based Inconel 740 superalloys at both room temperature (with carbon doping) and 750^oC (with Ti doping), and their native thermal oxides have achieved a high optical-to-thermal conversion efficiency of ηtherm=90.8% at 750°C under 1000x solar concentration ratio. In preliminary corrosion studies, two-phase Cr-modified HEAs have sustained bromide molten salts for 14 days at 750°C with <2% weight loss. The project has close collaborations with the Oak Ridge National Laboratory (ORNL) and the Ames Laboratory in computational materials science and atom probe tomography (APT) to understand the fundamental structure-property relationship in FeMnNiAlCr HEAs.

Funded by DOE.

High-temperature solar absorbers for solar thermal systems are complementary to solar cells as another way of harvesting solar energy. The energy conversion efficiency increases with the working temperature of solar thermal systems according to the second law of thermodynamics. A critical component to achieve high working temperature is solar absorber coatings, which should absorb sunlight and convert it to heat efficiently with minimal thermal radiation loss in the infrared regime. They should also be stable at high temperatures. In this research we are investigating optical absorbers based on oxidation-resistant cermet materials incorporating metallic nanoparticles in a ceramic matrix for a working temperature of >560 C. Instead of using metal nanoparticles which are either easily oxidized at this temperature or too expensive to apply in large quantities, we are exploring certain metallic compounds that combines metal-like optical absorption properties with ceramic-like chemical stability. We also engineer the microstructure of the coating to trap infrared thermal radiation within the solar absorber coating and further increase the working temperature.

This project develops and evaluates image-based registration methods involving stereovision for registration and image updating during open spine surgeries for degenerative spondylolisthesis.

Image-guided neurosurgery gives the surgeon the ability to track instruments in reference to subsurface anatomical structures. Using clinical brain displacement data, a computational technique is being developed to model the brain deformation that typically occurs during neurosurgery. The resulting deformation predictions are then used to update the patient's preoperative magnetic resonance images seen by the surgeon during the procedure.

Assessing muscle health is important diagnostic information in the treatment of numerous peripheral nerve diseases. This project aims to incorporate spatial information from ultrasound into the inverse problem of electrical impedance tomography in order to accurately image the electrical properties of the muscle (under the skin and adipose tissue).

Influence of culture and society on attitudes and behaviors aims to build and employ social, cultural, and political data-driven models to explore and explain attitudes and behaviors. The efforts involve classifying the factors that play significant roles in attitudes and behaviors, abstracting general rules from traditional research such as sociological case studies, studying the inferencing structures that allow different factors to influence decision-making, reasoning from different points of view, and applying them in predicting behavior.

In social systems, like transportation networks, human users make decisions despite having uncertainty about the underlying state of the system (e.g., the current state of traffic). If a central authority is able to gather this information (e.g., Google or Apple maps), then they are presented with the opportunity to reveal that information to the system's users or strategically reveal pieces of information as a mechanism to shape the users' beliefs and ultimate behavior. In this project, we seek to identify policies for strategically signaling information to incentivize more desirable decisions by the systems users. Intended outcomes will be insights on how to design information signaling policies as well as as formal analysis on the effects of heterogeneity among the human user population.

Information processing and summarization are critical areas of research that study how we can develop stand-alone algorithms as well as algorithms fused with humans to handle and process information in a variety of forms. The goal is to be able to extract the meaning (or semantics) of the information in order to better manipulate/reason and present it to the human user. This is fundamental to solving problems such as avoiding information overload and providing effective summarization.

Can we teach computers to think outside the box? In other words, is it possible to replicate the innovative decision process computationally for machine learning? Extant research in machine learning has typically either focused on (a) building predictive models of a single internally-consistent target or on (b) a single task or decision in isolation, and more rarely, both, given the difficulties already posed within these more restrictive problems. The successes and utility of modern machine learning is clearly evident in numerous applications across many domains, and ever more so now with Big Data. Yet, the former focus (a) has made machine learning of complex targets (e.g. systems of systems, complex systems, a human) very elusive because of their inherent assumptions of expectations. This precludes the ability to learn emergent, unexpected, or innovative behaviors.

Insider threat and deception detection are two areas that focus on user actions and their impacts upon the systems with which they interact. Insider threat aims to understand and prevent malicious activities that are instigated by "trusted" users on complex computer/information systems. Such activities cover a broad spectrum ranging from simple theft of confidential data to the more subtle alteration of system performance and/or information. For the latter, examples can include minor perturbation of a component specification in a manufacturing process resulting in a rippling effect of final component failure to influencing the decision-makers by modifying their information flow and content. The goal is to model insider threat in order to predict behavior and ultimately infer their goals and intentions.

We are developing algorithms and a complementary metal oxide semiconductor (CMOS) imager to perform high speed voltage imaging of brain activities.

Integrated nanophotonics devices for high-bandwidth, ultra-low energy photonic data links offer solutions to the increase in energy consumption from computation and communication systems that has come with the rapid growth of information technology in the 21st century. Data transmission starts to consume even more energy than data processing in microprocessors, servers and data centers due to resistive losses and RC delay in electrical interconnects. Optical interconnection, on the other hand, does not have bandwidth limit or resistive losses as its electrical counterpart, thereby providing an idea solution to high bandwidth, ultralow energy data links. Our research aims to integrate nanophotonic devices monolithically on silicon chips to achieve electronic-photonic synergy, combining the merits of photons in data transmission with electrons in data processing. Recent research focuses on low-temperature (<450C) integration of nanophotonic devices such as photodetectors and photonic modulators with back end of line (BEOL) CMOS technology.

The Empower Lab creates digital therapeutics and intervention technologies, with an emphasis on inclusive design, patient experience, and health equity.

We seek students to contribute to the design, development, and evaluation of our interactive tools, which aim to positively shape behaviors and health outcomes through an empowerment approach. Our work employs human-centered methods and explores a variety of form factors and interaction paradigms (e.g., AR/VR/XR, games, toys, tangibles, musical interfaces, narratives, psychotherapeutic visualization, socially assistive robots, and more).

We work closely with stakeholder populations on a range of health topics; key areas of focus include mental health, women’s health, physical mobility/pain, aging-in-place, and pediatrics/childhood development. More details about our lab and its active projects, including specific recruiting opportunities, can be found here: empowerlab.dartmouth.edu/projects

Generally speaking, our collaborative research style integrates interdisciplinary skills from both computing and engineering as well as the social sciences and humanities. Students typically focus on contributing to one or two aspects of a project’s research activities, which include iterative design (UX, prototyping), implementation (software programming, physical fabrication), data analysis (quantitative/statistical or qualitative), stakeholder engagement (conducting interviews, observational or ethnographic work), and running studies (user testing, lab experiments, online or survey studies, or field trials).

If interested in getting involved, you can fill out our application form dartgo.org/empowerlab-apply or email the lab director, Prof. Liz Murnane (emurnane@dartmouth.edu) to share your resume, relevant interests and background, and any questions you might have. We look forward to hearing from you!

Label free genome sequencing is an advancing technology to "read" the sequence in a single DNA molecule in a massively-parallel fashion. The technology combines concepts of single nucleotide addition (SNA) sequencing, near field optics, single molecule force spectroscopy, and microfluidics. This work is performed in collaboration with Professor Dmitri Vezonov at Lehigh University.

This project provides quantitative data analysis, physical modeling, and numerical methods support to the Dartmouth Center of Cancer Nanotechnology Excellence. The project employs physical modeling and numerical methods to understand the electromagnetic interactions that occur when magnetic nanoparticles are placed in an alternating magnetic field in biological environments, and investigates the impact that biological parameters (e.g., blood flow) have on the ability to increase tumor temperatures locally.

Magnetic resonance elastography is being developed as a technique to measure the elasticity of tissue in vivo by gently shaking the tissue in a magnetic resonance imager. The displacements measured are used to determine tissue mechanical properties which can help identify and classify breast lesions. See also Discovering at Thayer School.

Successful next generation AI/ML for real-world applications must be able to deal with incomplete, sparse and noisy data as well as unexpected or adversarial circumstances that might arise while solving real world military problems. Furthermore, successful models must be able to learn new concepts with few examples. Unfortunately, even in this age of commodity machine learning models, one still needs to train new models with a large amount of training examples to achieve the requisite performance from deep neural networks. The resulting models often lack robustness, explainability, and human-level intelligence.

One of the best models for intelligence is the one inspired by human brain itself, which can support robust and massive parallelism with ease. We are exploring two related aspects of human brain processing that will play a key role in a Third Wave AI revolution: (1) a meso-scale cortical computation that easily finds low dimensional manifolds where learning can naturally take place, and (2) recurrent module networks, as thin as one- or two-layer networks, akin to cellular automata, trained on time sequences rather than input-output pairs. We believe this process of simplification and automata implementation is a better model of the human neocortex than current state-of-the-art deep networks whose optimized architectures are often ad hoc and do not reflect biological reality. This project is being sponsored in part by DARPA.

At the core of novel wearable and implantable devices are cutting-edge sensing technologies that can be integrated with the body. In this project, we leverage MEMS, micro/nanofluidics, and other relevant technologies to develop advanced biophysical and biochemical sensors. One example of such efforts is nanoelectrokinetics-based devices for ultra-sensitive protein biomarker detection through electrokinetic enrichment of biomolecules (Lab Chip 2017, Anal. Chem. 2018, Nanoscale 2018, PNAS 2019, Angew. Chem. 2020, etc.). https://www.pnas.org/doi/abs/1...

Microfabricated magnetic components using nanomaterials make it possible to miniaturize power-handling magnetic components through taking advantage of the materials' high-flux-density and high-frequency capabilities. We are developing practical methods of depositing these materials and fabricating inductors and transformers on silicon chips or in other technologies.

Microwave electronics capable of fast data acquisition (approaching real-time) are being developed for brain imaging applications. Also, site-specific antenna arrays for microwave imaging are in development for heel imaging (screening for osteoporosis) and for brain imaging applications.

Microwave imaging spectroscopy (also commonly referred to as microwave computed tomography) presents the challenge of measuring the signal data necessary to produce meaningful images. Development of site-specific antenna arrays along with improved electronic signal detection technology is rapidly making this measurement feasible for breast cancer detection. A second-generation tomographic breast imaging system has been completed and the data are being used to recover permittivity and conductivity maps of the breast for evaluation by a clinician.

Real-world complex systems can be observed from many different angles or perspectives, and datasets collected from various perspectives often emphasize different types of features. This results in inconsistent beliefs about what is relevant to the system, how relevant features are related to one another, and what statistical properties these features possess. Many methods have been proposed to combine such diverse information sources. However, current algorithms only learn from each dataset separately and then combine individual outputs since this is easier to do with heterogeneous datasets with unknown feature correlations. This approach, although convenient and intuitive, cannot capture the logical linkages between various datasets. To understand variables’ interactions learned from datasets with noise and incompleteness, we are exploring algorithms that naturally fuse these datasets based on their shared variables and induce new variable relationships.

Near-infrared imaging (NIR) provides a way to quantify blood and water concentrations in tissue, as well as structural and functional parameters. Since normal tissue, benign tumors, and malignant tumors each carry different concentrations of both hemoglobin and water, and have different levels of oxygen demand and ultrastructural scattering, NIR spectroscopy can be combined into standard imaging systems as an effective method of to provide additional information for breast cancer detection and diagnosis. Work is ongoing to improve techniques for better image reconstruction, display and integration with magnetic resonance imaging (MRI) and computed tomography (CT) imaging.

See also Center for Imaging Medicine

Reinforcement learning has shown much success in games such as chess, backgammon, and Go. However, in most of these games, agents have full knowledge of the environment at all times. We describe a deep learning model that successfully maximizes its score using reinforcement learning in a game with incomplete and imperfect information. We apply our model to FlipIt (1), a two-player game in which both players, the attacker and the defender, compete for ownership of a shared resource and only receive information on the current state upon making a move. Our model is a deep neural network combined with Q-learning and is trained to maximize the defender's time of ownership of the resource. We extend FlipIt to a larger action-spaced game with the introduction of a new lower-cost move and generalize the model to multiplayer FlipIt.

(1) van Dijk, M, Juels, A, Oprea, A, Rivest, RL, FlipIt: The Game of "Stealthy Takeover." Journal of Cryptology, 26, 655–713 (2013)

Non-linear image reconstruction techniques is at the core of the medical imaging projects. Excitation-induced measurements from each instrument are compared with calculations from corresponding numerical models to compute updated property images of the biological target. As the images are progressively updated (or refined) in a non-linear iterative process, important features and functional information related to the objects physiological status—tumor, benign tissue, etc.—become more apparent. The computational core of the breast imaging project works synergistically with all four groups to improve our fundamental understanding of these mathematical systems to improve overall image quality and resolution. These processes have been developed for both 2D and 3D geometries in each modality and are being expanded to exploit emerging parallel computing capabilities.

To advance the science of decision-making as it pertains to how people learn to make decisions and how this process can be captured computationally, we are specifically addressing the challenge of how nonlinear decisions can be learned from data, experience, and even interactions with other decision-makers. Nonlinear thinking is a prized ability we, humans, have that is ubiquitously applied across any and all domains when the problems are challenging, and known solutions or ways of addressing the problems all fail to provide an adequate solution – e.g., All available choices are bad choices, must we settle for the least bad one? The ability to discover a new choice has been called being nonlinear, innovative, intuitive, emergent, or “outside-the-box.” It is well-documented that humans can often excel at such thinking in situations when there is a scarcity/overflow of data, significant uncertainty, and numerous contradictions in what is known or provided. However, how this can be replicated computationally for a machine has yet to be fully addressed or understood in extant research.

Optical molecular imaging is being used to provide molecular guidance in cancer surgery. Fluorescent contrast agents are in pre-clinical and clinical studies to image cancer tumors in vivo, with a dual focus, first on getting more accurate information out of the tissue, and secondly to provide better information about the specificity of the molecules as markers. Systems and algorithms for diffuse fluorescence imaging of tissue are studied, both as a stand-alone system, and as coupled to magnetic resonance imaging and computed tomography imaging. Tracer kinetic modeling is also being developed to allow quantitative imaging of molecular binding in vivo.

Passive high-frequency power components are often the limiting factors in reducing the power loss, size, cost, and weight of high-frequency electronic power converters. Through detailed analysis, modeling, and optimization of high-frequency effects in inductors, transformers, and capacitors, we are improving performance of these components and making it easier to design the efficient, low-cost power electronics needed for a wide range of applications including energy efficiency and renewable energy.

Photodynamic therapy (PDT) is a newly emerging therapy for displastic tissues, such as cancer, age-related blindness, pre-malignant transformation or psoriasis. The therapy involves the administration of a photosensitizing agent, together with the application of moderate intensity light to active the molecules to produce local doses of singlet oxygen. Ongoing research topics include, developing improved dosimetry instrumentation and software, fluorescence tomography imaging to sense drug localization, and assaying unique tumor biology and treatment effects in experimental cancers.

Porous thermoelectric cells (TECs) are being developed for waste heat recovery. This project seeks to convert waste thermal energy directly into electricity potentially increasing overall energy efficiency by 15-20% and providing new portable electric power sources, particularly for cold regions. The project is focusing on low cost, nanostructure-engineered TEC (NETECs) materials based on earth-abundant, highly machinable metallic alloys and intermetallic compounds by engineering the grain sizes, second phase precipitation, and nanopores in transition metal intermetallic compounds. The project is currently focusing on the compound Fe2AlV. The location of quaternary atoms in the lattice is being determined by the TEM-based technique ALCHEMI.

Funded by USA-CRREL

The goal of this project is to optimize and validate, in human and animal studies, intraoperative image updating methods that assimilate intraoperative data acquired with stereovision and ultrasound with computational modeling to update preoperative MRI in order to maintain image registration accuracy throughout surgery.

Process query systems have applications that involve using databases or datastreams of events to detect instances of processes. In those applications, events provide evidence that is used to infer the existence and estimate the states of the various processes of interest. Examples of such applications include: network and computer security; network management; sensor network tracking; military situational awareness; and critical infrastructure monitoring and protection.

Quantitative scatter imaging makes use of the fact that normal functioning cells and cancerous cells show differences both in the components within the cells and in the structural organization of the cells within the tissue. These physical distinctions in biological structure have been shown to scatter light differently. This study develops a new approach to imaging scatter-based contrast over a wide field of view in tissue using high frequency structured light. These scattering features may provide a method to diagnostically identify abnormal tissue without the need to administer targeted compounds. This approach has the potential to generate new diagnostic screenings and new approaches for surgical guidance.

Rare-earth selective emitters for high efficiency TPV systems address the major limiting factor for the efficiency of TPV cells: the broad emission spectrum from the heat source. Photons with energy smaller than the band gap of semiconductor PV materials cannot be absorbed, while for those with energy much greater than the band gap the excess energy is lost to heat instead of generating electricity. Although there is nothing we can do to change the solar spectrum, for thermal sources of TPV systems we may engineer the thermal emission spectrum by using certain coatings called "selective emitters." Using this approach, the broad blackbody emission spectrum can be condensed to a much narrower band that optimally matches the absorption spectrum of semiconductor TPV cells. In particular, we are investigating rare-earth ceramic selective emitters such as Yb and Er compounds to match the absorption of Si and Ge TPV cells. We are also engineering the microstructures of these compounds to manipulate photons for enhanced selective emitting performance.

Robot design and smart navigation focuses on developing affordable robot designs that employ "smart navigation" for path planning and mobility in extreme terrain, rather than complex and expensive vision systems. We are developing solar-powered robotic platforms for deploying scientific instrumentation over hundreds of kilometers in Arctic and Antarctic regions. These robots employ proprioceptive sensors to determine whether difficult terrain is passable, and if not, to navigate around such terrain.

Radiation therapy is used to treat cancer tumors by killing the tissue with high ionizing radiation doses. Modern external beam radiotherapy systems deliver high dose levels to precisely marked tumor volume in less time. As a mis-administration can have potentially severe impact to the surrounding healthy tissue, more stringent and complex quality assurance measurements are required in clinics. By developing a comprehensive optical dose imaging camera system, we aim to fundamentally simplify the quality assurance process and, in turn, to further promote the culture of safety in radiotherapy. By converting the dose to visible light using scintillation phantom, we can image and reconstruct 3D dose maps in real time, enabling complete and accurate verification in a fast enough timeframe for it to be useful in every procedure.

As systems grow large and additional communication channels are implemented, new avenues for noise, hazards, and bad actors are introduced. Unlike in a single-agent system, when many agents are connected, individual failures or attacks may be hard to detect, and local perturbations can cascade into much more expansive failures. This project studies different adversarial and risk-aware design environments through the lens of zero-sum game theory. Intended outcomes are secure strategies in networked adversarial environments and fundamental analysis of the relationship between information and the ability to guarantee security.

Self-assembled nanophotonic structures for light trapping in solar cells can help with the critical task of reducing $/W for large scale applications. In conventional wafer-based solar cells, material alone constitutes more than half of the cost. Thin-film solar cells have become an attractive solution in recent years due to a drastic decrease in material consumption by 100x. The trade-off, however, is that their efficiency is limited by the thickness of the cells which is insufficient to absorb all the sunlight. Light trapping is needed to increase the optical path length of sunlight in these thin-film solar cells in order to improve the efficiency. In this research we are investigating self-assembled nanophotonic structures integrated on the backside of thin-film solar cells for sunlight trapping. The nanophotonic structure diffracts the incident sunlight into oblique angles so that it propagates laterally in the thin films and gets completely absorbed. Recent work demonstrated nanophotonic structures using self-assembled porous anodic aluminum oxide as a fabrication template for effective light trapping.

We are developing a framework using advanced tools from random graph theory and spectral graph theory to carry out the quantitative analysis of the structure and dynamics of large networks—with the focus of graph merging and subgraph detection. This framework, using information theory as a demarcative tool, enables one to carry out analytic computations of observable network structures and capture the most relevant and refined quantities of real-world networks.

The resolution of ultrasound imaging is typically determined by the wavelength of sound. We are circumventing this limitation with a new class of phase change nanoparticle contrast agents, called laser-activated nanodetectors (LANDs). These contrast agents, which consist of a liquid perfluorocarbon core and encapsulated dye, undergo vaporization upon exposure to pulsed laser irradiation. Several milliseconds later, the LANDs recondense back to their stable liquid state. While in their gaseous state, the LANDs provide high contrast in ultrasound images. The end result is a triggerable "blinking" contrast agent. We are currently developing algorithms to harness the blinking to improve the resolution of the imaging system as well as probe the tissue properties.

Terrain identification research focuses on using small, lightweight robots to classify, characterize, and identify terrain properties necessary to predict mobility of these vehicles on the terrain. Terramechanics models for heavy vehicles are well understood, but similar comprehensive models do not exist for lightweight (sub-500 kg) vehicles. We are developing terrain models and modeling tools that can be used to asses mobility on a given terrain, while avoiding maneuvers that cause immobilization. We seek to integrate terrain identification and traction/stability control of the robots in order to allow autonomous or remote control of these robots at the maximum attainable speeds and accelerations achievable on the terrain.

By computing properties of all known materials, The Materials Project aims to remove guesswork from materials design in a variety of applications. Experimental research can be targeted to the most promising compounds from computational data sets. Researchers will be able to data-mine scientific trends in materials properties. By providing materials researchers with the information they need to design better, The Materials Project aims to accelerate innovation in materials research.

Therapy monitoring is an important emerging application of imaging modalities. These and other current research topics include:

• near-infrared imaging of brain tissue;
• near-infrared spectroscopy for diagnosing peripheral vascular disease;
• electrical impedance spectroscopy for radiation therapy monitoring;
• magnetic resonance elastography for detecting brain or prostate lesions; to follow the progression of diabetic damage in the foot; and to answer basic questions of wave propagation in tissue;
• microwave imaging spectroscopy for hyperthermia therapy monitoring, brain imaging, and detection of early-stage osteoporosis;
• electrical impedance tomography for monitoring traumatic brain injury progression and therapy.

We are investigating how to combine recent advances in topological data analysis with recent advances in machine learning to enhance multiple hypothesis tracking system. Our goal is to improve detecting, clustering, classifying and tracking of various patterns-of-life trajectories by developing the capability to distinguish behavioral types at all scales.

Photoacoustic imaging is a hybrid technique which relies on a pulsed laser to generate acoustic waves within tissue. It is capable of achieving high resolution images with optical contrast tens of millimeters deep in tissue. We are working to translate this technology to the clinic with an initial focus on identifying metastatic lymph nodes in breast cancer patients. This project involves the development of the imaging system, creation of spectroscopic imaging algorithms, and combining the two for imaging in the clinic.

Unexploded ordnance (UXO) detection and discrimination approaches are being developed to solve the Department of Defense's (DoD) most pressing environmental problems: UXO cleanup and humanitarian de-mining. The program combines advanced forward and inverse EM sensing approaches with statistical signal processing methodologies to solve these complex and challenging problems. See also UXO Research Group.

User modeling and user intent inferencing involves building dynamic cognitive user models that can predict the goals and intentions of a user in order to understand and ultimately provide proactive assistance with user tasks, such as information gathering. The key is to capture the user's intent by answering questions such as: what is the user's current focus, why is the user pursuing certain goals, and how will the user achieve them? The efforts involve machine learning, knowledge representation, intent inferencing, and establishment of proper evaluation metrics. This work has been applied to assisting with intelligent information retrieval and enhancing the effectiveness of intelligence analysts.

Visible-blind and solar-blind ultraviolet detectors for flame and oil spill detection are useful for solving the technical challenge of operating these detectors with a large background radiation of sunlight. UV detection in the wavelength range of 210-300 nm is useful for flame, oil spill and missile detection, and traditional approaches include wide band gap semiconductor UV photodetectors and vacuum-based photoemission detectors. We are investigating more effective and less costly technologies for UV detection, including self-assembled visible light filters and solid-state photoemission detectors based on metal nanoparticles with plasmonic enhanced responsivity.

Capabilities in continuous monitoring of key physiological parameters of disease have never been more important than in the context of the global COVID-19 pandemic. Soft, skin-mounted electronics that incorporate high-bandwidth, miniaturized motion sensors enable digital, wireless measurements of mechanoacoustic (MA) signatures of both core vital signs (heart rate, respiratory rate, and temperature) and underexplored biomarkers (coughing count) with high fidelity and immunity to ambient noises. This paper summarizes an effort that integrates such MA sensors with a cloud data infrastructure and a set of analytics approaches based on digital filtering and convolutional neural networks for monitoring of COVID-19 infections in sick and healthy individuals in the hospital and the home. The sensors, deployed on COVID-19 patients along with healthy controls in both inpatient and home settings, record coughing frequency and intensity continuously, along with a collection of other biometrics. The methodology creates opportunities to study patterns in biometrics across individuals and among different demographic groups. https://www.pnas.org/doi/abs/1...

Fully implantable wireless systems for the recording and modulation of neural circuits that do not require physical tethers or batteries allow for studies that demand the use of unconstrained and freely behaving animals in isolation or in social groups. Moreover, feedback-control algorithms that can be executed within such devices without the need for remote computing eliminate virtual tethers and any associated latencies. In this project, we develop a wireless and battery-less technology of this type, implanted subdermally along the back of freely moving small animals, for the autonomous recording of electroencephalograms, electromyograms and body temperature, and for closed-loop neuromodulation via optogenetics and pharmacology. The device incorporates a system-on-a-chip with Bluetooth Low Energy for data transmission and a compressed deep-learning module for autonomous operation, that offers neurorecording capabilities matching those of gold-standard wired systems. We also show the use of the implant in studies of sleep–wake regulation and for the programmable closed-loop pharmacological suppression of epileptic seizures via feedback from electroencephalography. The technology can support a broader range of applications in neuroscience and in biomedical research with small animals. https://www.nature.com/article...

Many communities are not prepared to understand the changes they may need to make under our currently changing climate in order to adapt and mitigate for a more resilient future. We are working with communities and with the U.S. Army Corps of Engineers to use published forecasts of climate change along with economics, planning policy, and engineering considerations in order to facilitate planning for a more resilient future for communities within the U.S. and elsewhere.

Band-engineered semiconductors and phosphor materials for efficient solid-state lighting contributes toward the development of sustainable lighting technology for the 21st century. Lighting consumes ~1 TW of power worldwide, with greenhouse gas emission equivalent to 70% of the world's car emission. Solid-state lighting using semiconductor light emitting diodes (LEDs) has been widely recognized as the next generation energy-efficient lighting technology, since LEDs feature superior efficiency, longer lifetime, and lower environmental hazard when compared with conventional lighting sources. Currently a blue LED is usually combined with phosphor coating to generate white light. There are two major challenges to this technology: (1) LED Efficiency droop at high current density required for lighting; (2) Optical energy losses due to converting blue photons to lower energy photons by phosphors. To attack the former issue we are exploring band-engineering of semiconductors near direct-indirect gap transition to achieve anti-droop properties. For the latter issue we are investigating more efficient phosphor materials especially in the yellow-red regime to enhance its conversion efficiency. We are also conducting research on long-afterglow phosphors which can store optical energy during the day and emit light at night for emergency lighting and decorative lighting.

Orthopedic implants are widely used to treat bone and joint disorders, such as fractures, osteoarthritis, and spinal deformities. However, conventional implant materials, such as stainless steel, titanium, and cobalt-chromium alloys, have several limitations: (i) they may cause adverse reactions and metal sensitivity due to their foreign ions and corrosion products; (ii) they may fail prematurely due to stress concentration and fatigue; (iii) they may interfere with bone remodeling and healing due to their mismatched mechanical properties with the host tissue; (iv) they can become a nidus for bacterial infection and biofilm colonization. Therefore, there is a need for novel implant materials that can overcome these challenges and improve the clinical outcomes of orthopedic surgery. Zinc is an attractive candidate for orthopedic implants because it is an essential trace element in the human body that plays a key role in bone metabolism and wound healing. Moreover, Zn is biodegradable and can be gradually resorbed by the body without leaving any permanent foreign material. The typical in vivo corrosion rates of unalloyed zinc in rats is 0.03 mm/yr, which is a useful rate for biodegradation. However, unalloyed Zn has very poor mechanical strength, which limits its application as an implant material. To address these issues, we are developing new Zn-based alloy that contains small amounts of silver, calcium, iron, magnesium, and manganese, and using a novel processing route. These alloying elements are chosen based on their beneficial effects on the biological and mechanical properties of Zn. Specifically, Ag has antibacterial activity and can reduce the risk of infection; Ca can promote bone formation and integration; Fe can enhance fracture fixation and blood compatibility; Mg can improve biocompatibility and corrosion resistance; Mn can increase ductility and strength. An adequate rate of implant degradation will allow the bone to heal properly before degrading and allowing the bone to support any loads.

Environmental fluid mechanics research studies natural fluid systems as agents for the transport and dispersion of environmental contamination. Understanding transport and dispersion processes in natural fluid flows, from the microscale to the planetary scale, serves as the basis for the development of models aimed at simulations, predictions, and ultimately sustainable environmental management. Research within this scope is diverse and can involve a variety of scientific and engineering disciplines such as civil, mechanical, and environmental engineering, meteorology, hydrology, hydraulics, limnology, and oceanography.

Environmental-benign Group IV (Si, Ge, Sn) nanomaterials for solar cells, thermophotovoltaic (TPV) cells, and infrared photodetectors are being used to address the issue of sustainability as, in the case of solar cells, it is highly desirable to use naturally abundant, environmentally-friendly materials. Currently Si, CdTe and CuInGaSe2 (CIGS) are dominating solar cell materials. Although thin-film CdTe and CIGS solar cells offer higher efficiency their Si rivals, the rarity of Cd, In, and Se elements makes it impossible for them to meet a meaningful portion of global energy needs. The toxicity of Cd also raises environmental concern. By comparison, Group IV materials (Si,Ge,Sn) are very attractive due to their high abundance and low toxicity. Furthermore, (Si,Ge,Sn) alloys and nanostructures have already demonstrated many interesting properties surpassing Si, such as wide range of band gap tenability and direct gap behavior. These properties can potentially lead to better performance than CdTe and CIGS cells. We are investigating material growth, defect passivation, and band-engineering of these environmental-benign Group IV nanomaterials for solar cells and thermophotovoltaic (TPV) cells (direct conversion of thermal radiation to electricity). Similarly, in mid and far infrared photodetectors the dominant HgCdTe material may also be substituted by environmental benign, cost-effective GeSn or Sn nanostructures to achieve IR sensing and night vision.

In large-scale industrial fermentations, it can be expensive to add oxygen, and to cool fermenters to mesophilic temperatures (20-45C). Using anaerobic thermophilic bacteria avoids both problems, however many genetic tools that have been originally developed for model organisms such as Saccharomyces cerevisiae and Escherichia coli do not work in these organisms. My group is working to develop several types of genetic tools necessary for domestication of thermophilic anaerobic bacteria, including:

• Ways to get foreign DNA into cells
• Tightly-controlled inducible promoters for reliable temporal control of gene expression
• Plasmid-based gene expression systems
• Chromosomal editing tools
• Ways to control the mutation rate

As perhaps the most dynamic of all naturally-deposited porous media, the changing structure of snow reflects changes in weather and the environment. A wealth of satellite imagery exists over the vast expanses of the Antarctic and Greenland ice sheets, yet there is an incomplete understanding of what variations in the images means. From snow and ice cores collected from traverses across the Antarctic Ice Sheet and from several locations on the Greenland Ice Sheet, we are developing the physical basis by which satellite imagery can be used to make high-resolution maps of snow accumulation over vast and remote areas of the ice sheets.

Analyzing potential synergies between deep decarbonization of the energy system and reducing (inequities in) negative health impacts. MORE

Soft magnets play a vital role in efficient energy conversion in a variety of important applications and industries including wide-bandgap semiconductors, electric vehicles, aeronautics, and aerospace, particularly at high temperatures. Improving the efficiency of modern power electronics and electrical machines via advanced soft magnets has the potential to significantly contribute to global energy savings, thereby leading to a reduction of the associated carbon footprint. In this project, we are working on two novel FeCoMnAl high-entropy alloy (HEA) soft magnets, one of which is single-phase B2 (Fe₃₀Co₄₀Mn₁₅Al₁₅) and the other consists of an ordered B2-phase matrix enriched with Co/Al and uniformly distributed BCC nanoprecipitates enriched with Fe/Mn (Fe₄₀Co₃₀Mn₁₅Al₁₅). The two HEAs show similar properties, viz., a high saturation magnetization of 158-162 Am² kg^-1, a high Curie temperature of 1020-1081 K, a low coercivity of 108-114 A m^-1, a high electrical resistivity of ~230 µΩ cm, and good thermal stability. We are processing these HEAs using both a powder metallurgy route and via additive manufacturing. The magnetic properties and microstructures of the resulting materials are being examined using combination of a VSM, TEM, SEM and XRD examinations.

This project is investigating the synergy between the excellent mechanical behavior of FeMnNiAlCr high entropy alloys (HEAs) and the high solar absorptance of their native thermal oxides for high efficiency concentrated solar thermal power (CSP) systems working at >700^oC. The alloy itself would be used in high-temperature tubing to carry molten salts or supercritical carbon dioxide (sCO2), while the native oxide would act as a high-efficiency solar thermal absorber. The oxide layer is also dense and protective against oxidation (parabolic growth kinetics) at 750 °C. This new Fe-Mn based HEA system has already demonstrated a higher tensile strength and ductility than more expensive Ni-based Inconel 740 superalloys at both room temperature (with carbon doping) and 750^oC (with Ti doping), and their native thermal oxides have achieved a high optical-to-thermal conversion efficiency of ηtherm=90.8% at 750°C under 1000x solar concentration ratio. In preliminary corrosion studies, two-phase Cr-modified HEAs have sustained bromide molten salts for 14 days at 750°C with <2% weight loss. The project has close collaborations with the Oak Ridge National Laboratory (ORNL) and the Ames Laboratory in computational materials science and atom probe tomography (APT) to understand the fundamental structure-property relationship in FeMnNiAlCr HEAs.

Funded by DOE.

High-temperature solar absorbers for solar thermal systems are complementary to solar cells as another way of harvesting solar energy. The energy conversion efficiency increases with the working temperature of solar thermal systems according to the second law of thermodynamics. A critical component to achieve high working temperature is solar absorber coatings, which should absorb sunlight and convert it to heat efficiently with minimal thermal radiation loss in the infrared regime. They should also be stable at high temperatures. In this research we are investigating optical absorbers based on oxidation-resistant cermet materials incorporating metallic nanoparticles in a ceramic matrix for a working temperature of >560 C. Instead of using metal nanoparticles which are either easily oxidized at this temperature or too expensive to apply in large quantities, we are exploring certain metallic compounds that combines metal-like optical absorption properties with ceramic-like chemical stability. We also engineer the microstructure of the coating to trap infrared thermal radiation within the solar absorber coating and further increase the working temperature.

Ice cores drilled in cold areas of the Greenland and Antarctic Ice Sheets provide high-resolution climate records that are essential for understanding abrupt climate change. The only remaining samples of the atmopshere from past centuries and millennia are contained in bubbles found deep in glacial ice. We are measuring the physical structure, transport properties, and microstructure from ice cores from Greenland and Antarctica to better understand mechanisms of the trapping of gases in ice cores for improved understanding of abrupt climate changes in the past.

Integrated nanophotonics devices for high-bandwidth, ultra-low energy photonic data links offer solutions to the increase in energy consumption from computation and communication systems that has come with the rapid growth of information technology in the 21st century. Data transmission starts to consume even more energy than data processing in microprocessors, servers and data centers due to resistive losses and RC delay in electrical interconnects. Optical interconnection, on the other hand, does not have bandwidth limit or resistive losses as its electrical counterpart, thereby providing an idea solution to high bandwidth, ultralow energy data links. Our research aims to integrate nanophotonic devices monolithically on silicon chips to achieve electronic-photonic synergy, combining the merits of photons in data transmission with electrons in data processing. Recent research focuses on low-temperature (<450C) integration of nanophotonic devices such as photodetectors and photonic modulators with back end of line (BEOL) CMOS technology.

The research lab at Dartmouth led by Professor Lee Lynd is active in research on the following topics:

• Microbial Cellulose Utilization, including fundamental and applied aspects
• Metabolic Engineering, focusing on thermophilic cellulolytic bacteria for fuel production
• Innovative Biomass Processing Technologies, including development, design, and evaluation
• Sustainable Bioenergy Futures, including resource, environment, and economic development

We approach these topics from a diversity of academic disciplines with molecular biology, microbiology, chemical/biochemical engineering providing the foundation for the first three. Consistent with the "Pasteur's Quadrant" model articulated by Donald Stokes (Brookings Institution Press, Washington, DC, 1997), we see advancing applied capability and increased fundamental understanding as having strong potential to be convergent and mutually-reinforcing, and we aspire to work in this mode.

A central theme of the Lynd group is processing cellulosic biomass in a single step without added enzymes. Such "consolidated bioprocessing" (CBP) is a potential breakthrough, and "is widely considered to be the ultimate low-cost configuration for cellulose hydrolysis and fermentation" (joint DOE/USDA Roadmap, 2007). We are focused on production of ethanol, a promising renewable fuel. The CBP strategy is however potentially applicable to a very broad range of fuels and chemicals.

