Winning Teams
1st Place
Universal Antigen Encoding of T-Cell Activation from High-dimensional Single-Cell Dynamics
Liamo Pennimpede, Abe Arafat, Ryan Reszetnik, and Romen Poirier (all from Electrical, Computer, and Software Engineering)
We have built an application to help cancer researchers speed up the drug development process. The application leverages a fully customised, unsupervised, machine learning model. Our supervisors are at Mila in Montreal and the National Cancer Institute in Washington, D.C.聽The use of our solution allows cancer researchers to test a high number of neoantigens at a very low cost, and very quickly. Then, only top candidates move on to be validated using the more costly mouse-model, before finally moving on to human trials.
2nd Place
Multi-Camera Surgical Tool Localization Setup for Image-Guided Neurosurgery
George Sideris, Justin Cree, Andrew Stirling, and Mamadou Ly (all from Mechanical Engineering)
Our project included the design (both mechanical and software) and fabrication of a multi-camera surgical tool localization setup for image-guided neurosurgery. Image-guided neurosurgery entails the modelling of critical structures in the brain, such as tumors and blood vessels, and the tracking of surgical tools in relation to these models. This tracking enables the surgeons to visualize the position of their tools in relation to the critical structures in real-time, allowing them to plan tumor removal approaches before performing a craniotomy (analogous to 'x-ray' vision). Our solution has the potential to greatly improve the user experience of neurosurgeons by limiting any obstructions to their view or mobility in comparison to state-of-the-art tool localization systems. Further, its greatly reduced cost will enable image-guided surgery technologies to become accessible to many hospitals within Canada and around the world which lack the funding for currently available systems.
3rd Place
Multiplexed Colorimetric Biosensor on Contact-Mode CMOS Image Sensor
Laura Camila Penuela Cardenas, Nassib Jr. Hassouna, Young chae Han, Lan Anh Huynh, and Mary Wan (all聽from Bioengineering)
We have designed and built a colorimetric biosensor on a contact-mode CMOS image sensor for the monitoring of chronic kidney disease (CKD) via the multiplex detection of sweat CKD biomarkers (e.g., uric acid, creatinine, chloride).聽Our proposed solution offers end users the potential to monitor their own progression of CKD from the comfort of their home.
