Heart failure & cardiomyopathies
Automated Assessment of Right Atrial Pressure From Ultrasound Videos Using Machine Learning
Authors' conclusion
Machine learning is capable of accurately and robustly interpreting RAP from echocardiogram videos.
Study context
- Population
- Echocardiogram studies with IVC ultrasound videos; subset with right heart catheterization15,828 IVC videos
- Comparison
- Deep learning model vs cardiologist estimates of right atrial pressure (RAP) from IVC ultrasound
- Outcome
- Agreement of model RAP estimates with cardiologist-assessed RAP
Methodological detailsexternal validation
- External validation
Performance
As reported in the abstract
Agreement with cardiologist RAP estimates (test data set)
| Test or model | Value | AUC |
|---|---|---|
| Model | agreement 80.3% | 0.844 |
Model
Valueagreement 80.3%
AUC0.844
Further findings
| Outcome | Model |
|---|---|
| External data set, different institutionCompared to cardiologist RAP estimates | |
| AUC | 0.854 |
- RAP estimates vs right heart catheterization (subset)
- P = 0.98
- RAP estimates vs right heart catheterization (subset): statistically indistinguishable vs catheterization gold standard
- RAP estimates vs right heart catheterization (subset): 319 RAP measurements from right heart catheterization
Original publication
Automated Assessment of Right Atrial Pressure From Ultrasound Videos Using Machine Learning. JACC Adv. Published 28 August 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →