AI may spot deadly heart risk in a routine ECG
UC Berkeley researchers developed an AI model that can identify a hidden pattern in routine ECGs associated with sudden cardiac death. The model was trained on over 440,000 Swedish ECGs linked to mortality data and validated on separate cohorts from the United States and Taiwan. Findings suggest the approach could help detect at‑risk individuals who appear normal on standard heart screenings.
- ▪The AI system was trained using more than 440,000 ECG recordings from Sweden paired with death certificates and health records.
- ▪When tested on independent patient data from the United States and Taiwan, the model maintained its predictive performance.
- ▪Sudden cardiac death can affect people with no known heart problems, including younger athletes, making early detection crucial.
- ▪Current routine ECG interpretation may miss the subtle waveform patterns that the AI model can uncover.
Opening excerpt (first ~120 words) tap to expand
Artificial Intelligence AI may spot deadly heart risk in a routine ECG UC Berkeley researchers say a new AI model found a hidden ECG signal that could help doctors identify more people at risk before sudden cardiac death By Kurt Knutsson, CyberGuy Report Fox News Published June 29, 2026 6:32am EDT Facebook Twitter Threads Flipboard Comments Print Email Add Fox News on Google close Video Kurt Knutsson analyzes social media addiction trial nearing verdict Kurt "CyberGuy" Knutsson discusses the social media addiction trial against Meta and YouTube, the jury’s questions on liability and damages, and what a possible verdict could mean on "Fox & Friends." NEWYou can now listen to Fox News articles! A routine heart test may be hiding a warning sign that doctors have missed for years.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Fox News.
Discussion
0 commentsMore from Fox News
