Skip to main content
. 2026 Apr 1;17:1782968. doi: 10.3389/fphar.2026.1782968

FIGURE 1.

ROC curve line graph showing model performance with true positive rate versus false positive rate, area under the curve equals zero point nine three one, and a diagonal reference line labeled random.

ROC curve of the primary logistic regression model trained on the patient-level 80–20 split stratified by whether a patient had ≥ 1 high-risk PARS visit; all visits from a patient assigned to a single split and evaluated on the held-out test set.