Researchers at UC Berkeley used artificial intelligence to identify warning signs associated with sudden cardiac death from routine ECGs. The model was trained on more than 440,000 ECGs and death certificates from Sweden and evaluated on thousands of patient records from the United States and Taiwan. The AI-identified high-risk group had an annual mortality rate of 7%, compared with 4.6% in the group identified through conventional methods. The results were published in Nature, and the team is exploring integrating the algorithm into hospital ECG databases to support patient monitoring and implantable defibrillator decisions.

Source:ScienceDaily · Read Original
Leave a Reply