Fig. 1: SHAP summary plot for 10-year mortality prediction in PD using XGBoost algorithm. | npj Parkinson's Disease

Fig. 1: SHAP summary plot for 10-year mortality prediction in PD using XGBoost algorithm.

From: Prediction of all-cause mortality in Parkinson’s disease with explainable artificial intelligence using administrative healthcare data

Fig. 1

Each point represents a SHAP value for a single participant. The colors indicate the feature values, with purple denoting higher values and yellow denoting lower values. The features are ranked in descending order of importance on the y-axis. The plot highlights the 20 most significant predictors of 10-year mortality after PD, including age, sex, comorbidities, body mass index, and hemoglobin level.

Back to article page