LIVE
OpenAI Partners with Chilean Ministry to Advance Public Sector AI Ambience Healthcare Expands Clinical AI Suite for Hospital Operations AI Models Vulnerable to Manipulation in Simulated Warfare Scenarios, Research Shows AWS Highlights Enterprise Governance and AI Workspace Quick Ahead of re:Invent 2026 LoanPro Integrates Amazon Bedrock to Automate Loan Servicing Workflows Google Brings Wallet App to iOS Users Across the United States Microsoft Reveals Cause of Global Outlook Outage AWS Outlines Phased Roadmap for Responsible Agentic AI Governance Google updates Gemini settings to separate chat history from AI model training Uber Eats and Zipline launch autonomous drone food delivery in Houston OpenAI Partners with Chilean Ministry to Advance Public Sector AI Ambience Healthcare Expands Clinical AI Suite for Hospital Operations AI Models Vulnerable to Manipulation in Simulated Warfare Scenarios, Research Shows AWS Highlights Enterprise Governance and AI Workspace Quick Ahead of re:Invent 2026 LoanPro Integrates Amazon Bedrock to Automate Loan Servicing Workflows Google Brings Wallet App to iOS Users Across the United States Microsoft Reveals Cause of Global Outlook Outage AWS Outlines Phased Roadmap for Responsible Agentic AI Governance Google updates Gemini settings to separate chat history from AI model training Uber Eats and Zipline launch autonomous drone food delivery in Houston

UC Berkeley Researchers Train AI on EKGs to Spot Hidden Sudden Cardiac Risks

🤖 AI Models 2 min read 👁 3

Scientists at UC Berkeley have developed an artificial intelligence model using hundreds of thousands of electrocardiograms to detect previously hidden warning signals for sudden cardiac arrest.

UC Berkeley Researchers Train AI on EKGs to Spot Hidden Sudden Cardiac Risks🤖AI Models

Photo: ScienceDaily AI

Key points

  • AI model trained on over 440,000 EKGs from Sweden to detect hidden cardiac risks.
  • Identified a patient group with an annual sudden cardiac death rate of 7%, outperforming traditional screening.
  • Works on routine EKG images already collected globally in medical centers.

Researchers at the University of California, Berkeley, have successfully trained an artificial intelligence system to uncover a previously overlooked warning signal in routine heart scans. The technology is designed to identify patients at high risk of sudden cardiac arrest before a fatal event occurs. Annually, more than 300,000 individuals in the United States die from sudden cardiac arrest, a medical emergency caused by a sudden failure in the heart's electrical system rather than a blocked blood vessel.

Led by associate professor Ziad Obermeyer, the team trained the algorithm on more than 440,000 electrocardiograms gathered in Sweden. By cross-referencing these electrical patterns with death certificate data, the AI learned to differentiate between healthy individuals, those with known heart conditions, and patients who ultimately experienced fatal cardiac events. The researchers subsequently tested the system using thousands of patient records from the United States and Taiwan to ensure its accuracy across diverse populations.

Traditional screening methods primarily measure how much blood the heart pumps with each contraction to determine who might need an implanted defibrillator. While these standard techniques isolate a patient group with an annual sudden cardiac death rate of 4.6%, the AI model flagged a cohort with a 7% annual death rate. Furthermore, the algorithm identified thousands of individuals at elevated risk who would have otherwise been classified as low risk under existing medical guidelines.

Because the algorithm operates on standard EKG images already gathered routinely in healthcare settings, the innovation could improve patient screening without necessitating new medical procedures. The findings were published in the journal Nature, offering physicians a potential new tool to determine device placement and better understand pre-arrest cardiac activity.

What to watch next: Further clinical validation and investigation into the specific internal heart changes that generate the electrical patterns recognized by the algorithm.

Source ScienceDaily AI — sciencedaily.com/releases/2026/11/261112230456.htm ↗

Contact the newsroom

Got a tip, a proposal or a partnership question? Write to us and we will reply shortly.