Heart failure & cardiomyopathies
Evaluation of diastolic function: machine learning improves classification of left ventricular filling pressure
Authors' conclusion
ML can improve classification of LVFP, particularly with a higher feasibility. The study unveiled less used parameters as some of the most valuable for evaluating LVFP.
Study context
- Population
- Patients undergoing echocardiography and heart catheterization
- Participants
- 250 patients
- Comparison
- Machine learning (ML) classification of left ventricular filling pressure (LVFP) vs the 2016 ASE/EACVI guidelines algorithm, against heart catheterization
- Outcome
- Accuracy of classification of LVFP as normal or elevated
Methodological detailsnested cross-validation, echocardiography simultaneously with, or within 24 h of, heart catheterization
- Nested cross-validation
- Echocardiography simultaneously with, or within 24 h of, heart catheterization
Performance
As reported in the abstract
ML models
82% to 86%
accuracy
2016 ASE/EACVI guidelines algorithm
81%
accuracy in the remaining patients
- Guidelines algorithm left unclassified due to missing values
- 13%
Further findings
Parameters selected by the ML models
- Average number of parameters selected
- 13
- Models including left atrial strain
- three
- Five highest ranked parameters: mitral E/left atrial reservoir strain, log(NT-proBNP), tricuspid regurgitation velocity, septal E/e’, and E/A
Original publication
Evaluation of diastolic function: machine learning improves classification of left ventricular filling pressure. Eur Heart J Cardiovasc Imaging. Published 28 January 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →