Skip to Main Content
An official website of the United States government

AI Risk Prediction for Breast Cancer

Principal Investigator

Name
Anne Martel

Degrees
Ph.D.

Institution
Sunnybrook Research Institute

Position Title
Senior Scientist

Email
anne.martel@sunnybrook.ca

About this CDAS Project

Study
PLCO (Learn more about this study)

Project ID
PLCOI-1924

Initial CDAS Request Approval
Nov 13, 2025

Title
AI Risk Prediction for Breast Cancer

Summary
We have trained a series of risk prediction models on a private dataset taking approaches using both cell and tissue morphological features to predict risk, as well as more modern end-to-end methods including graph-based multiple instance learning and foundation models. We aim to validate these models independently as well as compare them to each other, classic pathology risk scores such as Nottingham grading, and more broad clinical variables like stage.

Aims

We would like to use the PLCO breast data as an independent validation method for our models. This would give us a more reliable measure of which methods are able to make reliable risk predictions for breast cancer patients.

Collaborators

None