Valvular & structural heart disease
Development and External Validation of an Echocardiography-Based AI Model for Predicting Aortic Stenosis Progression
Authors' conclusion
Models based on routinely available structured echocardiographic data predicted progression from mild or moderate to severe AS and showed consistent performance in an independent external cohort. These findings support the potential use of structured-data risk prediction to inform personalized surveillance strategies for patients with AS.
Study context
- Population
- Patients with mild or moderate aortic stenosis (AS) with structured echocardiographic report dataechocardiographic visits · derivation 9,247 visits · external validation 3,873 visits
- Exposure
- Three models using 21 routinely available echocardiographic and demographic variables to predict progression to severe AS at 1-, 3-, and 5-year horizons
- derivationn = 9,247 visits
- external validationn = 3,873 visits
Methodological details2 independent large tertiary centers, external validation
- 2 independent large tertiary centers
- External validation
Main finding
As reported in the abstract
1-year AUROC
0.90
95% CI 0.86–0.93
3-year AUROC
0.91
95% CI 0.88–0.93
5-year AUROC
0.91
95% CI 0.87–0.95
Observational data: associations, not effects.
Further findings
Discrimination for progression to severe AS (AUROC), external validation · 1-, 3-, and 5-year prediction horizons
- 1-year AUROC
- 0.91 (95% CI 0.90-0.93)
- 3-year AUROC
- 0.88 (95% CI 0.87-0.90)
- 5-year AUROC
- 0.88 (95% CI 0.87-0.90)
Original publication
Summarised from the published abstract; every number is checked against it