Skip to main content
. 2023 Jul 28;2(6):100452. doi: 10.1016/j.jacadv.2023.100452

Figure 3.

Figure 3

Alluvial Plot Demonstrating Reclassification of Patients Using Clinical Scores Compared to the AI HFpEF Model

Patients in the independent testing data set were given a categorical score based on relevant functional and morphological echocardiographic, and biomarker parameters. Patients were categorized as unlikely (0 or 1), indeterminate (2-4), or probable (5-6) likelihood of heart failure with preserved ejection fraction for the Heart Failure Association-Pretest Assessment, Echocardiographic and Natriuretic Peptide Score, Functional Testing, and Final Etiology (HFA-PEFF) score (A), and low-probability (0 or 1), indeterminate (2-5), or high-probability (6-9) of heart failure with preserved ejection fraction for the Heavy, Hypertensive, Atrial Fibrillation, Pulmonary Hypertension, Elder, and Filling Pressure (H2FPEF) score (B). AI = artificial intelligence; HFpEF = heart failure with preserved ejection fraction.