UC Berkeley (Calif.) researchers have developed an AI model that identifies sudden cardiac death risk from standard electrocardiograms more accurately than current clinical tests, according to a study published in Nature.
The team trained the algorithm on more than 440,000 EKGs from Sweden, pairing the scans with death certificate data to teach the model to recognize waveform patterns associated with sudden cardiac death. Researchers then validated the tool against de-identified patient data from a San Diego hospital system and a hospital in Taiwan.
The current standard identified a high-risk population with a 4.6% annual rate of sudden cardiac death, while the AI model identified a high-risk group with a 7% annual rate. This means thousands of people who are high-risk appeared low-risk under the current screening criteria. The model can better identify those patients using widely available EKGs, potentially expanding the pool of patients who qualify for an implantable defibrillator.
“We can not only make better decisions, but also start to understand what’s actually going on with these patients before their heart stops,” lead author Ziad Obermeyer, MD, an associate professor at UC Berkeley’s School of Public Health, said in a June 24 university news release.
More than 300,000 people in the U.S. die from sudden cardiac arrest annually, and roughly 90% of out-of-hospital cases are fatal.
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.
