Exploration of machine/deep learning techniques using Raman spectroscopy for the identification and classification of brain tumors and prostate cancer

Cancer cells are nearly impossible to distinguish from normal tissue without sophisticated detection methods and conventional methods fail to provide real-time information during the surgical procedure. ODS Medical is developing an exclusive Raman spectroscopy system that collects data through an optical probe and is interfaced with sophisticated machine learning algorithm to identify tissues abnormalities in real-time. This technology has proven to be effective for brain tumor and prostate cancer detection. As a result much larger datasets have been acquired both for brain and prostate. Those datasets will be analyzed in order to improve the current model and develop new models based on convolutional neural networks. A global objective of this research project is to build robust mathematical models on which ODS Medical can rely on.

Frédérick Dallaire
Faculty Supervisor: 
Samuel Kadoury
Project Year: