AI discovers a hidden heart signal that could predict sudden death
Scientists have uncovered a hidden warning sign in routine heart scans that could predict sudden cardiac death before it strikes. The AI-powered discovery outperformed conventional screening methods, potentially revealing thousands of people at risk of a fatal heart malfunction that doctors might otherwise miss.
Over 300,000 U.S. deaths annually stem from sudden cardiac arrest, an emergency stemming from the heart's electrical system abruptly failing. Implanted defibrillators can avert some of these fatalities, yet doctors have encountered difficulties pinpointing which patients require them most. UC Berkeley researchers leveraged artificial intelligence (AI) trained on data from over 440,000 Swedish electrocardiograms (EKGs) to discover a previously unnoticed warning signal in routine heart scans.
By analyzing electrical patterns alongside death certificate information, the AI system distinguished between healthy individuals, those with known heart risks, and people who perished from sudden cardiac causes. Subsequent analysis of U.S. and Taiwanese patient records confirmed the AI's reliability in identifying heightened-risk patients.
The AI outperformed standard screening methods that focus on heart pumping efficiency at predicting annual sudden cardiac death rates. Moreover, the AI detected a larger group of people at elevated risk, including those typically deemed low-risk by current guidelines. The technology's reliance on existing EKG images could enhance screening capabilities without necessitating new tests.
The findings, published in Nature, could aid physicians in making better decisions regarding implanted defibrillators and shed light on the mechanisms behind the AI-recognized electrical patterns.
Rewritten by urgent.news; may contain errors — check the original.
Source: sciencedaily.com
This story
This is one outlet's version. Read the fullest account.
- AI discovers a hidden heart signal that could predict sudden death sciencedaily.com