Development of AI Solution for intra-operative Optical Coherence Tomography based classification of cancerous lesions

Breast cancer was the most common cancer in 109 countries including Canada (excluding skin cancer) in 2020 (WHO). Early detection is critical in prognosis. The standard-of-care in early-stage breast cancer is breast conserving surgery (BCS), typically followed by radiotherapy and/or chemotherapy and/or endocrine therapy. The gold standard for margin assessment for BCS is histopathology. Results from histopathology analysis typically require 2-4 days, resulting in the need for a second surgery if a margin is positive. Despite decades of progress, reoperation due to tumor at or close to the margin is estimated to occur in over 20% of BCS patients. Our main goal is to improve this clear unmet need. Founded in 2013, Perimeter Medical Imaging AI (https://perimetermed.com/) is a Toronto-based medical device company with U.S. Headquarters in Dallas, Texas. Perimeter. Our Optical Coherence Tomography B-Series device embedded with our ImgAssist AI tool was granted Breakthrough Device Designation by the US FDA, designed to speed up availability of disruptive new technologies. We are currently conducting a randomized, multi-site, pivotal study to test our solution against the current standard of care and assess the impact on re-operation rates. The scope of this project involves 1) improving the current AI model to demonstrate improvements on performance metrics such as: accuracy, sensitivity, specificity, F1 score, AUC and 2) streamline the presentation of findings to the surgeons for quick decision making for next round of product release. Projects provide interns experience in AI development in healthcare with distinct deliverables to improve future product.

Faculty Supervisor:

Ervin Sejdi?

Student:

Partner:

Perimeter

Discipline:

Computer science

Sector:

Artificial Intelligence; Health and Related Sciences & Technology

University:

University of Toronto

Program: