AI Mammography
Overview
This project focuses on transforming breast cancer screening by automating the interpretation of mammograms using AI, addressing the global shortage of specialized breast imaging radiologists.
The Issue
Breast cancer screening with mammography has proven effective in reducing mortality, yet screening programs worldwide remain inefficient. In the U.S. alone, over $7 billion is spent annually on mammograms, often interpreted by radiologists with varying diagnostic accuracy. With a shortage of specialized breast imaging radiologists, sensitivity ranges from 66.7% to 98.6% and specificity from 71.2% to 96.9%. This discrepancy highlights the need for a more effective solution.
Approach
Our solution involves using AI to selectively automate mammography interpretation. By developing AI tools that identify a significant fraction of mammograms as cancer-free with near-perfect accuracy, we aim to fully automate image interpretation for these cases, freeing radiologists to focus on more complex cases.
MTM Student Engagement
Our MTM team will be integral in moving this project toward commercial translation. Responsibilities include clinical needs assessment, competitive market analysis, regulatory strategy, IP planning, prototyping software deployment, and developing a business model. Students will work alongside experts in AI and radiology, contributing to the design and deployment of cutting-edge AI tools in the clinical setting.
Project Brief
