Valvular & structural heart disease
Machine learning identifies pathophysiologically and prognostically informative phenotypes among patients with mitral regurgitation undergoing transcatheter edge-to-edge repair
Authors' conclusion
ML-enabled phenotyping captures the complexity of extra-mitral valve cardiac damage, which does not necessarily occur in a sequential fashion.
Study context
- Population
- Patients undergoing mitral valve TEER for mitral regurgitation (MR)609 derivation, 817 external validation
- Participants
- 1,426 patients
- Exposure
- Machine-learning (ML)-derived echocardiographic phenotypes
- Outcome
- Survival
- Follow-up
- 5 years
Methodological detailsmachine learning unsupervised agglomerative clustering, external validation (two institutions)
- Machine learning unsupervised agglomerative clustering
- External validation (two institutions)
Main finding
As reported in the abstract
5-year survival (derivation cohort)
Cluster 2 vs Cluster 1: P-value: 0.032 · Cluster 3 vs Cluster 1: P-value: <0.001 · Cluster 4 vs Cluster 1: P-value: <0.001
Observational data: associations, not effects.
Further findings
Prognostic significance of phenotypes in external validation cohort
- Prognostic significance of ML-derived phenotypes externally confirmed
Original publication
Machine learning identifies pathophysiologically and prognostically informative phenotypes among patients with mitral regurgitation undergoing transcatheter edge-to-edge repair. Eur Heart J Cardiovasc Imaging. Published 3 February 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →