Mechanical probe housing validation for minimally invasive EIT probe

Faculty: Ryan Halter

Mentor: Allie Doussan

This project will focus on developing an electrical impedance tomography probe for use in minimally invasive surgical procedures to determine the presence of cancer along the surgical margins. There is currently a concept of all of the components and 3D housing and we are seeking help to validate this design. This project would include 3D printing, post processing materials, and validating that the mechanical probe housing works as expected with incorporated components and with the da Vinci surgical robotic system. Opportunities to adjust the housing using Solidworks based on testing results is available but not a requirement. Other opportunities to engage in data collection/experiments exist based on interest and availability.

Tissue growth monitoring with EIT

Faculty: Ryan Halter

Mentor: Mimi Lan

Recommended background:

• Matlab or programming basics
• Cell culture experience preferred but not required

This project aims to non-invasively monitor the growth of cultured tissue in the lab using electrical impedance. Students interested in this project will possibly work on cell culture, phantom imaging experiments, MATLAB reconstructions, and experimental refinement. Some examples include exploring ways to improve cell adhesion to electrode surfaces, exploring phantom imaging with a novel transimpedance electrode array, and running experiments monitoring 2D or 3D cells for 7+ days in the cell incubator.

Construction of permanent phantoms for electrical impedance tomography

Faculty: Ryan Halter

Mentor: Ethan Murphy

Electrical impedance tomography produces images of the electrical properties of tissues under study from boundary data. An important part of validation experiments is producing phantoms that mimic the tissue of interest (eg. chest, breast, head and brain, or prostate). Our agar- and gelatin-based phantoms have a short life and their electrical properties can diffuse/change with time. We are interested in refining an epoxy and/or silicone-based recipe as a more permanent approach to phantom making. The work will involve 3D printing molds, constructing epoxy and silicone phantoms, measuring electrical properties, and adjusting ingredient proportions.

Human-centered inquiry-based engineering activities

Faculty: Petra Bonfert-Taylor and Vicki May

Thayer is creating a new human-centered pathway to the engineering sciences major as an alternative to the standard prerequisite-focused pathway. This pathway will include three new courses that teach foundational mathematical concepts in the context of engaging, hands-on, project-based activities with human-centered topics that resonate with students’ experiences, interests, and backgrounds. We are looking for first-year students to help develop lesson plans with activities that can be explained using physics and engineering concepts in conjunction with calculus and linear algebra-based mathematics.

Microstructural evolution of a high-entropy alloy

Faculty: Ian Baker

Mentor: Edwin Jiang

High entropy alloys (HEAs) are an exciting new type of material that, unlike traditional alloys, contain large amounts of multiple elements. In this project, the student will study the microstructural evolution of the HEA Fe28.2Ni18.8Mn32.9Al14.1Cr6 heated to temperature in the range of 650-750oC. The student will determine the relationship between temperature, heating time, and microstructure, and will operate a high-speed saw, polishing wheels, furnaces, and examine the specimens in a scanning electron microscope.

Recrystallization of cold-rolled polyethylene

Faculty: Ian Baker

Ultra-high molecular weight polyethylene (UHMWPE) consists of crystalline regions surrounded by amorphous regions. By cold-rolling the UHMWPE, the molecules in the amorphous regions can be aligned. The UHMWPE is then in a high-energy state, and upon heating, releases energy and can recrystallize. In this project, the student will cold roll UHMWPE and then run the material through a differential scanning calorimeter at different heating rates to determine the activation energy of the resulting recrystallization. X-ray diffraction will be performed before and after cold rolling to determine the degree of crystallinity. Time-permitting, the student will perform mechanical testing on the heat-treated UHMWPE.

Gastroenterology environmental impact assessment

Faculty: Emily Monroe and Solomon Diamond

Recommended background: An interest in biomedical, energy, and environmental engineering as well as applied math.

An estimated 17 million endoscopies are performed each year in the US alone, but little is known about the carbon emissions and other environmental impacts involved in performing these procedures. A team of Dartmouth-Health physicians would like to build a comprehensive model that captures the carbon emissions from cradle to grave in the endoscopy procedure. This effort will involve understanding the procedure itself as well as the logistics, material, and transportation that precedes and follows the actual endoscopy.

Data science for optimization of healthcare operations

Faculty: Vikrant Vaze

Recommended background: Some experience with Python coding

We use machine learning, simulation and optimization methods for improving patient access to primary and specialty care services at low cost through care coordination, improved scheduling, better capacity utilization, and optimized matching of patient needs with clinical resources. This project involves Python coding, regular meetings with healthcare practitioners, and use of anonymized actual patient datasets, and has high potential for impactful publications at the intersection of engineering, medicine, computer science, and business.

Artificial intelligence for fighting wildfires

Faculty: Vikrant Vaze

Mentor: Spencer Bertsch

Recommended background: Some experience with Python coding

With climate change creating warmer, drier conditions that lead to hotter and longer wildfire seasons, effectively allocating firefighting resources, such as trucks and planes, is getting harder. This becomes a dynamic resource allocation problem in which a constrained number of resources must be used to contain wildfires before they grow out of control. We aim to use distributed, deep multi-agent reinforcement learning methods to help firefighters make better tactical decisions. If you want to apply data science to fight climate change and learn some hands-on AI skills, then this project is for you.

