AI-powered ECG diagnosis and prognosis for hypertrophic cardiomyopathy
Precision Diagnosis and Prognostic Prediction of Hypertrophic Cardiomyopathy Using Artificial Intelligence: A Multicenter Study
This project will test whether artificial intelligence reading standard 12-lead ECGs can identify different types of hypertrophic cardiomyopathy and tell them apart from look‑alike heart conditions in adults.
Quick facts
| Study type | Observational |
|---|---|
| Enrollment | 15000 (estimated) |
| Ages | 18 Years and up |
| Sex | All |
| Sponsor | Second Affiliated Hospital, School of Medicine, Zhejiang University Academic / other |
| Locations | 1 site (Hangzhou, Zhejiang) |
| Trial ID | NCT07263204 on ClinicalTrials.gov |
What this trial studies
Using a large multicentre historical cohort anchored on the inexpensive and widely available 12‑lead ECG, researchers will train deep‑learning models augmented with attention mechanisms to produce precise ECG‑based phenotyping of HCM, including septal, apical, and other morphologies. The team will build a discriminative model to separate HCM from phenocopies such as hypertensive heart disease and aortic stenosis, and an algorithmic framework designed for stability across devices and populations. Model governance will be embedded through version‑controlled releases, cloud‑edge deployment, and an "offline replay" evaluation loop to create an end‑to‑end evidence chain that mirrors clinical workflows. Participants include adults with guideline‑defined HCM, adults with LV hypertrophy as phenocopies, and healthy adult controls who can provide analyzable ECGs.
Who should consider this trial
Good fit: Ideal candidates are adults (≥18) with a guideline diagnosis of HCM, adults with LV wall thickness ≥13 mm as phenocopies, and healthy adult controls who can provide usable 12‑lead ECG recordings.
Not a fit: Patients who cannot provide analyzable ECG data, children under 18, or those with types of cardiac disease not represented in the training data may not receive benefit from this approach.
Why it matters
Potential benefit: If successful, this approach could enable earlier, cheaper, and more reproducible ECG‑based detection and subtype classification of HCM, helping patients get appropriate follow-up and care sooner.
How similar studies have performed: Previous AI‑ECG work has shown promising results for detecting HCM from ECGs, but large‑scale multicentre phenotyping and robust cross‑device deployment are still relatively novel.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: 1. Adults aged ≥ 18 years. 2. HCM cohort: Adults diagnosed with hypertrophic cardiomyopathy in accordance with the \*2023 Chinese Guidelines for the Diagnosis and Treatment of Hypertrophic Cardiomyopathy in Adults\*. 3. HCM phenocopy cohort: Adults with an LV wall thickness ≥ 13 mm at any site on echocardiography. 4. Healthy-control cohort: Adults with no history of cardiac disease and no evidence of myocardial hypertrophy on echocardiography. Exclusion Criteria: Patients from whom analyzable ECG data cannot be obtained.
Where this trial is running
Hangzhou, Zhejiang
- Second Affiliated Hospital, Zhejiang University School of Medicine — Hangzhou, Zhejiang, China (Recruiting)
Study contacts
- Study coordinator: Xiaojie Xie, MD, PhD
- Email: xiexj@zju.edu.cn
- Phone: (+86)0571-87784700
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.