Arrhythmias & electrophysiology
Clinical Applications, Methodology, and Scientific Reporting of Electrocardiogram Deep-Learning Models: A Systematic Review
Authors' conclusion
While ECG deep-learning models are increasingly clinically relevant, their reporting is highly variable, and few publications provide sufficient detail for methodologic reproduction or model validation by external groups.
Study context
- Population
- Clinically relevant electrocardiogram (ECG) deep-learning models published through July 1, 2022 (PubMed, EMBASE)44 manuscripts
- Participants
- 53 participants
- Outcome
- Clinical applications and reporting practices of ECG deep-learning models
Pooled result
As reported in the abstract
- Cardiomyopathy identification (most common application)
- 14/53 [26%]
- Arrhythmia detection
- 9/53 [17%]
- Publications with model architecture diagrams
- 33/44 (75%)
- Complete information to reproduce models
- 10/44 (23%)
- Saliency analysis performed
- 20/44 (46%)
- Models tested in external validation cohorts
- 18/53 (34%)
- Model code or resources available for external implementation
- 5/44 (11%)
Original publication
Clinical Applications, Methodology, and Scientific Reporting of Electrocardiogram Deep-Learning Models: A Systematic Review. JACC Adv. Published 8 November 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →