New AI tool matches human precision in assessing surgical margins for breast cancer

Editor's Note

Researchers from the University of North Carolina developed an artificial intelligence (AI) model that can predict whether cancerous tissue has been fully removed from the body during breast cancer surgical procedures. Their findings were published in the Annals of Surgical Oncology August 2023 issue. 

During the procedure, surgeons take a small amount of the surrounding healthy tissue, which is then photographed using a mammography machine and reviewed to make sure all abnormalities are removed. This specimen is then sent to pathology for further analysis. In this study, the researchers trained the AI model with hundreds of specimen mammogram images and matched these with the final specimen reports from pathologists. The researchers further developed the AI model by including demographic data from patients, such as age, race, tumor type, and tumor size.

The AI model's accuracy in predicting pathologic margins was found to perform as well as humans, if not better.

The researchers noted that the tool could be particularly useful for patients with higher breast density and for hospitals with fewer resources that may lack specialist surgeons, radiologists, or pathologists in-house.