Heart failure & cardiomyopathies
Deep Learning-Based Prediction of Right Ventricular Ejection Fraction Using 2D Echocardiograms
Authors' conclusion
Using 2D echocardiographic videos alone, the proposed DL-based tool can accurately assess RV function, with similar diagnostic and prognostic power as 3D imaging.
Study context
- Population
- Patients with right ventricular ejection fraction (RVEF) measured by 3D echocardiographyexternal set 365
- Participants
- 831 patients
- Comparison
- Deep learning (DL)-based RVEF prediction from 2D echocardiographic videos vs 3D echocardiography-derived RVEF and expert visual assessment
- Outcome
- Accuracy of DL-predicted RVEF vs 3D echocardiography-derived RVEF
Methodological detailsretrospective, internal and external validation
- Retrospective
- Internal and external validation
Performance
As reported in the abstract
RVEF prediction, mean absolute error
| Test or model | Mean absolute error, internal validation | Mean absolute error, external validation |
|---|---|---|
| Ensemble model | 4.57 percentage points | 5.54 percentage points |
Ensemble model
Mean absolute error, internal validation4.57 percentage points
Mean absolute error, external validation5.54 percentage points
Further findings
MACE, by DL-predicted RVEF (adjusted for age, sex, LV systolic function) · median follow-up 1.9 years
| HR (95% CI) | P value | ||
|---|---|---|---|
| DL-predicted RVEF | 0.924 (0.862–0.990) | P = 0.025 | |
| 0.512Hazard ratio (log scale) |
| Outcome | Expert reader | P value |
|---|---|---|
| RV dysfunction (RVEF <45%), external validation | ||
| Accuracy | 77.0% | P = 0.678 |
- Right ventricular (RV) dysfunction (RVEF <45%), external validation: DL ensemble model Accuracy 78.4%
Original publication
Deep Learning-Based Prediction of Right Ventricular Ejection Fraction Using 2D Echocardiograms. JACC Cardiovasc Imaging. Published 10 May 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →