
Grand Challenge: Cancer Digital Intelligence (CDI): Responsible Artificial Intelligence (AI)
Princess Margaret Cancer Centre (PM) Grand Challenges support bold, innovative, and high-impact projects across the spectrum of cancer care. This year’s PM Grand Challenge, powered by Cancer Digital Intelligence, focuses on supporting projects that responsibly use artificial intelligence (AI) in cancer care. It also features an exciting partnership with the University of Waterloo, further strengthening our commitment to driving innovation in research and cancer care.
Congratulations to our
2024 – 2025 winning teams!
Dr. Scott Hopkins at the University of Waterloo, Dr. Arash Zarrine-Afsar at UHN, and their respective teams for their project, An AI-Driven Intraoperative Diagnostic Tool for Cancer Surgery

Dr. Scott Hopkins

Dr. Arash Zarrine-Afsar
Project Description:
Central Nervous System (CNS) cancer impacts thousands of Canadians each year, leading to poor health outcomes and low survival rates. While neurosurgery remains one of the primary treatment options, current diagnostic methods rely on post-operative analysis, which can delay decision-making and limit surgical precision.
As part of the Grand Challenge, the team is developing an innovative intraoperative diagnostic tool designed to provide rapid and accurate cancer classification. Using infrared laser mass spectrometry (PIRL-MS), the tool generates unique “fingerprints” of cancerous tissues, comparing them to known tumour profiles for faster, more precise identification. To further enhance accuracy, the team is integrating machine learning (ML) to improve classification, particularly for complex and rare data types.
This collaboration represents a major step forward in the use of AI for cancer diagnostics, potentially transforming surgical decision-making to improve patient outcomes.
Dr. Rajesh Bhayana and his team at the Princess Margaret Cancer Centre for their project, Artificial Intelligence for Risk Stratification of Liver and Ovarian Lesions (LI-RADS and O-RADS MRI) from Radiology Reports

Dr. Rajesh Bhayana
Project Description:
Assessing the probability of cancer detected in medical imaging is a crucial step in cancer diagnosis. However, radiology reports can be lengthy, and at times fail to clearly convey cancer risk levels. While tools such as the American College of Radiology (ACR) Reporting and Data Systems (RADS) help assess malignancy risk, they are often underutilized, reducing their effectiveness.
Dr. Bhayana’s team alongside CDI will work to integrate AI-driven automation into the interpretation of radiology reports. By leveraging Large Language Models (LLMs), the project aims to ensure that RADS is applied more consistently, leading to more accurate classification of liver and ovarian lesions, improved communication of cancer risk, and improved treatment decisions.
By implementing this AI-driven approach in supervised clinical settings, the team aims to enhance diagnostic accuracy and provide timely and accurate cancer management.
How to Apply
Our application process for the 2024-2025 Grand Challenge is now closed, thank you to everyone for applying.
Questions? Email pmcdi@uhn.ca.
Previous Grand Challenge winners:
2023-2024 Project Winners:
Computational & Bench Scientist Ecosystem (CoBE)
CoBE is a web portal recommendation engine allowing scientists to rapidly find or contribute self-contained, fully reproducible software tools & bioinformatic pipelines that address the most pressing analytical needs for computational biology at the Princess Margaret Cancer Centre and UHN.
Scalable Integrated Radiation Therapy Autosegmentation and Decision-Support for Individualized Cancer Care (SIRTADICC)
The SIR TADICC project will provide new tools to safely automate the way doctors identify the areas that need radiation treatment for head-and-neck cancers and assist doctors in making high quality treatment decisions. These segmentation tools and decision supports will allow for rapid adjustments to patient radiation plans according to the tumor response or other factors.
2022-2023 Project Winner:
Clinical Trial Integrated Matching System (CTIMS)
CTIMS intoduces an innovative approach to harnessing a patient’s ‘digital fingerprint’ to pinpoint trials they may be eligible for. CTIMS offers a new trial match process that significantly minimizes the time and resources required to identify patients eligible for clinical trials.
More Information
Direct questions about funding opportunity details and eligibility to alejandro.berlin@rmp.uhn.ca, Medical Director, CDI or benjamin.haibe-kains@uhnresearch.ca, Scientific Director, CDI.
Direct questions about the application process to pmcdi@uhn.ca