Fig. 1: Feature importance of categorical variables across four ensemble learning models for early CKD detection using phenotypic data. | npj Digital Medicine

Fig. 1: Feature importance of categorical variables across four ensemble learning models for early CKD detection using phenotypic data.

From: Ensemble learning approaches for early prediction of chronic kidney disease based on polysomnographic phenotype analysis

Fig. 1: Feature importance of categorical variables across four ensemble learning models for early CKD detection using phenotypic data.

a Random Forest (RF), b XGBoost, c LightGBM, and d CatBoost. Each plot highlights the most influential categorical variables, ranked by their contribution to model predictions. These features are critical in CKD classification, providing insights into key phenotypic markers associated with disease severity.

Back to article page