Clinical implementation of an AI-enabled ECG for hypertrophic cardiomyopathy detection

Area of Science:

• Cardiology
• Medical Informatics
• Artificial Intelligence in Medicine

Background:

• Hypertrophic cardiomyopathy (HCM) is frequently underdiagnosed.
• Artificial intelligence (AI) offers a potential solution for early HCM detection using ECGs.
• Clinical implementation of AI for HCM suspicion has not been previously studied.

Purpose of the Study:

• To assess the clinical implementation of an AI-ECG software for identifying patients with suspected HCM.
• To evaluate the effectiveness of AI in alerting clinicians to potential HCM cases.

Main Methods:

• A prospective, open-label, multicentre cohort study was conducted.
• Viz HCM, an AI-ECG software, was implemented across five healthcare systems from January to December 2023.
• The study focused on patients over 18 without a prior HCM diagnosis, analyzing alert viewing rates and follow-up actions.

Main Results:

• Out of 145,848 screened ECGs, 3% (4,348) triggered an HCM suspicion alert.
• Users viewed 69% of alerted cases, with 217 patients enrolled for follow-up.
• A total of 17 new HCM diagnoses (7.8%) were confirmed among enrolled patients.

Conclusions:

• AI-based ECG analysis can be successfully integrated into clinical workflows for identifying new HCM patients.
• Further research is needed to evaluate the scalability and compare this AI approach to the standard of care.