Figure 4.
Decision Curve Analysis for Patient Management Decisions Incorporating the AI HFpEF Model
Decision curve analysis comparing net benefit of patient management decisions are based on the output of HFA-PEFF score, (A) and score (B) and/or the artificial intelligence heart failure with preserved ejection fraction model. Assuming a 20% prevalence of heart failure with preserved ejection fraction in the target population, and a threshold probability of 30%, managing a patient based on the combined information provided by the PEFF or H2FPEF score and the artificial intelligence heart failure with preserved ejection fraction model results in 5 to 6 more patients with heart failure with preserved ejection fraction (out of assumed 20 in the population) being managed appropriately compared to managing based on the PEFF or H2FPEF scores alone. Further details are provided in the Supplemental Appendix. AI = artificial intelligence; H2FPEF = Heavy, Hypertensive, Atrial Fibrillation, Pulmonary Hypertension, Elder, and Filling Pressure; HFA-PEFF = Heart Failure Association-Pretest Assessment, Echocardiographic and Natriuretic Peptide Score, Functional Testing, and Final Etiology; HFpEF = heart failure with preserved ejection fraction.