Researching complex interactions between climate hazards and communities to inform governance of coastal risk. MORE

Metabolism can be understood at many levels of aggregation from individual enzymes to the whole organism. An important intermediate level of aggregation is the metabolic pathway. Developing pathways that enable rapid production of desired compounds at high yield and titer requires a detailed understanding of both the individual components and the systems-level behavior that results from the interaction of these components, including:

• Identification of constituent enzymes in a pathway and the stoichiometry of the reactions they mediate
• Characterization of enzyme inhibition and regulation
• Development of screens and selections to improve properties of key enzymes.
• Protein engineering to increase activity, decrease inhibition, or change substrate specificity

Microfabricated magnetic components using nanomaterials make it possible to miniaturize power-handling magnetic components through taking advantage of the materials' high-flux-density and high-frequency capabilities. We are developing practical methods of depositing these materials and fabricating inductors and transformers on silicon chips or in other technologies.

The microstructual evolution of snow under a temperature gradient has been of interest for many years since this can lead to persistent weak layers, which are possible microstructural causes of avalanches. Ice crusts can form on top or within a snowpack from a variety of meteorological conditions including significant melt/freeze or freezing rain events, and once buried, they can persist throughout the entire winter season and act as an ideal sliding surface for dangerous slab avalanches in seasonal mountain snowpacks. Both of these phenomena are important because the number of fatalities from avalanches in the US has increased annually since the 1970s. Avalanches can also have substantial economic impacts due to road closures, the costs of rescue and building damage, and, with continued global warning, more avalanches are expected in Arctic regions. To understand avalanche nucleation, we are deforming two types of specimens (heterogeneously-layered snow and snow containing an ice lens) in a micro CT located in a cold room, in which the specimens are repeatedly imaged during loading. We are also performing more macroscopic deformation experiments on larger samples at both different rates and different temperatures, which are imaged using a high-speed video camera during loading. The final deformed microstructures in both cases are imaged at high resolution using a scanning electron microscope, which provides information on both the effects of crystal orientation on deformation while clearly delineating one ice crystal orientation from another. Based on the experimental observations, a multiscale computational model is being built to understand crack initiation/crack propagation as well as the deformation mechanisms in heterogenously-layered snow samples containing persistent weak layers and ice/snow interfaces.

Passive high-frequency power components are often the limiting factors in reducing the power loss, size, cost, and weight of high-frequency electronic power converters. Through detailed analysis, modeling, and optimization of high-frequency effects in inductors, transformers, and capacitors, we are improving performance of these components and making it easier to design the efficient, low-cost power electronics needed for a wide range of applications including energy efficiency and renewable energy.

Advancing the quantitative understanding of multisector dynamics. MORE

Currently, most ethanol is produced using various strains of yeast. These organisms are very good at producing ethanol but have no native ability to consume lignocellulose. This has made it difficult to develop cost-effective yeast-based processes for lignocellulosic ethanol production. My group takes an alternative approach of starting with organisms that natively ferment lignocellulosic biomass and engineering them for efficient biofuel formation. To do this, we need to understand key aspects of the physiology of these native biomass-fermenting organisms, including:

• Which substrates they can consume, and how these substrates are used for growth, energy production, and product formation.
• Factors that limit growth and fermentation.
• Genetic adaptation to stresses associated with industrial fermentation

Porous thermoelectric cells (TECs) are being developed for waste heat recovery. This project seeks to convert waste thermal energy directly into electricity potentially increasing overall energy efficiency by 15-20% and providing new portable electric power sources, particularly for cold regions. The project is focusing on low cost, nanostructure-engineered TEC (NETECs) materials based on earth-abundant, highly machinable metallic alloys and intermetallic compounds by engineering the grain sizes, second phase precipitation, and nanopores in transition metal intermetallic compounds. The project is currently focusing on the compound Fe2AlV. The location of quaternary atoms in the lattice is being determined by the TEM-based technique ALCHEMI.

Funded by USA-CRREL

Rare-earth selective emitters for high efficiency TPV systems address the major limiting factor for the efficiency of TPV cells: the broad emission spectrum from the heat source. Photons with energy smaller than the band gap of semiconductor PV materials cannot be absorbed, while for those with energy much greater than the band gap the excess energy is lost to heat instead of generating electricity. Although there is nothing we can do to change the solar spectrum, for thermal sources of TPV systems we may engineer the thermal emission spectrum by using certain coatings called "selective emitters." Using this approach, the broad blackbody emission spectrum can be condensed to a much narrower band that optimally matches the absorption spectrum of semiconductor TPV cells. In particular, we are investigating rare-earth ceramic selective emitters such as Yb and Er compounds to match the absorption of Si and Ge TPV cells. We are also engineering the microstructures of these compounds to manipulate photons for enhanced selective emitting performance.

Self-assembled nanophotonic structures for light trapping in solar cells can help with the critical task of reducing $/W for large scale applications. In conventional wafer-based solar cells, material alone constitutes more than half of the cost. Thin-film solar cells have become an attractive solution in recent years due to a drastic decrease in material consumption by 100x. The trade-off, however, is that their efficiency is limited by the thickness of the cells which is insufficient to absorb all the sunlight. Light trapping is needed to increase the optical path length of sunlight in these thin-film solar cells in order to improve the efficiency. In this research we are investigating self-assembled nanophotonic structures integrated on the backside of thin-film solar cells for sunlight trapping. The nanophotonic structure diffracts the incident sunlight into oblique angles so that it propagates laterally in the thin films and gets completely absorbed. Recent work demonstrated nanophotonic structures using self-assembled porous anodic aluminum oxide as a fabrication template for effective light trapping.

Demand for high-performance permanent magnets for motors is increasing rapidly for applications such as wind turbine generators and motors in both electric and hybrid cars. Samarium-Cobalt and Neodymium-Iron-Boron Rare Earth magnets are generally used for such challenging applications. While Rare Earth magnets are the best currently available permanent magnets, they are not without problems such as being brittle, suffering from thermal shock, and experiencing corrosion. Further, over 95% of Rare Earths are produced in China and there has been substantial price volatility. Finally, Rare Earth mining has been associated with severe environmental degradation and large energy usage. NiFe, which has been identified in meteorites as the compound Tetrataenite where it transformed from the high temperature (disordered) f.c.c. phase over thousands of years, has magnetic properties comparable to that of Rare Earth magnets. This research award develops new, low-cost, environmentally-friendly NiFe-based permanent magnets, using both quantum-mechanical calculations to predict the effects of alloying with other elements coupled with experiments to verify the effects of these additional elements on the transformation kinetics and magnetic properties. It uses the novel approach of pulsed electrical heating, which has been shown to accelerate transformations. Both women and under-represented minorities will be engaged in the research. The development of novel NiFe magnets will enable production of permanent magnets to be relocated to the USA. As part of the work, the project develops a web site that offers simple virtual experiments to explain to a wide audience magnetism and the materials science of permanent magnets.

The L1₀-structured compound Nickel-Iron (NiFe) has the potential to replace Rare Earth (RE) magnets at low cost: NiFe has a magnetic anisotropy energy, k_u, of 1.3 x 10⁶ J.m^-3 and a saturation magnetization m₀M_S of 1.59 Tesla, which is comparable to that of Nd₂Fe₁₄B. In addition, it has good corrosion resistance. The challenge is that the binary L1₀ compound has a very low transformation temperature from the high-temperature f.c.c. phase of about 320^oC that forms on casting and, thus, orders very slowly at temperatures where it is stable. This project combines ab initio quantum mechanical calculations and experimental work to design new L1₀-structured NiFe magnets with ternary elemental additions. These ternary compounds potentially have a significantly higher f.c.c.-to-L1₀ transformation temperatures and higher diffusivities than binary NiFe, but have similar saturation magnetizations. Thus, the L1₀ phase can be produced at higher temperature in short, commercially-viable times utilizing electro-pulse annealing of cold-worked material, which has also been shown to dramatically accelerate recrystallization in NiFe. The TEM-based technique ALCHEMI is used to determine the atom site locations of these ternary additions in the L1₀ unit cell. This work demonstrates a practical paradigm for designing magnets and leads to new commercially-viable permanent magnet. Commercially, NiFe can be manufactured by continuous electro-pulse annealing of rolls of sheet material or of rods. NiFe is very ductile and can easily be machined into various shapes.

Visible-blind and solar-blind ultraviolet detectors for flame and oil spill detection are useful for solving the technical challenge of operating these detectors with a large background radiation of sunlight. UV detection in the wavelength range of 210-300 nm is useful for flame, oil spill and missile detection, and traditional approaches include wide band gap semiconductor UV photodetectors and vacuum-based photoemission detectors. We are investigating more effective and less costly technologies for UV detection, including self-assembled visible light filters and solid-state photoemission detectors based on metal nanoparticles with plasmonic enhanced responsivity.

Band-engineered semiconductors and phosphor materials for efficient solid-state lighting contributes toward the development of sustainable lighting technology for the 21st century. Lighting consumes ~1 TW of power worldwide, with greenhouse gas emission equivalent to 70% of the world's car emission. Solid-state lighting using semiconductor light emitting diodes (LEDs) has been widely recognized as the next generation energy-efficient lighting technology, since LEDs feature superior efficiency, longer lifetime, and lower environmental hazard when compared with conventional lighting sources. Currently a blue LED is usually combined with phosphor coating to generate white light. There are two major challenges to this technology: (1) LED Efficiency droop at high current density required for lighting; (2) Optical energy losses due to converting blue photons to lower energy photons by phosphors. To attack the former issue we are exploring band-engineering of semiconductors near direct-indirect gap transition to achieve anti-droop properties. For the latter issue we are investigating more efficient phosphor materials especially in the yellow-red regime to enhance its conversion efficiency. We are also conducting research on long-afterglow phosphors which can store optical energy during the day and emit light at night for emergency lighting and decorative lighting.

Orthopedic implants are widely used to treat bone and joint disorders, such as fractures, osteoarthritis, and spinal deformities. However, conventional implant materials, such as stainless steel, titanium, and cobalt-chromium alloys, have several limitations: (i) they may cause adverse reactions and metal sensitivity due to their foreign ions and corrosion products; (ii) they may fail prematurely due to stress concentration and fatigue; (iii) they may interfere with bone remodeling and healing due to their mismatched mechanical properties with the host tissue; (iv) they can become a nidus for bacterial infection and biofilm colonization. Therefore, there is a need for novel implant materials that can overcome these challenges and improve the clinical outcomes of orthopedic surgery. Zinc is an attractive candidate for orthopedic implants because it is an essential trace element in the human body that plays a key role in bone metabolism and wound healing. Moreover, Zn is biodegradable and can be gradually resorbed by the body without leaving any permanent foreign material. The typical in vivo corrosion rates of unalloyed zinc in rats is 0.03 mm/yr, which is a useful rate for biodegradation. However, unalloyed Zn has very poor mechanical strength, which limits its application as an implant material. To address these issues, we are developing new Zn-based alloy that contains small amounts of silver, calcium, iron, magnesium, and manganese, and using a novel processing route. These alloying elements are chosen based on their beneficial effects on the biological and mechanical properties of Zn. Specifically, Ag has antibacterial activity and can reduce the risk of infection; Ca can promote bone formation and integration; Fe can enhance fracture fixation and blood compatibility; Mg can improve biocompatibility and corrosion resistance; Mn can increase ductility and strength. An adequate rate of implant degradation will allow the bone to heal properly before degrading and allowing the bone to support any loads.

Biomechanics analysis and monitoring following joint arthroplasty is valuable for achieving optimal recovery. Our laboratory has developed and implemented a novel method for monitoring continuous long term joint function using inertial measurement units (IMUs). Prospective studies are in progress to compare knee and shoulder function before and after arthroplasty. This data can be compared to a cohort of healthy individuals with no known joint arthropathy.

See more about biomechanics.

Demand for high-performance permanent magnets is increasing rapidly for applications such as wind turbine generators and motors in both electric and hybrid cars. This market is served by rare earth (RE) magnets based on Nd2Fe14B and Sm2Co17. RE magnets are not without issues; they can chip, suffer thermal shock, and can suffer grain boundary corrosion. However, their biggest problems are: price volatility; that China largely controls the RE metals market; and that the extraction of RE metals creates severe environmental degradation. L10-structured Tau-MnAl has been of interest as a permanent magnet since the early 1960s. It has a theoretical energy product, (BH)max, between that of AlNiCo magnets and RE magnets with a value (12 MGOe) comparable to that of bonded Nd2Fe14B magnets. Further, it does not suffer from the issues associated with RE magnets, and potentially has the lowest cost per MGOe of any permanent magnet. The enigma is that the theoretical (BH)max has never been achieved: mechanically-milled particulates can show high coercivity (HC) but low saturation magnetization (MS) while warm-extruded material can show high MS but low HC. Tau-MnAlis a metastable phase that transforms from the high temperature ε phase, during which anti-phase boundaries (APBs), twins, stacking faults and dislocations are created. Depending on the processing conditions, the equilibrium β and γ2phases can also form. The fundamental difficulty with improving the magnetic properties of Tau-MnAl is that there is no clear understanding on how they depend on the defect structure. The grain size can also influence the magnetic properties either directly or by affecting the β and γ2 arrangement and defect formation.

We are using equal channel angular extrusion (ECAE) to process Tau-MnAl billets over a range of temperatures. Multiple passes will be performed utilizing Routes A (extruded billet is fed back into the ECAE jig in the same orientation) and BC(extruded billet is rotated 90o clockwise about its axis between passes). We will determine the texture of the extruded billets, and the density and arrangement of APBs, twins, stacking faults, dislocations and second phases after each pass, and relate these to the magnetic properties. The local chemistry will be explored at high resolution using atom probe tomography via collaboration with Dr. Baptiste Gault, Max-Planck-Institut fur Eisenforschung, Germany. To extend the strain range and, hence, defect densities studied we will also explore using severe plastic deformation through a collaboration with Prof. Gheoghe Gurau, University of Galati, Romania. Our working hypothesis is that we need a strong, c-axis alignment and a low density of APBs, twins and stacking faults (which locally disorder the material) for a high MS, while a low density of APBs, twins and stacking faults but a high dislocation density are required for a high HC. It is thought that a fine distribution of β and γ2 phases will also contribute to a high HC through magnetic domain wall pinning.

Funded by the National Science Foundation.

Environmental-benign Group IV (Si, Ge, Sn) nanomaterials for solar cells, thermophotovoltaic (TPV) cells, and infrared photodetectors are being used to address the issue of sustainability as, in the case of solar cells, it is highly desirable to use naturally abundant, environmentally-friendly materials. Currently Si, CdTe and CuInGaSe2 (CIGS) are dominating solar cell materials. Although thin-film CdTe and CIGS solar cells offer higher efficiency their Si rivals, the rarity of Cd, In, and Se elements makes it impossible for them to meet a meaningful portion of global energy needs. The toxicity of Cd also raises environmental concern. By comparison, Group IV materials (Si,Ge,Sn) are very attractive due to their high abundance and low toxicity. Furthermore, (Si,Ge,Sn) alloys and nanostructures have already demonstrated many interesting properties surpassing Si, such as wide range of band gap tenability and direct gap behavior. These properties can potentially lead to better performance than CdTe and CIGS cells. We are investigating material growth, defect passivation, and band-engineering of these environmental-benign Group IV nanomaterials for solar cells and thermophotovoltaic (TPV) cells (direct conversion of thermal radiation to electricity). Similarly, in mid and far infrared photodetectors the dominant HgCdTe material may also be substituted by environmental benign, cost-effective GeSn or Sn nanostructures to achieve IR sensing and night vision.

Soft magnets play a vital role in efficient energy conversion in a variety of important applications and industries including wide-bandgap semiconductors, electric vehicles, aeronautics, and aerospace, particularly at high temperatures. Improving the efficiency of modern power electronics and electrical machines via advanced soft magnets has the potential to significantly contribute to global energy savings, thereby leading to a reduction of the associated carbon footprint. In this project, we are working on two novel FeCoMnAl high-entropy alloy (HEA) soft magnets, one of which is single-phase B2 (Fe₃₀Co₄₀Mn₁₅Al₁₅) and the other consists of an ordered B2-phase matrix enriched with Co/Al and uniformly distributed BCC nanoprecipitates enriched with Fe/Mn (Fe₄₀Co₃₀Mn₁₅Al₁₅). The two HEAs show similar properties, viz., a high saturation magnetization of 158-162 Am² kg^-1, a high Curie temperature of 1020-1081 K, a low coercivity of 108-114 A m^-1, a high electrical resistivity of ~230 µΩ cm, and good thermal stability. We are processing these HEAs using both a powder metallurgy route and via additive manufacturing. The magnetic properties and microstructures of the resulting materials are being examined using combination of a VSM, TEM, SEM and XRD examinations.

This project is investigating the synergy between the excellent mechanical behavior of FeMnNiAlCr high entropy alloys (HEAs) and the high solar absorptance of their native thermal oxides for high efficiency concentrated solar thermal power (CSP) systems working at >700^oC. The alloy itself would be used in high-temperature tubing to carry molten salts or supercritical carbon dioxide (sCO2), while the native oxide would act as a high-efficiency solar thermal absorber. The oxide layer is also dense and protective against oxidation (parabolic growth kinetics) at 750 °C. This new Fe-Mn based HEA system has already demonstrated a higher tensile strength and ductility than more expensive Ni-based Inconel 740 superalloys at both room temperature (with carbon doping) and 750^oC (with Ti doping), and their native thermal oxides have achieved a high optical-to-thermal conversion efficiency of ηtherm=90.8% at 750°C under 1000x solar concentration ratio. In preliminary corrosion studies, two-phase Cr-modified HEAs have sustained bromide molten salts for 14 days at 750°C with <2% weight loss. The project has close collaborations with the Oak Ridge National Laboratory (ORNL) and the Ames Laboratory in computational materials science and atom probe tomography (APT) to understand the fundamental structure-property relationship in FeMnNiAlCr HEAs.

Funded by DOE.

High-temperature solar absorbers for solar thermal systems are complementary to solar cells as another way of harvesting solar energy. The energy conversion efficiency increases with the working temperature of solar thermal systems according to the second law of thermodynamics. A critical component to achieve high working temperature is solar absorber coatings, which should absorb sunlight and convert it to heat efficiently with minimal thermal radiation loss in the infrared regime. They should also be stable at high temperatures. In this research we are investigating optical absorbers based on oxidation-resistant cermet materials incorporating metallic nanoparticles in a ceramic matrix for a working temperature of >560 C. Instead of using metal nanoparticles which are either easily oxidized at this temperature or too expensive to apply in large quantities, we are exploring certain metallic compounds that combines metal-like optical absorption properties with ceramic-like chemical stability. We also engineer the microstructure of the coating to trap infrared thermal radiation within the solar absorber coating and further increase the working temperature.

Integrated nanophotonics devices for high-bandwidth, ultra-low energy photonic data links offer solutions to the increase in energy consumption from computation and communication systems that has come with the rapid growth of information technology in the 21st century. Data transmission starts to consume even more energy than data processing in microprocessors, servers and data centers due to resistive losses and RC delay in electrical interconnects. Optical interconnection, on the other hand, does not have bandwidth limit or resistive losses as its electrical counterpart, thereby providing an idea solution to high bandwidth, ultralow energy data links. Our research aims to integrate nanophotonic devices monolithically on silicon chips to achieve electronic-photonic synergy, combining the merits of photons in data transmission with electrons in data processing. Recent research focuses on low-temperature (<450C) integration of nanophotonic devices such as photodetectors and photonic modulators with back end of line (BEOL) CMOS technology.

Material behavior of medical grade ultra-high molecular weight polyethylene (UHMWPE) was identified as a serious concern as it limits the overall lifetime and success of a joint replacement. Although total joint arthroplasty involving UHMWPE as a bearing surface has been one of the most successful procedures of the last century, issues of wear, oxidation, and fatigue failure remain obstacles to the longevity of joint replacements.

See more about UHMWPE material behavior.

Bearing function of retrieved knee devices sent to us by orthopaedic surgeons are assessed for damage, and also quantitatively assessed for wear. Dimensions of retrievals are compared to design specifications or shorter in-vivo duration devices to calculate both articular and backside wear. Wear and wear rate are correlated with variables including polyethylene pedigree, articular bearing geometry, device fixation, and patient factors.

Current work also includes examination of a series of reverse and total shoulders to determine the incidence of abrasive and adhesive wear and determine typical locations for these wear patterns on polyethylene components.

See more about knee/shoulder implant bearing function.

Research is conducted to determine physical processes that underlie brittle failure on scales large (Arctic) and small (laboratory). The current goal is to relate failure of the arctic sea ice cover and fracture during ice interaction with off-shore engineered structures to processes such as wing-crack and comb-crack formation and the development of shear faults. The underlying hypothesis is that brittle compressive failure is a scale-independent process driven by intermittent frictional sliding and stable crack growth. The hypothesis is applicable to other brittle materials as well, such as ceramics, rock, and minerals.

New devices research and developent for total joint arthroplasty includes:

• Testing of a new bi-material bearing for a total hip arthroplasty (THA) device against conventional bearing designs to compare levels of bearing surface damage and wear;
• Development of an intraoperative method for quantifying the orientation of prosthetic components used in total knee arthroplasty (TKA) that is efficient, easy to use, cost effective, and quick with respect to total surgical time.

See more about new devices.

New materials research and development for orthopaedic implants includes:

• Evaluation of equal channel angular extrusion (ECAE)-processed ultra-high molecular weight polyethylene (UHMWPE) for joint arthroplasty and industrial applications;
• Investigation of off-label use of a resorbable calcium sulfate antibiotic carrier in single stage and two-stage procedures to determine the potential of this use to change damage patterns or wear rates of artificial joints.

See more about new materials.

The microstructual evolution of snow under a temperature gradient has been of interest for many years since this can lead to persistent weak layers, which are possible microstructural causes of avalanches. Ice crusts can form on top or within a snowpack from a variety of meteorological conditions including significant melt/freeze or freezing rain events, and once buried, they can persist throughout the entire winter season and act as an ideal sliding surface for dangerous slab avalanches in seasonal mountain snowpacks. Both of these phenomena are important because the number of fatalities from avalanches in the US has increased annually since the 1970s. Avalanches can also have substantial economic impacts due to road closures, the costs of rescue and building damage, and, with continued global warning, more avalanches are expected in Arctic regions. To understand avalanche nucleation, we are deforming two types of specimens (heterogeneously-layered snow and snow containing an ice lens) in a micro CT located in a cold room, in which the specimens are repeatedly imaged during loading. We are also performing more macroscopic deformation experiments on larger samples at both different rates and different temperatures, which are imaged using a high-speed video camera during loading. The final deformed microstructures in both cases are imaged at high resolution using a scanning electron microscope, which provides information on both the effects of crystal orientation on deformation while clearly delineating one ice crystal orientation from another. Based on the experimental observations, a multiscale computational model is being built to understand crack initiation/crack propagation as well as the deformation mechanisms in heterogenously-layered snow samples containing persistent weak layers and ice/snow interfaces.

Implant failure analysis in the Dartmouth Biomedical Engineering Center for Orthopaedics is ongoing and plays a key role in identifying failure modes and relating them to various designs and materials being used in the industry. In fact, in 2000, NIH's Consensus Development Program produced a technology assessment statement acknowledging the value of implant retrieval programs:

• Implant retrieval and analysis is of critical importance in the process of improving care of patients in need of implants.
• Attention needs to be directed toward reducing various obstacles to implant retrieval and analysis, particularly legal and economic disincentives.
• The failure to appreciate the value of implant retrieval and analysis is a serious impediment to research in devices. A focused educational program will provide the information necessary for improving the quality of future devices.

See more about implant failure analysis.

Porous thermoelectric cells (TECs) are being developed for waste heat recovery. This project seeks to convert waste thermal energy directly into electricity potentially increasing overall energy efficiency by 15-20% and providing new portable electric power sources, particularly for cold regions. The project is focusing on low cost, nanostructure-engineered TEC (NETECs) materials based on earth-abundant, highly machinable metallic alloys and intermetallic compounds by engineering the grain sizes, second phase precipitation, and nanopores in transition metal intermetallic compounds. The project is currently focusing on the compound Fe2AlV. The location of quaternary atoms in the lattice is being determined by the TEM-based technique ALCHEMI.

Funded by USA-CRREL

Rare-earth selective emitters for high efficiency TPV systems address the major limiting factor for the efficiency of TPV cells: the broad emission spectrum from the heat source. Photons with energy smaller than the band gap of semiconductor PV materials cannot be absorbed, while for those with energy much greater than the band gap the excess energy is lost to heat instead of generating electricity. Although there is nothing we can do to change the solar spectrum, for thermal sources of TPV systems we may engineer the thermal emission spectrum by using certain coatings called "selective emitters." Using this approach, the broad blackbody emission spectrum can be condensed to a much narrower band that optimally matches the absorption spectrum of semiconductor TPV cells. In particular, we are investigating rare-earth ceramic selective emitters such as Yb and Er compounds to match the absorption of Si and Ge TPV cells. We are also engineering the microstructures of these compounds to manipulate photons for enhanced selective emitting performance.

Self-assembled nanophotonic structures for light trapping in solar cells can help with the critical task of reducing $/W for large scale applications. In conventional wafer-based solar cells, material alone constitutes more than half of the cost. Thin-film solar cells have become an attractive solution in recent years due to a drastic decrease in material consumption by 100x. The trade-off, however, is that their efficiency is limited by the thickness of the cells which is insufficient to absorb all the sunlight. Light trapping is needed to increase the optical path length of sunlight in these thin-film solar cells in order to improve the efficiency. In this research we are investigating self-assembled nanophotonic structures integrated on the backside of thin-film solar cells for sunlight trapping. The nanophotonic structure diffracts the incident sunlight into oblique angles so that it propagates laterally in the thin films and gets completely absorbed. Recent work demonstrated nanophotonic structures using self-assembled porous anodic aluminum oxide as a fabrication template for effective light trapping.

By computing properties of all known materials, The Materials Project aims to remove guesswork from materials design in a variety of applications. Experimental research can be targeted to the most promising compounds from computational data sets. Researchers will be able to data-mine scientific trends in materials properties. By providing materials researchers with the information they need to design better, The Materials Project aims to accelerate innovation in materials research.

Demand for high-performance permanent magnets for motors is increasing rapidly for applications such as wind turbine generators and motors in both electric and hybrid cars. Samarium-Cobalt and Neodymium-Iron-Boron Rare Earth magnets are generally used for such challenging applications. While Rare Earth magnets are the best currently available permanent magnets, they are not without problems such as being brittle, suffering from thermal shock, and experiencing corrosion. Further, over 95% of Rare Earths are produced in China and there has been substantial price volatility. Finally, Rare Earth mining has been associated with severe environmental degradation and large energy usage. NiFe, which has been identified in meteorites as the compound Tetrataenite where it transformed from the high temperature (disordered) f.c.c. phase over thousands of years, has magnetic properties comparable to that of Rare Earth magnets. This research award develops new, low-cost, environmentally-friendly NiFe-based permanent magnets, using both quantum-mechanical calculations to predict the effects of alloying with other elements coupled with experiments to verify the effects of these additional elements on the transformation kinetics and magnetic properties. It uses the novel approach of pulsed electrical heating, which has been shown to accelerate transformations. Both women and under-represented minorities will be engaged in the research. The development of novel NiFe magnets will enable production of permanent magnets to be relocated to the USA. As part of the work, the project develops a web site that offers simple virtual experiments to explain to a wide audience magnetism and the materials science of permanent magnets.

The L1₀-structured compound Nickel-Iron (NiFe) has the potential to replace Rare Earth (RE) magnets at low cost: NiFe has a magnetic anisotropy energy, k_u, of 1.3 x 10⁶ J.m^-3 and a saturation magnetization m₀M_S of 1.59 Tesla, which is comparable to that of Nd₂Fe₁₄B. In addition, it has good corrosion resistance. The challenge is that the binary L1₀ compound has a very low transformation temperature from the high-temperature f.c.c. phase of about 320^oC that forms on casting and, thus, orders very slowly at temperatures where it is stable. This project combines ab initio quantum mechanical calculations and experimental work to design new L1₀-structured NiFe magnets with ternary elemental additions. These ternary compounds potentially have a significantly higher f.c.c.-to-L1₀ transformation temperatures and higher diffusivities than binary NiFe, but have similar saturation magnetizations. Thus, the L1₀ phase can be produced at higher temperature in short, commercially-viable times utilizing electro-pulse annealing of cold-worked material, which has also been shown to dramatically accelerate recrystallization in NiFe. The TEM-based technique ALCHEMI is used to determine the atom site locations of these ternary additions in the L1₀ unit cell. This work demonstrates a practical paradigm for designing magnets and leads to new commercially-viable permanent magnet. Commercially, NiFe can be manufactured by continuous electro-pulse annealing of rolls of sheet material or of rods. NiFe is very ductile and can easily be machined into various shapes.

Visible-blind and solar-blind ultraviolet detectors for flame and oil spill detection are useful for solving the technical challenge of operating these detectors with a large background radiation of sunlight. UV detection in the wavelength range of 210-300 nm is useful for flame, oil spill and missile detection, and traditional approaches include wide band gap semiconductor UV photodetectors and vacuum-based photoemission detectors. We are investigating more effective and less costly technologies for UV detection, including self-assembled visible light filters and solid-state photoemission detectors based on metal nanoparticles with plasmonic enhanced responsivity.

With support from the Association of American Universities (AAU), we are creating a system for teaching evaluation that integrates evidence-based evaluation data from three sources: student course assessments, peer observation, and self-reflection. Existing student course assessments are being augmented to focus on observable best practices, rather than students' intuitive impression of the instructor. Peer observations are conducted in accordance with a consistent protocol, which includes guided pre-briefing and debrief sessions between the observer and observee. Self-reflection allows each instructor to note and report on their progress toward the implementation of best teaching practices. The system is currently being piloted and assessed en route to broader implementation. Learn more

Acoustics and signal processing research focuses on active noise control and distributed sensing. Active noise control reduces noise in hearing protection and communication systems to reduce noise induced hearing loss and to enhance the ability to communicate. Distributed sensing research uses signal processing to focus listening in a specified direction. This research blends mechatronics—the design of mechanical and electrical systems—with high performance signal processing and control algorithms to improve communication in noisy environments.

Many communities are not prepared to understand the changes they may need to make under our currently changing climate in order to adapt and mitigate for a more resilient future. We are working with communities and with the U.S. Army Corps of Engineers to use published forecasts of climate change along with economics, planning policy, and engineering considerations in order to facilitate planning for a more resilient future for communities within the U.S. and elsewhere.

Adversary intent inferencing and adversarial modeling investigates the feasibility of developing and utilizing an adversary intent inferencing model as a core element for predictive analyses and simulations to establish emergent adversarial behavior. It is our desire to use this intelligent adversary to predict adversary intentions, explain adversary goals, and predict enemy actions in an effort to generate alternative futures critical to performing course of action (COA) analysis. Such a system will allow planners to gauge and evaluate the effectiveness of alternative plans under varying actions and reactions to friendly COAs. This can also be applied in a broad range of areas.

Agent-based systems engineering aims to successfully cross-fertilize the fields of systems engineering and artificial intelligence. Systems engineering (control, signal processing and communications) focuses primarily on physical domains that can be characterized by rich mathematical dynamics while artificial intelligence deals with human perception, decision making and action. Goals of such cross-fertilization are to explore the modeling, performance and scientific foundations of software agent systems using ideas from classical systems engineering and computer engineering.

What is the nature of intelligence? Can we make machines that are intelligent? Machines that think like human beings or think differently? Can machines think even better than humans? What are the implications? These and other questions are being investigated.

Bayesian knowledge bases, engineering, verification, and validation focuses on the fundamental problem of probabilistic modeling of knowledge in order to represent and reason about information in a theoretically sound manner. The world is replete with issues such as incompleteness, impreciseness, and inconsistency which makes the task of capturing even everyday tasks, processes, and activities very difficult, let alone trying to capture that of decision-making by experts or other complex phenomena. Improperly modeling uncertainty leads to numerous anomalies in reasoning as well as increased computational difficulties.

In large-scale multi-agent systems like robotic fleets, distributing the computational and decision-making tasks to be performed locally by individual robots (or agents) is a popular method of reducing the complexity and communication overhead of a fleet level control policy. However, the emergent behavior of these distributed systems can be inferior to those that are controlled in a centralized manner. This project seeks to understand how agents can partially coordinate their behavior by collaborating on group control actions. Intended outcomes are control algorithms for new, collaborative paradigms for multi-agent systems and formal guarantees on the behavior of these systems.

Biomechanics analysis and monitoring following joint arthroplasty is valuable for achieving optimal recovery. Our laboratory has developed and implemented a novel method for monitoring continuous long term joint function using inertial measurement units (IMUs). Prospective studies are in progress to compare knee and shoulder function before and after arthroplasty. This data can be compared to a cohort of healthy individuals with no known joint arthropathy.

See more about biomechanics.

In existing attempts to model complex systems, one critical aspect that has not been clearly ad-dressed involves the underlying mechanism for integrating the numerous “pieces” and “parts” that make up the target. Combining pieces is the process of aggregation and must handle inconsistencies among the pieces. Combining parts is the process of composition in which the parts are encapsulations of information with a set of meaningful operations defined on them. Parts are functional in nature and thus are driven by function composition. Extant research has not directly addressed this resulting in mathematically ad-hoc models opaque to analysis. We propose to develop a singular, rigorous, comprehensive computation framework that is axiomatic and provides the capabilities needed to model complex systems based on a new model of complex adaptive Bayesian Knowledge Bases and a novel, powerful analytical framework capable of wholistic end-to-end quantitative analysis of performance, robustness, vulnerability, and impacts of change on our targets being modeled. Furthermore, our results will be applicable to numerous domains of public purpose from crisis and catastrophe management for natural disasters and disease outbreaks to assessing the well-being of our financial system and national infrastructures.

Connections Hypothesis Provider (CHP) service built by Dartmouth College (PI – Dr. Eugene Santos) and Tufts University (Co-PI – Joseph Gormley) in collaboration with the National Center for Advancing Translational Sciences (NCATS). CHP aims to leverage clinical data along with structured biochemical knowledge to derive a computational representation of pathway structures and molecular components to support human and machine-driven interpretation, enable pathway-based biomarker discovery, and aid in the drug development process. In its current version, CHP supports queries relating to genetic, therapeutic, and patient clinical features (e.g. tumor staging) contribution toward patient survival, as computed within the context of our test pilot: a robust breast cancer dataset from The Cancer Genome Atlas (TCGA). We are using this as a proving ground for our system’s basic operations as we work to incorporate structured pathway knowledge and pathway analysis methods into the tool.

Convection electric field climatology and variability focuses on the large-scale convection of plasma in the Earth's polar ionospheres. Measurements from the SuperDARN network of HF radars are used to look at statistical properties of convection associated with various drivers as well as the observed small-scale variability.

Cooperative control of multi-robot systems focuses on modeling and control of groups of high-speed mobile robots while accommodating communication latencies and nonlinear vehicle dynamics. In distributed cooperative control, robots communicate information about their state to each other; communication latencies and error depends on the amount of information communicated and the number of robots. We are developing distributed control system modeling and design tools that seek to maximize control bandwidth for a given information set. These tools will also assist in assessing the value of information transmitted in maintaining stability and performance of group dynamics. Both potential function path planning and control and predictive control methods are being developed.

Deception detection aims to automatically detect and infer the intentions behind deceptive actions. Our objectives are to 1) develop a framework for categorizing and classifying errors that may be committed by an expert, since not all errors are deception; and 2) design algorithms for automatic deception detection capable of providing detailed evidential information and explanation of deception intent, plus analysis of the deception's impact. Like insider threat, deception detection can occur in any number of scenarios and domains, and insider threat and deception detection are often interrelated.

Distributed information retrieval aims to develop a large-scale information retrieval architecture that can be effectively and efficiently deployed in distributed environments. Heterogeneous information (such as content, formats and sources) is the typical issue that needs to be identified and handled in the distributed environment. Our objective is to develop a unified architecture called I-FGM (intelligent foraging, gathering and matching) for dealing with the massive amount of information in a dynamic search space within large-scale distributed platforms. The system will proceed to explore the information space, and continuously identify and update promising candidate information. Specific metrics are also being developed for performance evaluation.

Environmental fluid mechanics research studies natural fluid systems as agents for the transport and dispersion of environmental contamination. Understanding transport and dispersion processes in natural fluid flows, from the microscale to the planetary scale, serves as the basis for the development of models aimed at simulations, predictions, and ultimately sustainable environmental management. Research within this scope is diverse and can involve a variety of scientific and engineering disciplines such as civil, mechanical, and environmental engineering, meteorology, hydrology, hydraulics, limnology, and oceanography.

As perhaps the most dynamic of all naturally-deposited porous media, the changing structure of snow reflects changes in weather and the environment. A wealth of satellite imagery exists over the vast expanses of the Antarctic and Greenland ice sheets, yet there is an incomplete understanding of what variations in the images means. From snow and ice cores collected from traverses across the Antarctic Ice Sheet and from several locations on the Greenland Ice Sheet, we are developing the physical basis by which satellite imagery can be used to make high-resolution maps of snow accumulation over vast and remote areas of the ice sheets.

Analyzing potential synergies between deep decarbonization of the energy system and reducing (inequities in) negative health impacts. MORE

High performance search and optimization aims to develop new models and algorithms for solving challenging engineering problems in domains such as mission planning and logistics, manufacturing process optimization, composite materials production, distributed plant scheduling and management, and policy evaluation, to name a few.

Ice cores drilled in cold areas of the Greenland and Antarctic Ice Sheets provide high-resolution climate records that are essential for understanding abrupt climate change. The only remaining samples of the atmopshere from past centuries and millennia are contained in bubbles found deep in glacial ice. We are measuring the physical structure, transport properties, and microstructure from ice cores from Greenland and Antarctica to better understand mechanisms of the trapping of gases in ice cores for improved understanding of abrupt climate changes in the past.

