AI Finds Hidden Heart Risks in ECGs | Phoenix News

A groundbreaking AI tool developed at UC Berkeley can analyze ECGs to detect hidden patterns predicting sudden cardiac death — often missing by current medical standards. Trained on 440,000 Swedish ECGs and validated across U.S. and Taiwanese data, the model outperformed traditional risk assessments like LVEF, identifying a high-risk group previously overlooked. The AI zeroed in on a previously undocumented signal in the QRS complex, offering a new frontier in heart health monitoring. While not yet for home use, this could soon help doctors decide who needs life-saving defibrillators — and raise critical questions about data privacy in medical AI. Support the show:Get a discount at https://solipillow.com/discount/dnn. Advertise on DNN:[email protected] This is an automated, high-level news summary based on public reporting.Report issues to [email protected]. View sources & latest updates:https://sources.thednn.ai/42fefff570596d57