Fig. 6 | Scientific Reports

Fig. 6

From: Development and validation of interpretable machine learning models to predict distant metastasis and prognosis of muscle-invasive bladder cancer patients

Fig. 6

ML model interpretation. (A) Importance ranking of features in 13 ML prediction algorithms. (B) The importance ranking of different variables according to the mean (|SHAP value|) in the optimal CatBoost model. (C) The importance ranking of different risk factors predicting metastasis with stability and interpretation in the optimal CatBoost model. The higher SHAP value of a feature is given, the higher risk of distant metastasis the patient would have. The yellow part in feature value represents higher value. (D) SHAP value explanation in a classical sample without distant metastasis. (E) SHAP value explanation in a classical sample with distant metastasis.

Back to article page