Influence of culture and society on attitudes and behaviors aims to build and employ social, cultural, and political data-driven models to explore and explain attitudes and behaviors. The efforts involve classifying the factors that play significant roles in attitudes and behaviors, abstracting general rules from traditional research such as sociological case studies, studying the inferencing structures that allow different factors to influence decision-making, reasoning from different points of view, and applying them in predicting behavior.

In social systems, like transportation networks, human users make decisions despite having uncertainty about the underlying state of the system (e.g., the current state of traffic). If a central authority is able to gather this information (e.g., Google or Apple maps), then they are presented with the opportunity to reveal that information to the system's users or strategically reveal pieces of information as a mechanism to shape the users' beliefs and ultimate behavior. In this project, we seek to identify policies for strategically signaling information to incentivize more desirable decisions by the systems users. Intended outcomes will be insights on how to design information signaling policies as well as as formal analysis on the effects of heterogeneity among the human user population.

Information processing and summarization are critical areas of research that study how we can develop stand-alone algorithms as well as algorithms fused with humans to handle and process information in a variety of forms. The goal is to be able to extract the meaning (or semantics) of the information in order to better manipulate/reason and present it to the human user. This is fundamental to solving problems such as avoiding information overload and providing effective summarization.

Can we teach computers to think outside the box? In other words, is it possible to replicate the innovative decision process computationally for machine learning? Extant research in machine learning has typically either focused on (a) building predictive models of a single internally-consistent target or on (b) a single task or decision in isolation, and more rarely, both, given the difficulties already posed within these more restrictive problems. The successes and utility of modern machine learning is clearly evident in numerous applications across many domains, and ever more so now with Big Data. Yet, the former focus (a) has made machine learning of complex targets (e.g. systems of systems, complex systems, a human) very elusive because of their inherent assumptions of expectations. This precludes the ability to learn emergent, unexpected, or innovative behaviors.

Insider threat and deception detection are two areas that focus on user actions and their impacts upon the systems with which they interact. Insider threat aims to understand and prevent malicious activities that are instigated by "trusted" users on complex computer/information systems. Such activities cover a broad spectrum ranging from simple theft of confidential data to the more subtle alteration of system performance and/or information. For the latter, examples can include minor perturbation of a component specification in a manufacturing process resulting in a rippling effect of final component failure to influencing the decision-makers by modifying their information flow and content. The goal is to model insider threat in order to predict behavior and ultimately infer their goals and intentions.

We are developing algorithms and a complementary metal oxide semiconductor (CMOS) imager to perform high speed voltage imaging of brain activities.

The Empower Lab creates digital therapeutics and intervention technologies, with an emphasis on inclusive design, patient experience, and health equity.

We seek students to contribute to the design, development, and evaluation of our interactive tools, which aim to positively shape behaviors and health outcomes through an empowerment approach. Our work employs human-centered methods and explores a variety of form factors and interaction paradigms (e.g., AR/VR/XR, games, toys, tangibles, musical interfaces, narratives, psychotherapeutic visualization, socially assistive robots, and more).

We work closely with stakeholder populations on a range of health topics; key areas of focus include mental health, women’s health, physical mobility/pain, aging-in-place, and pediatrics/childhood development. More details about our lab and its active projects, including specific recruiting opportunities, can be found here: empowerlab.dartmouth.edu/projects

Generally speaking, our collaborative research style integrates interdisciplinary skills from both computing and engineering as well as the social sciences and humanities. Students typically focus on contributing to one or two aspects of a project’s research activities, which include iterative design (UX, prototyping), implementation (software programming, physical fabrication), data analysis (quantitative/statistical or qualitative), stakeholder engagement (conducting interviews, observational or ethnographic work), and running studies (user testing, lab experiments, online or survey studies, or field trials).

If interested in getting involved, you can fill out our application form dartgo.org/empowerlab-apply or email the lab director, Prof. Liz Murnane (emurnane@dartmouth.edu) to share your resume, relevant interests and background, and any questions you might have. We look forward to hearing from you!

Iterative learning control refers to the mechanism by which the necessary control can be synthesized by repeated trials—based on the fundamental recognition that repeated practice is a common mode of human learning. Learning control is most suitable for operations where the same task is to be performed over and over again, e.g., robots in a manufacturing line. Available learning techniques range from those requiring no knowledge of the system dynamics to more sophisticated methods involving system identification to make the learning process efficient and successful on difficult problems. Our research finds ways to design optimal iterative learning controllers that are robust to model uncertainty, and capable of producing monotonic convergence.

Material behavior of medical grade ultra-high molecular weight polyethylene (UHMWPE) was identified as a serious concern as it limits the overall lifetime and success of a joint replacement. Although total joint arthroplasty involving UHMWPE as a bearing surface has been one of the most successful procedures of the last century, issues of wear, oxidation, and fatigue failure remain obstacles to the longevity of joint replacements.

See more about UHMWPE material behavior.

Bearing function of retrieved knee devices sent to us by orthopaedic surgeons are assessed for damage, and also quantitatively assessed for wear. Dimensions of retrievals are compared to design specifications or shorter in-vivo duration devices to calculate both articular and backside wear. Wear and wear rate are correlated with variables including polyethylene pedigree, articular bearing geometry, device fixation, and patient factors.

Current work also includes examination of a series of reverse and total shoulders to determine the incidence of abrasive and adhesive wear and determine typical locations for these wear patterns on polyethylene components.

See more about knee/shoulder implant bearing function.

Researching complex interactions between climate hazards and communities to inform governance of coastal risk. MORE

Successful next generation AI/ML for real-world applications must be able to deal with incomplete, sparse and noisy data as well as unexpected or adversarial circumstances that might arise while solving real world military problems. Furthermore, successful models must be able to learn new concepts with few examples. Unfortunately, even in this age of commodity machine learning models, one still needs to train new models with a large amount of training examples to achieve the requisite performance from deep neural networks. The resulting models often lack robustness, explainability, and human-level intelligence.

One of the best models for intelligence is the one inspired by human brain itself, which can support robust and massive parallelism with ease. We are exploring two related aspects of human brain processing that will play a key role in a Third Wave AI revolution: (1) a meso-scale cortical computation that easily finds low dimensional manifolds where learning can naturally take place, and (2) recurrent module networks, as thin as one- or two-layer networks, akin to cellular automata, trained on time sequences rather than input-output pairs. We believe this process of simplification and automata implementation is a better model of the human neocortex than current state-of-the-art deep networks whose optimized architectures are often ad hoc and do not reflect biological reality. This project is being sponsored in part by DARPA.

A network of HF radars are being built at several mid-latitude locations to measure ionospheric electric fields. This project involves the construction, maintenance and operation of several HF SuperDARN radars that have recently been built or are in the planning stages.

Model predictive control is control action based on a prediction of the system output a number of time steps into the future. Originated from chemical process engineering, model predictive control has found its way into virtually all areas of control engineering. Our research focuses on the development of a general formulation of predictive control that subsumes both the input-output and state-space perspectives. We seek comprehensive answers to questions such as: What is the simplest way to justify the existence and structures of various input-output predictive models? How does one arrive at an input-output controller if the starting point of the derivation is a state-space model? Can explicit state-space model identification be avoided? What is an efficient strategy to synthesize a predictive controller from input-output data directly without having to resort to model identification? What is the role of predictive control in the disturbance rejection problem? How can we design model predictive controllers for a swarm of robots?

Real-world complex systems can be observed from many different angles or perspectives, and datasets collected from various perspectives often emphasize different types of features. This results in inconsistent beliefs about what is relevant to the system, how relevant features are related to one another, and what statistical properties these features possess. Many methods have been proposed to combine such diverse information sources. However, current algorithms only learn from each dataset separately and then combine individual outputs since this is easier to do with heterogeneous datasets with unknown feature correlations. This approach, although convenient and intuitive, cannot capture the logical linkages between various datasets. To understand variables’ interactions learned from datasets with noise and incompleteness, we are exploring algorithms that naturally fuse these datasets based on their shared variables and induce new variable relationships.

New devices research and developent for total joint arthroplasty includes:

• Testing of a new bi-material bearing for a total hip arthroplasty (THA) device against conventional bearing designs to compare levels of bearing surface damage and wear;
• Development of an intraoperative method for quantifying the orientation of prosthetic components used in total knee arthroplasty (TKA) that is efficient, easy to use, cost effective, and quick with respect to total surgical time.

See more about new devices.

New materials research and development for orthopaedic implants includes:

• Evaluation of equal channel angular extrusion (ECAE)-processed ultra-high molecular weight polyethylene (UHMWPE) for joint arthroplasty and industrial applications;
• Investigation of off-label use of a resorbable calcium sulfate antibiotic carrier in single stage and two-stage procedures to determine the potential of this use to change damage patterns or wear rates of artificial joints.

See more about new materials.

Reinforcement learning has shown much success in games such as chess, backgammon, and Go. However, in most of these games, agents have full knowledge of the environment at all times. We describe a deep learning model that successfully maximizes its score using reinforcement learning in a game with incomplete and imperfect information. We apply our model to FlipIt (1), a two-player game in which both players, the attacker and the defender, compete for ownership of a shared resource and only receive information on the current state upon making a move. Our model is a deep neural network combined with Q-learning and is trained to maximize the defender's time of ownership of the resource. We extend FlipIt to a larger action-spaced game with the introduction of a new lower-cost move and generalize the model to multiplayer FlipIt.

(1) van Dijk, M, Juels, A, Oprea, A, Rivest, RL, FlipIt: The Game of "Stealthy Takeover." Journal of Cryptology, 26, 655–713 (2013)

To advance the science of decision-making as it pertains to how people learn to make decisions and how this process can be captured computationally, we are specifically addressing the challenge of how nonlinear decisions can be learned from data, experience, and even interactions with other decision-makers. Nonlinear thinking is a prized ability we, humans, have that is ubiquitously applied across any and all domains when the problems are challenging, and known solutions or ways of addressing the problems all fail to provide an adequate solution – e.g., All available choices are bad choices, must we settle for the least bad one? The ability to discover a new choice has been called being nonlinear, innovative, intuitive, emergent, or “outside-the-box.” It is well-documented that humans can often excel at such thinking in situations when there is a scarcity/overflow of data, significant uncertainty, and numerous contradictions in what is known or provided. However, how this can be replicated computationally for a machine has yet to be fully addressed or understood in extant research.

Implant failure analysis in the Dartmouth Biomedical Engineering Center for Orthopaedics is ongoing and plays a key role in identifying failure modes and relating them to various designs and materials being used in the industry. In fact, in 2000, NIH's Consensus Development Program produced a technology assessment statement acknowledging the value of implant retrieval programs:

• Implant retrieval and analysis is of critical importance in the process of improving care of patients in need of implants.
• Attention needs to be directed toward reducing various obstacles to implant retrieval and analysis, particularly legal and economic disincentives.
• The failure to appreciate the value of implant retrieval and analysis is a serious impediment to research in devices. A focused educational program will provide the information necessary for improving the quality of future devices.

See more about implant failure analysis.

Advancing the quantitative understanding of multisector dynamics. MORE

Process query systems have applications that involve using databases or datastreams of events to detect instances of processes. In those applications, events provide evidence that is used to infer the existence and estimate the states of the various processes of interest. Examples of such applications include: network and computer security; network management; sensor network tracking; military situational awareness; and critical infrastructure monitoring and protection.

Robot design and smart navigation focuses on developing affordable robot designs that employ "smart navigation" for path planning and mobility in extreme terrain, rather than complex and expensive vision systems. We are developing solar-powered robotic platforms for deploying scientific instrumentation over hundreds of kilometers in Arctic and Antarctic regions. These robots employ proprioceptive sensors to determine whether difficult terrain is passable, and if not, to navigate around such terrain.

As systems grow large and additional communication channels are implemented, new avenues for noise, hazards, and bad actors are introduced. Unlike in a single-agent system, when many agents are connected, individual failures or attacks may be hard to detect, and local perturbations can cascade into much more expansive failures. This project studies different adversarial and risk-aware design environments through the lens of zero-sum game theory. Intended outcomes are secure strategies in networked adversarial environments and fundamental analysis of the relationship between information and the ability to guarantee security.

Soft computing has experienced major advances in actuator and sensor technology, computing technology, and the emergence of a collection of new tools that can solve problems in an unconventional yet effective way:

• Artificial neural networks are constructed from identical data processing elements arranged in some regular pattern. These networks exhibit surprising abilities to capture non-linear relationships among variables, perform pattern classification and feature extraction, and encode associate memory, among others.
• Fuzzy logic can emulate human-like rule-based operations using linguistic terms such as "if it starts to become hot, turn the temperature down a little bit."
• Genetic algorithms give us a new way to perform optimization without actually solving equations in the traditional sense.

Other soft computing techniques such as DNA computing and simulated annealing are also very intriguing. Our research finds ways to apply these tools to problems such as the control of a magneto-hydrodynamic power generators for hypersonic aircraft, and the evolution of a robot's rule base for obstacle avoidance and target acquisition.

We are developing a framework using advanced tools from random graph theory and spectral graph theory to carry out the quantitative analysis of the structure and dynamics of large networks—with the focus of graph merging and subgraph detection. This framework, using information theory as a demarcative tool, enables one to carry out analytic computations of observable network structures and capture the most relevant and refined quantities of real-world networks.

Sub-auroral convection electric fields focuses on the study of ionospheric plasma convection in the plasmasphere boundary layer region where complex coupling between the ionosphere and magnetotail occur during geomagnetically disturbed periods. Measurements from the mid-latitude SuperDARN radars are used in conjunction with other ground and space-based observations of sub-auroral phenomena.

System identification refers to the general process of extracting information about a system from measured input-output data. A typical outcome is an input-output model which may be static or dynamic, deterministic or stochastic, linear or nonlinear. One can use such an input-output model for simulation, controller design, or analysis. System identification can extract the physical properties of a system such as its mass, stiffness, and damping distribution. System identification methods can also be applied to obtain information other than a model of a system. For example, it can be used to identify an observer or Kalman filter gain, existing feedback controller gain, disturbance environment, or to detect actuator and sensor failure. The same theory can even be used to synthesize feedback or feedforward controller gains directly from input-output data without having to obtain an intermediate model of the system first. System identification has widespread applications in virtually all areas of engineering including chemical, electrical, mechanical, biomedical, aerospace engineering, and economics.

Terrain identification research focuses on using small, lightweight robots to classify, characterize, and identify terrain properties necessary to predict mobility of these vehicles on the terrain. Terramechanics models for heavy vehicles are well understood, but similar comprehensive models do not exist for lightweight (sub-500 kg) vehicles. We are developing terrain models and modeling tools that can be used to asses mobility on a given terrain, while avoiding maneuvers that cause immobilization. We seek to integrate terrain identification and traction/stability control of the robots in order to allow autonomous or remote control of these robots at the maximum attainable speeds and accelerations achievable on the terrain.

We are investigating how to combine recent advances in topological data analysis with recent advances in machine learning to enhance multiple hypothesis tracking system. Our goal is to improve detecting, clustering, classifying and tracking of various patterns-of-life trajectories by developing the capability to distinguish behavioral types at all scales.

User modeling and user intent inferencing involves building dynamic cognitive user models that can predict the goals and intentions of a user in order to understand and ultimately provide proactive assistance with user tasks, such as information gathering. The key is to capture the user's intent by answering questions such as: what is the user's current focus, why is the user pursuing certain goals, and how will the user achieve them? The efforts involve machine learning, knowledge representation, intent inferencing, and establishment of proper evaluation metrics. This work has been applied to assisting with intelligent information retrieval and enhancing the effectiveness of intelligence analysts.

With support from the Association of American Universities (AAU), we are creating a system for teaching evaluation that integrates evidence-based evaluation data from three sources: student course assessments, peer observation, and self-reflection. Existing student course assessments are being augmented to focus on observable best practices, rather than students' intuitive impression of the instructor. Peer observations are conducted in accordance with a consistent protocol, which includes guided pre-briefing and debrief sessions between the observer and observee. Self-reflection allows each instructor to note and report on their progress toward the implementation of best teaching practices. The system is currently being piloted and assessed en route to broader implementation. Learn more

Biomechanics analysis and monitoring following joint arthroplasty is valuable for achieving optimal recovery. Our laboratory has developed and implemented a novel method for monitoring continuous long term joint function using inertial measurement units (IMUs). Prospective studies are in progress to compare knee and shoulder function before and after arthroplasty. This data can be compared to a cohort of healthy individuals with no known joint arthropathy.

See more about biomechanics.

A significant barrier to discovering contraceptives is the lack of systems for exploring the molecular features of ovarian follicles (an oocyte surrounded by somatic cells in the ovary), the drivers of ovulation, and for screening for contraceptive potential. My lab, in collaboration with a well-established network of ovarian collaborators will leverage these systems to enable contraceptive discovery. We will do this by identifying early signals of ovulation prior to tissue remodeling to identify targets that can preserve fertility. This project will involve substantial analyses and integration of transcriptomic data, with opportunities to develop rational frameworks for discovering druggable targets from these data types.

DIFUSE is an NSF-funded Dartmouth project aimed at creating opportunities for undergraduates to learn and use data science in introductory STEM courses and beyond. We work with teams of undergraduates, PhD students, and faculty to develop data science "modules" to integrate into existing course curriculum. We also offer opportunities for undergraduate and PhD students to apply data science and data visualization through Internships and work in the DALI Lab.

Goal: to provide an imaging tool to effectively monitor the long-term physiological effects

of deep space and automate diagnosis, enabling crew members to be proactive in the

event of injury. More specifically, we are designing an integrated US-EIT system and

demonstrate proof-of-concept of US-EIT for enhanced ultrasound imaging capabilities of

deep internal bleeding.

Environmental fluid mechanics research studies natural fluid systems as agents for the transport and dispersion of environmental contamination. Understanding transport and dispersion processes in natural fluid flows, from the microscale to the planetary scale, serves as the basis for the development of models aimed at simulations, predictions, and ultimately sustainable environmental management. Research within this scope is diverse and can involve a variety of scientific and engineering disciplines such as civil, mechanical, and environmental engineering, meteorology, hydrology, hydraulics, limnology, and oceanography.

Analyzing potential synergies between deep decarbonization of the energy system and reducing (inequities in) negative health impacts. MORE

The interaction between the endocrine system and immune system remains poorly understood. The goal of this project is to use single-cell methods to better understand how hormones impact immune cell differentiation and function, focusing initially on innate immune cells. This will ultimately allow us to tailor immune responses based on hormonal contexts and sex differences, while enabling the creation of a more integrated and complete model of immune cell functions.

The Empower Lab creates digital therapeutics and intervention technologies, with an emphasis on inclusive design, patient experience, and health equity.

We seek students to contribute to the design, development, and evaluation of our interactive tools, which aim to positively shape behaviors and health outcomes through an empowerment approach. Our work employs human-centered methods and explores a variety of form factors and interaction paradigms (e.g., AR/VR/XR, games, toys, tangibles, musical interfaces, narratives, psychotherapeutic visualization, socially assistive robots, and more).

We work closely with stakeholder populations on a range of health topics; key areas of focus include mental health, women’s health, physical mobility/pain, aging-in-place, and pediatrics/childhood development. More details about our lab and its active projects, including specific recruiting opportunities, can be found here: empowerlab.dartmouth.edu/projects

Generally speaking, our collaborative research style integrates interdisciplinary skills from both computing and engineering as well as the social sciences and humanities. Students typically focus on contributing to one or two aspects of a project’s research activities, which include iterative design (UX, prototyping), implementation (software programming, physical fabrication), data analysis (quantitative/statistical or qualitative), stakeholder engagement (conducting interviews, observational or ethnographic work), and running studies (user testing, lab experiments, online or survey studies, or field trials).

If interested in getting involved, you can fill out our application form dartgo.org/empowerlab-apply or email the lab director, Prof. Liz Murnane (emurnane@dartmouth.edu) to share your resume, relevant interests and background, and any questions you might have. We look forward to hearing from you!

Material behavior of medical grade ultra-high molecular weight polyethylene (UHMWPE) was identified as a serious concern as it limits the overall lifetime and success of a joint replacement. Although total joint arthroplasty involving UHMWPE as a bearing surface has been one of the most successful procedures of the last century, issues of wear, oxidation, and fatigue failure remain obstacles to the longevity of joint replacements.

See more about UHMWPE material behavior.

Bearing function of retrieved knee devices sent to us by orthopaedic surgeons are assessed for damage, and also quantitatively assessed for wear. Dimensions of retrievals are compared to design specifications or shorter in-vivo duration devices to calculate both articular and backside wear. Wear and wear rate are correlated with variables including polyethylene pedigree, articular bearing geometry, device fixation, and patient factors.

Current work also includes examination of a series of reverse and total shoulders to determine the incidence of abrasive and adhesive wear and determine typical locations for these wear patterns on polyethylene components.

See more about knee/shoulder implant bearing function.

The research lab at Dartmouth led by Professor Lee Lynd is active in research on the following topics:

• Microbial Cellulose Utilization, including fundamental and applied aspects
• Metabolic Engineering, focusing on thermophilic cellulolytic bacteria for fuel production
• Innovative Biomass Processing Technologies, including development, design, and evaluation
• Sustainable Bioenergy Futures, including resource, environment, and economic development

We approach these topics from a diversity of academic disciplines with molecular biology, microbiology, chemical/biochemical engineering providing the foundation for the first three. Consistent with the "Pasteur's Quadrant" model articulated by Donald Stokes (Brookings Institution Press, Washington, DC, 1997), we see advancing applied capability and increased fundamental understanding as having strong potential to be convergent and mutually-reinforcing, and we aspire to work in this mode.

A central theme of the Lynd group is processing cellulosic biomass in a single step without added enzymes. Such "consolidated bioprocessing" (CBP) is a potential breakthrough, and "is widely considered to be the ultimate low-cost configuration for cellulose hydrolysis and fermentation" (joint DOE/USDA Roadmap, 2007). We are focused on production of ethanol, a promising renewable fuel. The CBP strategy is however potentially applicable to a very broad range of fuels and chemicals.

Researching complex interactions between climate hazards and communities to inform governance of coastal risk. MORE

Microfabricated magnetic components using nanomaterials make it possible to miniaturize power-handling magnetic components through taking advantage of the materials' high-flux-density and high-frequency capabilities. We are developing practical methods of depositing these materials and fabricating inductors and transformers on silicon chips or in other technologies.

There is a need to create versatile tools for exploring the interplay between peripheral, tissue and immune factors in tissue remodeling. To create such a platform in the context of reproductive health, this project will develop methods for culturing organoids derived from patient material, with an initial focus on organoids derived from progenitor cells isolated from mammary tissue and human breast milk (hBM). These systems will be used to better understand lactation and cancer development, and ultimately, inspire methods for engineering "true to natural" formula production.

New devices research and developent for total joint arthroplasty includes:

• Testing of a new bi-material bearing for a total hip arthroplasty (THA) device against conventional bearing designs to compare levels of bearing surface damage and wear;
• Development of an intraoperative method for quantifying the orientation of prosthetic components used in total knee arthroplasty (TKA) that is efficient, easy to use, cost effective, and quick with respect to total surgical time.

See more about new devices.

New materials research and development for orthopaedic implants includes:

• Evaluation of equal channel angular extrusion (ECAE)-processed ultra-high molecular weight polyethylene (UHMWPE) for joint arthroplasty and industrial applications;
• Investigation of off-label use of a resorbable calcium sulfate antibiotic carrier in single stage and two-stage procedures to determine the potential of this use to change damage patterns or wear rates of artificial joints.

See more about new materials.

Implant failure analysis in the Dartmouth Biomedical Engineering Center for Orthopaedics is ongoing and plays a key role in identifying failure modes and relating them to various designs and materials being used in the industry. In fact, in 2000, NIH's Consensus Development Program produced a technology assessment statement acknowledging the value of implant retrieval programs:

• Implant retrieval and analysis is of critical importance in the process of improving care of patients in need of implants.
• Attention needs to be directed toward reducing various obstacles to implant retrieval and analysis, particularly legal and economic disincentives.
• The failure to appreciate the value of implant retrieval and analysis is a serious impediment to research in devices. A focused educational program will provide the information necessary for improving the quality of future devices.

See more about implant failure analysis.

Passive high-frequency power components are often the limiting factors in reducing the power loss, size, cost, and weight of high-frequency electronic power converters. Through detailed analysis, modeling, and optimization of high-frequency effects in inductors, transformers, and capacitors, we are improving performance of these components and making it easier to design the efficient, low-cost power electronics needed for a wide range of applications including energy efficiency and renewable energy.

Advancing the quantitative understanding of multisector dynamics. MORE

Maintaining immune tolerance during pregnancy is essential and evidence suggests that imbalances between T cell subsets can lead to fatal pregnancy events. This project, using existing and in-house generated single-cell datasets, will seek to better understand how peripheral immune cell features correlate with reproductive and uterine health, and in the long term, pregnancy outcomes or fertility.