AI predicts cardiac arrest risk better than doctors

An artificial intelligence model has proven significantly more accurate than doctors in predicting which patients are at risk of sudden cardiac death. This breakthrough could save lives and reduce unnecessary medical interventions. The AI, called MAARS, analyzes medical records and heart imaging, including previously underutilized MRI data, to identify hidden patterns of heart health. It achieved 89% accuracy overall, and 93% for those aged 40-60, outperforming current clinical guidelines. The AI model's success stems from its ability to detect heart scarring, a key indicator of risk, in patients with hypertrophic cardiomyopathy. Researchers plan to expand the model's use to other heart conditions.