AI Flags Women’s Heart Disease on 97,364 Mammograms With Up to 86% Accuracy
Updated
Updated · The Guardian · Aug 27
AI Flags Women’s Heart Disease on 97,364 Mammograms With Up to 86% Accuracy
3 articles · Updated · The Guardian · Aug 27
Summary
97,364 mammograms from 29,921 women let Israeli researchers train an AI system that identified stroke, high blood pressure and coronary heart disease from breast scans alone.
86% accuracy for prior stroke, 79% for high blood pressure and 78% for coronary heart disease suggests routine breast screening could also flag cardiovascular risk without extra imaging.
16% of the women had high blood pressure, while 2.5% had coronary heart disease and 2.5% had suffered a stroke; results held across ages and whether women also had cancer.
Munich conference presenters and outside cardiology experts called the findings compelling but said the model still needs better accuracy, fewer false results and clinical validation before rollout.
Hundreds of millions of women undergo mammograms each year, giving the approach broad potential as heart disease remains the leading and often underdiagnosed cause of death in women.