Menu Close

MIT is Using AI to Predict Breast Cancer 5 Years Before it Starts with 99% Accuracy


MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and Massachusetts General Hospital (MGH) have co-developed a model to predict the likelihood that breast cancer or tumor will develop by scanning a mammogram.

The AI model was trained using over 90,000 mammogram scans, learned patterns, and precursors that led to malignant tumors. It also factors in age, family history of breast and ovarian cancer, hormonal and reproductive factors, and breast density.

The goal is to make early diagnosis more accurate and less expensive, especially for racial minorities. This deep-learning model is also being tested in other fields, including cardiovascular disease, pancreatic cancer, and other diseases.


MIT Professor Regina Barzilay, a breast cancer survivor as well as a lead researcher, has this to say:

“Rather than taking a one-size-fits-all approach, we can personalize screening around a woman’s risk of developing cancer,”

“For example, a doctor might recommend that one group of women get a mammogram every other year, while another higher-risk group might get supplemental MRI screening.”

Lead author Adam Yala, a CSAIL PhD student:

“Our goal is to make these advancements a part of the standard of care,”

“By predicting who will develop cancer in the future, we can hopefully save lives and catch cancer before symptoms ever arise.”


For the full breakdown, head on over to MIT’s site here.