A machine learning diagnostic signature using multi-modality electronic health record data detected HFpEF with an AUROC of 90% (P<0.001) and average precision of 74% in a validation cohort.
Observational (n=1,854)
No
Does a machine learning-based diagnostic signature using multi-modality EHR data accurately predict the likelihood of HFpEF in patients with unexplained dyspnea compared to existing clinical scores?
An automated machine learning pipeline using structured and unstructured EHR data can accurately identify patients with HFpEF among those presenting with unexplained dyspnea, outperforming the existing H2FPEF score.
Estimación del efecto: AUROC 0.90 (95% CI 0.898-0.902)
valor p: p=<0.001
ABSTRACT Aims Heart failure with preserved ejection fraction (HFpEF) is thought to be highly prevalent yet remains underdiagnosed. We sought to develop a data-driven diagnostic model to predict from electronic health records (EHR) the likelihood of HFpEF among patients with unexplained dyspnea and preserved left ventricular EF. Methods control cohort II). The ability of clinical parameters and investigations to discriminate cases from controls was evaluated by extreme gradient boosting. A likelihood scoring system was developed and validated in a separate test cohort. The derivation cohort included 1585 consecutive patients: 133 cases of HFpEF (9%), 194 non-HF cases (Control cohort I) and 1258 HFrEF cases (Control cohort II). Two HFpEF diagnostic signatures were derived, comprising symptoms, diagnoses and investigation results. A final prediction model was generated based on the averaged likelihood scores from these two models. In a validation cohort consisting of 269 consecutive patients (with 66 HFpEF cases (24.5%)), the diagnostic power of detecting HFpEF had an AUROC of 90% (P<0.001) and average precision (AP) of 74%. Conclusion This diagnostic signature enables discrimination of HFpEF from non-cardiac dyspnea or HFrEF from EHR and can assist in the diagnostic evaluation in patients with unexplained dyspnea.
Farajidavar et al. (Fri,) conducted a observational in Heart Failure with Preserved Ejection Fraction (HFpEF) (n=1,854). Machine learning diagnostic signature vs. Expert clinical consensus and H2FPEF score was evaluated on Diagnostic power of detecting HFpEF (AUROC) (AUROC 0.90, 95% CI 0.898-0.902, p=<0.001). A machine learning diagnostic signature using multi-modality electronic health record data detected HFpEF with an AUROC of 90% (P<0.001) and average precision of 74% in a validation cohort.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: