Heart failure & cardiomyopathiesMeta-analysis
Artificial Intelligence-Enabled Electrocardiograms and Pregnancy-Related Cardiomyopathy
- Univariate and Reitsma bivariate random effects model meta-analysis
- systematic search of PubMed, Scopus, and Cochrane
Authors' conclusion
AI-ECG models demonstrate good performance for PRCM detection and may represent a promising adjunct for earlier identification of PRCM. Additional studies are needed to evaluate their impact on clinical outcomes in pragmatic settings.
Study context
- Population
- Studies investigating the use of AI-ECG models for pregnancy-related cardiomyopathy (PRCM) detection, reporting complete 2×2 confusion matrix datapatients in 6 included studies (716 identified) · PubMed 84 · Scopus 544 · Cochrane 88
- Participants
- 2,928 participants
- Comparison
- Pooled diagnostic performance of AI-ECG models for PRCM detection
Pooled result
As reported in the abstract
Pooled diagnostic performance (univariate random effects model)
| Group | sensitivity | specificity | diagnostic OR |
|---|---|---|---|
| AI-ECG models | 0.808 (0.656-0.902) | 0.944 (0.889-0.973) | 54.017 (37.918-76.952) |
Further pooled results
| Outcome | Values |
|---|---|
| Pooled diagnostic performance for PRCM detection (Reitsma bivariate random effects model) | AI-ECG models sensitivity 0.775 (0.623-0.878) · specificity 0.936 (0.871-0.969) · area under the curve 0.932 (0.813-0.953) |
Parts of this Brief are withheld: the automatic checks could not confirm them against the abstract.
Original publication
Artificial Intelligence-Enabled Electrocardiograms and Pregnancy-Related Cardiomyopathy. JACC Adv. Published 24 August 2026. doi:10.1016/j.jacadv.2026.103180
Summarised from the published abstract; every number is checked against it