Massachusetts: One of the biggest challenges with breast cancer is the delayed diagnosis. To help doctors identify at-risk patients early, researchers are taking the help of artificial intelligence.
MIT researchers have built a new deep learning model that can predict from a mammogram if a patient is likely to develop breast cancer in the future. As their official release explains, the system has been trained on mammograms and known outcomes from over 60,000 patients to learn the subtle patterns in breast tissue.
The AI was able to accurately access 31 percent of all cancer patients in its highest-risk compared to 18 percent by traditional models. Systems like these could help doctors in early diagnosis, much before the disease actually develops.