Heart failure & cardiomyopathies
A Machine Learning-Derived Score to Effectively Identify Heart Failure With Preserved Ejection Fraction
Main finding · Diagnostic accuracy for HFpEFResult — go to the resultAUC 0.88 vs 0.74Authors' conclusion
In a HFpEF cohort with obesity, BMI, estimated glomerular filtration rate, left ventricular mass index, and left atrial to left ventricular volume ratio most correlated with the identification of HFpEF, and a score based on these variables (HFpEF-JH score) outperformed the currently used H2PEF score.
Study context
- Population
- Patients with obesity with heart failure with preserved ejection fraction (HFpEF) (mean BMI 36.6 kg/m²) and controls404 HFpEF, 67 controls
- Exposure
- HFpEF-JH score vs H2FPEF score for diagnosis of HFpEF
- Outcome
- Diagnosis of HFpEF
Methodological detailsmachine learning (Gradient Boosting Machine), multivariate logistic model, internal validation
- Machine learning (Gradient Boosting Machine)
- Multivariate logistic model
- Internal validation
Main finding
Diagnostic accuracy for HFpEF
AUC
0.88 vs 0.74
P < 0.001
By group
HFpEF-JH score
- Sensitivity
- 0.83
- Specificity
- 0.82
- AUC
- 0.88
H2FPEF score
0.74
AUC
As reported in the abstract
Observational data: associations, not effects.
Further findings
Internal validation of the model (optimism-adjusted AUC)
- Optimism-adjusted AUC of the model
- 0.87
Original publication
A Machine Learning-Derived Score to Effectively Identify Heart Failure With Preserved Ejection Fraction. JACC Adv. Published 7 June 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →