Skip to main content
. 2025 Oct 13;12:1655302. doi: 10.3389/fmed.2025.1655302

FIGURE 7.

Side-by-side calibration curves display predicted versus observed probabilities. Panel A (Train) shows three lines: clinical (blue), image (orange), and combined (green). Panel B (Test) also shows clinical, image, and combined lines. Each line is compared against the diagonal line representing perfect calibration. Both panels highlight discrepancies between predicted and observed values across different data sets.

Calibration curves for predicting treatment outcome. Calibration curves for the clinical model, image model, and combined model in the training (A) and test (B) cohorts. Calibration curves plot observed probability versus predicted probability, with a perfectly calibrated model following the diagonal line.