AI-powered ECG analysis could boost care for patients with hypertrophic cardiomyopathy
An advanced artificial intelligence (AI) algorithm examining electrocardiograms (ECG) can help clinicians diagnose and treat hypertrophic cardiomyopathy (HCM), according to a new study published in the Journal of the American College of Cardiology.[1]
“Although hypertrophic cardiomyopathy causes significant morbidity and is a leading cause of sudden death in adolescents, initial detection remains difficult,” wrote lead author Geoffrey H. Tison, MD, a cardiologist with the University of California San Francisco (UCSF), and colleagues. “Although echocardiography is an important diagnostic modality for HCM, the electrocardiogram is more widely accessible.”
Tison et al. focused on one AI-based ECG analysis algorithm (AI-ECG) developed at UCSF and another AI-ECG developed by specialists with Mayo Clinic. Each algorithm was independently developed, trained and validated. When it came to identifying signs of HCM in a 12-lead ECG, the UCSF algorithm had an area under the receiver operating characteristic (ROC) curve curve of 0.938, sensitivity of 84.6% and specificity of 96.3%. The Mayo Clinic algorithm had an area under the ROC curve of 0.979, sensitivity of 92.3% and specificity of 94.1%.
The group applied these AI-ECGs to data from the PIONEER-OLE trial, focusing on 12-lead ECGs from more than 200 HCM patients treated through January 2020. The mean patient age was nearly 58 years old, and 69.2% were men. The median follow-up period was 79 weeks.
Overall, the authors found, both algorithms were able to capture key data related to “obstructive HCM physiology and pathophysiology” that is not identified during a traditional ECG interpretation.
“AI-ECGs might hold promise as a potential tool for monitoring disease status, cardiac hemodynamics, and drug therapeutic response,” the authors wrote.
The team did note that their study had a small sample size. However, they also emphasized their work’s strengths, including its long-term follow-up.
“Future studies can evaluate this approach as a guide for drug titration to enhance safety,” they concluded.
Related Cardiomyopathy Content:
U.S. expected to see large rise in cardiomyopathy cases over next decade
FDA approves mavacamten for obstructive hypertrophic cardiomyopathy
Metoprolol demonstrates value as a treatment for obstructive hypertrophic cardiomyopathy
Diabetes with cardiomyopathy associated with heightened heart failure risk
‘Surprising’ trends in ICD use among HCM patients
Combined genetic testing for cardiomyopathies and arrhythmias yields positive results
COVID-19 leads to sharp rise in stress cardiomyopathy
LAA occlusion reduces stroke risk among patients with HCM and AFib
MRI detects heart failure risk in patients with dilated cardiomyopathy
Reference:
Michael has more than 18 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.