Enhanced AI Decision Support Module

AI Decision Support Module: An enhanced AI Decision Support Module aimed at streamlining the surgical review process through the identification of suspicious excision margins within medical imaging. Building upon an existing AI framework, which utilizes labeled data for the identification of potentially problematic regions, this project seeks to improve the model’s accuracy and reliability. The key responsibilities involve increasing the training dataset via data augmentation techniques and the development of a new AI model. The expected outcome is a model that demonstrates superior evaluation metrics, including F1 score and AUPRC, on blinded test datasets, thereby facilitating a more efficient and accurate surgical planning and review process.

Faculty Supervisor:

Paul Yoo

Student:

Partner:

Perimeter

Discipline:

Engineering

Sector:

Biotechnology

University:

University of Toronto

Program: