Fig. 6

SHAP decision plot of model for different samples. The SHAP values accumulation process for a single sample is represented by the colored lines indicating whether each feature moves the output value above or below the average prediction, with the gray vertical line marking the model’s baseline value. The feature values are provided for reference next to the prediction lines. Starting from the bottom of the graph, the prediction lines show how the SHAP values accumulate from the baseline to the top of the graph, representing the model’s final score (predicted concentration value). Each line represented a single sample of the testing group. CL, clearance; TSLD, time since last dose; Cr, creatinine; UA, uric acid; Hgb, hemoglobin; SOFA, sequential organ failure assessment; Vd, volume of distribution; TP, total protein; ALT, alanine aminotransferase; Cys-C, cystatin C.