Heart failure & cardiomyopathiesMeta-analysis
Artificial Intelligence–Enhanced Electrocardiogram Models for Detection of Left Ventricular Dysfunction
- systematic review and external validation (independent validation and head-to-head comparison study)
- risk of bias assessment
- well-phenotyped registry of patients undergoing routine clinical cardiac magnetic resonance imaging with cardiologist-adjudicated reports and paired ECGs
- all consecutive patients
Authors' conclusion
In this first-in-kind independent validation and head-to-head comparison study, AI-ECG for LVSD detection demonstrated strong performance despite training on disparate populations. However, the limited-availability of models hinders independent validation.
Study context
- Population
- Patients undergoing routine clinical cardiac magnetic resonance imaging with paired ECGsall patients in external validation cohort, mean age 59 ± 15 years, 450 [35%] female
- Participants
- 1,203 patients
- Comparison
- Published AI-ECG models for left ventricular systolic dysfunction (LVSD) detection compared head-to-head against cardiologist-adjudicated cardiac magnetic resonance imaging reports
- All patientsn = 1,203
Pooled result
As reported in the abstract
Systematic review: published AI-ECG models predicting LVSD
Studies identified
35
Models described
51
Groups sharing models for independent testing (from Korea, the United States, Taiwan, and the Netherlands)
Four
Further pooled results
| Outcome | All patients | Lower-complexity subgroup |
|---|---|---|
| AUROC for LVSD detection | AUROC range 0.83 to 0.93 | AUROC range 0.87 to 0.96 |
Performance across subgroups
- Performance remained consistent across subgroups, with slight decreases in ECGs showing wide QRS complexes or atrial fibrillation
Original publication
Artificial Intelligence–Enhanced Electrocardiogram Models for Detection of Left Ventricular Dysfunction. JACC Adv. Published 20 January 2026. doi:10.1016/j.jacadv.2025.102572
Summarised from the published abstract; every number is checked against it