Automated Detection of Clinically Significant Mitral Regurgitation from Single-View B-mode Echocardiography Using Deep Learning
An AI model trained on nearly 30,000 heart ultrasounds can detect dangerous heart valve disease from basic images without requiring specialized Doppler ultrasound, potentially enabling broader screening.
MitralVision was trained on 28,487 apical four-chamber cine loops from 11,244 studies across 20 US states and externally validated on 629 studies from 26 independent sites, demonstrating AUROC 0.91 for detecting moderate/severe MR using only grayscale B-mode echo without Doppler—removing a major technical barrier to AI-assisted MR screening. Performance was comparable to inter-reader variability (expert agreement only 75.2%), suggesting the model could standardize assessments in high-throughput or resource-limited settings.
What the study was
- Study design
- external_validation_diagnostic_study
- Category
- ai_ml_clinical_diagnostics
- Maturity
- Validated
- Journal
- Journal of the American Society of Echocardiography
Why it surfaced
Doppler-free deep learning for MR detection from single-view echo with strong multi-site external validation is a clinically deployable approach that could democratize valvular disease screening. Score 7 (N=3, D=2, P=1, E=1) reflecting strong external validation (D=2) with population impact partially limited by existing diagnostic pathways.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.