Effect of oxygen on alcohol tolerance of E. coli to improve cellulosic biofuel production

Faculty: Daniel Olson

Mentor: Bishal Sharma

Alcohols such as ethanol are potential biofuels that could be produced from cellulose. Understanding how ethanol inhibits the growth and metabolism of microbes is important for developing engineering strategies to improve their ethanol tolerance. Our previous experiments have shown that some anaerobic cellulolytic organisms show high sensitivity to ethanol under anaerobic conditions. We are interested to see if this is true in E. coli as well, and if so, whether this sensitivity is affected by the presence of oxygen or other electron acceptors. Students will learn basic microbiology skills, prepare growth medium, perform batch fermentations, and measure metabolite concentrations by HPLC.

Characterizing ethanol production in clostridium thermocellum

Faculty: Daniel Olson

Mentor: Ty Lanahan

Cellulose is an abundant resource that could be used to produce biofuels and chemicals. My group is engineering native cellulose-fermenting organisms to produce ethanol. Clostridium thermocellum is one of the best cellulose-fermenting organisms in nature, however it only produces small amounts of biofuels such as ethanol. To improve ethanol production we are testing several new metabolic pathways. Students will learn basic microbiology and molecular biology skills while they assist in characterizing novel metabolic pathways that have been recently introduced into C. thermocellum.

Examination of biomaterials and biomechanics associated with total joint replacement

Faculty: Doug Van Cittters

Recommended background: Interest in biomedical engineering with particular interest in biomaterials and biomechanics

Over one million total joint replacements are performed in the US annually. For patients to regain activities of daily living (ADL), it is paramount that joint prostheses use materials that elicit appropriate responses in the body. Working in collaboration with the Dartmouth Biomedical Engineering Center (DBEC), orthopedic surgeons, and the medical device industry, Thayer’s Orthopedic Biomaterials Laboratory (OBL) examines the biomaterials and biomechanics of these devices, as well as novel assessment methods. Research may involve materials analysis using microscopy, spectroscopy, and mechanical testing, or examining the biomechanics of a certain joint or set of joints. Students will work closely with faculty, graduate students, research engineers, and orthopedic surgeons, and should be comfortable learning to use a variety of hands-on laboratory equipment.

Advanced manufacturing of stretchable bioelectronics for wearable healthcare

Faculty: Wei Ouyang

Mentor: Xianling Li

We aim to develop soft wearable patches for profiling biochemical markers in body fluids as a novel technology for at-home health monitoring. Toward this goal, we will leverage 3D printing, laser cutting, and microfabrication to manufacture various functional components of the system. Students will be exposed to topics in advanced manufacturing, biochemical assays, circuit design, embedded systems, and digital health.

Antibody engineering to improve HIV bNAb activity in vivo

Faculty: Margie Ackerman

Mentor: Natasha Kelkar

In 2022, 39 million people globally were living with HIV-1 infection. As of today, there is no successful HIV vaccine, but HIV antibodies are now in clinical trials. This project involves screening a panel of HIV-1 antibodies and their EG mutants for complement activity and other immune effector functions. The next step would be to screen combinations of the EG mutants, to study enhanced immune functions due to formation of heterohexamers. The project also involves transducing a human cell line in order to express HIV envelope protein on the surface. This cell line would be used as a tool for testing the different antibodies.

Game-theoretic reinforcement learning for cyber security

Faculty: Peter Chin

It is crucial for computer networks to be secure, trustworthy, and reliable, and to respond to or even predict ever-increasing cyber attacks. Advancements in deep learning, especially reinforcement learning (RL), offer hope to keep up with this challenge. However, the application of RL to cybersecurity has not achieved its full potential. Cyber-defense teams often fail to understand that their actions to defend the network are exactly what their adversaries are expecting. An adaptive and decentralized RL framework will be needed to combat the sophistication level of the cyber attacks of today. The student will work in the LISP (Learning, Intelli gence + Signal Processing) Lab and help develop adaptive, hierarchical game-theoretic training frameworks that can predict future cyber attacks.

Making federated learning more robust

Faculty: Peter Chin

Distributed learning paradigms such as federated learning often involve transmission of model updates, or gradients, over a network to avoid transmission of private data. However, it is possible for sensitive information about the training data to be revealed from such gradients. We have developed a method to discover the set of labels of training samples from only the gradient of the last layer and the id to label mapping. Our method is applicable to a wide variety of model architectures across multiple domains. The project will be to improve the effectiveness of our method for model training in various domains—image classification, automatic speech recognition, and more.

Deep routing in the real world

Faculty: Peter Chin

Routing through networks is a fundamental part of our lives, whether it’s the route we take to a new restaurant, or the data that’s routed to us through the internet. For both kinds of networks, all routing approaches rely on the ability to approximate the distance between a node in the network and the target node, which can be accomplished by embedding the network into some metric space. Road networks are inherently embedded onto the surface of the earth, but there is no natural embedding for artificially constructed networks like the internet. It was observed, however, that routing on internet topologies works well when they’re embedded into the hyperbolic plane. We want to explore how deep learning can help generate or adjust embeddings, to improve the performance of routing algorithms. To start, we will evaluate existing deep learned embeddings for routing on a variety of real-world networks, and use the resulting insights to adapt and improve the embedding method.

2023–2024 application deadline: October 24

1. Apply to either the FYREE program or the Women in Science Project (WISP), but not both.
2. Review list of available projects above and make a selection.
3. Submit an online application form for each project before the deadline listed above.
4. Not all applicants will be contacted for an interview, nor admitted. (We always have more applicants than we do projects. If you are not selected, we encourage you to reach out to faculty directly if interested in their research.)
5. Faculty mentors will interview applicants and select a student for each project.
6. Decisions will be made prior to the start of Winter term.