Fig. 5: Feature importance and global interpretations based on SHAP method. | npj Precision Oncology

Fig. 5: Feature importance and global interpretations based on SHAP method.

From: Development and validation of machine learning models for young-onset colorectal cancer risk stratification

Fig. 5

A, B Global summary of the SHAP value distribution for all enrolled features. The top five informative features for YOCRC risk stratification were CRP, FIT, Lym, CEA, and LDH. SHAP SHapley Additive explanation, YOCRC Young-onset colorectal cancer, CRP C-reactive protein, FIT Fecal immunochemical test, Lym lymphocytes, CEA carcinoembryonic antigen, LDH lactate dehydrogenase, Lipa Lipoprotein a, Hb Hemoglobin, Mono Monocyte, A/G Albumin/globulin ratio, HDL high density lipoprotein, ALT alanine aminotransferase, ALP alkaline phosphatase, TC total cholesterol, PLT platelet, RDW red blood cell distribution width, WBC white blood cell.

Back to article page