Cancer tissue classification using machine learning/deep learning algorithms

Raman spectroscopy is a non-destructive laser-based optical technique that provides information on the molecular composition of biological tissue. Using this technique, it is possible to distinguish different types of tissue and use this information to develop prognostic tests to evaluate, for example, how a patient will respond to a specific therapy. Combined with other spectroscopy techniques, it is possible to acquires a wide range of information about the cancer/normal tissues in different organs and use this database to perform a very efficient tissue classification using the machine/deep learning algorithms. This research will lead to the development of new prognostic system, using spectroscopy techniques to predict and categorize tissues to for different types of cancers. These developments will setup the stage for future clinical trials using spectroscopy to improve treatment for patients affected with cancer, for example, the potential efficacy of immunotherapies.

Faculty Supervisor:

Frédéric Leblond

Student:

Ehsan Edjlali

Partner:

Institut national d'optique

Discipline:

Visual arts

Sector:

Medical devices

University:

École Polytechnique de Montréal