Fig. 3
From: Machine learning-based prediction of respiratory depression during sedation for liposuction
SHAP summary plot showing the impact of features on the model’s predictions. This summary plot visualizes the distribution of SHAP values for each feature across all samples, highlighting how different values of each feature (indicated by color) influence the model’s predictions. Each point represents a SHAP value for an individual patient, with red points indicating higher feature values and blue points indicating lower feature values. The plot demonstrates that higher values of features like UADV, AV, and BMI tend to increase the risk of respiratory depression, as indicated by their positive SHAP values, while other features such as TC and TV can have a mitigating effect depending on their value. All abbreviations in this figure are clarified in the Supplementary information.
