PhD Thesis Proposal: Mikael Marois

Thursday, February 20, 2020, 8:00–10:00am

Borwell 658W - DHMC

"Improvements in Hyperspectral Imaging Methods to Increase Speed and Accuracy of Fluorescence Guided Neurosurgery"

Fluorescence guidance is an important tool that allows neurosurgeons to better differentiate between tumor and important functional tissue during surgical resection. To highlight tumors, the fluorescent tracer protoporphyrin IX (PpIX) is often used, as it preferentially aggregates in cancerous tissue. This fluorophore is able to absorb blue light and re-emit it in the pink region of the color spectrum.In order to quantify the amount of PpIX present in a given sample during surgery, the hyperspectral acquisition of multiple wavelengths is required. Most hyperspectral imaging methods are time consuming, as they rely on temporal scanning of wavelengths, and can therefore inhibit the flow of surgery.

This thesis aims to improve hyperspectral acquisition techniques by reducing recording time and increasing accuracy. An algorithm that reduces the number of wavelengths needed to produce accurate concentration estimates was developed. This technique allows the acquisition of hyperspectral stacks in half the time usually required, without compromising the quality of the measurement. A novel snapshot hyperspectral camera that uses birefringence in order to acquire 64 distinct channels in a fraction of a second was characterized and calibrated for use with PpIX imaging. This snapshot camera is compared to commonly used temporal scanning systems, and is shown to increase sensitivity in fluorescence detection because of its high optical efficiency, while reducing acquisition time by a factor of a hundred. Future steps to move this system towards real-time quantitative imaging of PpIX are discussed. Finally, stability of measurements taken using a probe based quantification system are studied, and alternatives to make this system more accurate are explored.

Thesis Committee

For more information, contact Daryl Laware at daryl.a.laware@dartmouth.edu.