Preoperative prediction of textbook outcome in intrahepatic cholangiocarcinoma by interpretable machine learning: A multicenter cohort study

Figure 4 Machine learning prediction process diagram. Taking No. 1 patient as an example, the patient has a tumor of 8 cm (-0.154), hepatitis B virus negative (+0.197), and liver function Child A grade (+0.136). Eastern Cooperative Oncology Group 0 points (-0.922). The overall score f (x) = -0.743 points, so the model believes that the patient cannot achieve the textbook outcome, which is consistent with the actual situation. ECOG score: Eastern Cooperative Oncology Group score; SHAP: The SHapley Additive exPlanations.