For info on how to attend this videoconference, please email sidhartha.jandhyala.TH@dartmouth.edu.

"Multimodal ultrasound imaging for improved metastatic lymph node detection"

Abstract

Head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer worldwide and is complex in nature due to the variety of organs located in the head and neck region. Knowing the metastatic state of the lymph nodes is paramount in accurately staging and treating HNSCC patients. Currently, metastatic lymph node detection involves the use of magnetic resonance imaging and/or x-ray computed tomography, followed by biopsies for histological confirmation. The main diagnostic criteria is the size of the nodes; however, current imaging methods are not 100% accurate due natural lymph node variability. Ultrasound imaging is able to provide additional biological information in addition to lymph node size such as the hilus state, presence of necrosis and vascular information, but it is hindered by poor resolution and limited contrast. Augmenting ultrasound for metastatic lymph node detection has clinical potential due to the availability of ultrasound in the clinic, reduced radiation exposure and minimized patient morbidity.

This thesis focuses on augmenting ultrasound with photoacoustic imaging or with nanoparticle contrast agents for improved detection of lymph node metastasis. First, the development of an ultrasound-photoacoustic (USPA) imaging system is described. The USPA system is capa- ble of imaging blood oxygen saturation (sO2), a promising criteria to differentiate between metastatic and healthy lymph nodes. To correct for tissue-dependent attenuation of light in tissue, a deep neural network was developed and trained using Monte-Carlo simulated and experimentally acquired photoacoustic data for better sO2 predictions. Secondly, to improve ultrasound sensitivity to metastatic cells, molecularly targeted phase change perfluorohexane nanodroplets conjugated to epidermal growth factor receptor (EGFR) antibodies (PFHnD-Abs) were developed. It is shown that the PFHnD-Abs are able to specifically bind to HNSCC cells and improve the ultrasound contrast of the cells, opening the door to targeted metastatic lymph node detection. Lastly, to validate the use of the PFHnD-Abs in-vivo, a paired agent imaging approach was adopted by using a perfluoropentane core nanodroplet (PFPnD) as a non-targeted imaging agent to enable multiplex ultrasound imaging in vivo. Overall, this work expands the potential of ultrasound for metastatic lymph node detection.

Thesis Committee

• Geoffrey Luke, PhD (Chair)
• Kimberley Samkoe, PhD
• Jonathan Elliott, PhD
• Stanislav Emelianov, PhD

Contact

For more information, contact Theresa Fuller at theresa.d.fuller@dartmouth.edu.