Home / Science / AI Spots Hidden Heart Risk on ECGs
AI Spots Hidden Heart Risk on ECGs
29 Jun
Summary
- AI model trained on ECGs to predict sudden cardiac death.
- AI identified a higher-risk group than standard LVEF method.
- A visible ECG feature in the 'aVL' view signals high risk.
A groundbreaking AI model developed by UC Berkeley researchers can now detect hidden signals in electrocardiograms (ECGs) that may indicate a heightened risk for sudden cardiac death. Trained on over 440,000 ECGs, the AI identified a previously unrecognized high-risk group, showing a 7% annual rate of sudden cardiac death, significantly higher than the 4.6% rate associated with the standard LVEF screening method.
The AI's detection focuses on a specific, previously undescribed feature within the 'aVL' view of the ECG's QRS complex. This discovery offers a new avenue for earlier intervention, potentially preventing invasive procedures by identifying individuals who might benefit most from closer monitoring or other interventions before a cardiac event occurs.
This AI tool, tested successfully on data from both the U.S. and Taiwan, is currently undergoing further testing in health systems across Sweden, Taiwan, and the U.S. While not yet available for home use, this innovation promises to enhance routine cardiac screening, offering a vital second chance by spotting danger signs that current methods might overlook. Researchers emphasize the importance of patient privacy and data security as these advanced medical AI tools are developed and deployed.
