AI forecasts bleeding risk post-heart attack
Researchers at Indiana University have developed an artificial intelligence-based system that is interpretable (XAI) to identify patients at risk of internal bleeding in the heart muscle following a severe heart attack, using clinical information collected during cardiac catheterization. The study demonstrated that the model achieves an accuracy of over 84% in predicting bleeding risk before re-opening the artery, thereby enhancing physicians’ ability to make early treatment decisions and reduce the risks of death and heart failure. This system represents an important step toward improving early diagnosis and reducing reliance on later tests, and it can be adapted for other areas of healthcare.
Size: 120%Text Size
Notice: This Is an AI-Generated Summary
Comments (0)