A new artificial intelligence model outperformed the most common methods of detecting heart attacks by using electrocardiogram readings to more efficiently spot heart attack indicators, according to a study performed by researchers at the University of Pittsburgh.
The model was able to quickly and accurately analyze ECG readings, saving up to 24 hours that additional tests may take when doctors cannot verify a patient is experiencing a heart attack, according to a June 29 news release from UPMC.
The model was developed from ECGs from 4,026 patients with chest pain at three hospitals in Pittsburgh and validated with 3,287 patients from another hospital system, according to the release.
The study observed 7,313 patients from multiple clinical sites and revealed that the AI model outperformed clinical interpretation of ECG, commercial ECG algorithms and the HEART score, a scoring system that assigns points toward a patient's risk factors when experiencing chest pain, according to the study published June 29 in Nature Medicine.
The researchers conducting the study wanted to determine whether the AI model would achieve a level of accuracy similar to HEART, but it surpassed their expectations by exceeding this score, according to Salah Al-Zaiti, PhD, RN, associate professor in the Pittsburgh School of Nursing and of emergency medicine and cardiology in the School of Medicine.