HOSPITAL 30 DAY PATIENT READMISSION RISK PREDICTION USING CLINICAL AND DEMOGRAPHIC FEATURES WITH SHAP EXPLAINABILITY

Abstract

Hospital readmissions within 30 days represent a significant challenge in modern healthcare systems, as they increase operational costs and often indicate gaps in the quality of patient care. This project presents a predictive analytics framework aimed at identifying patients who are at high risk of readmission by utilizing both clinical and demographic data. The dataset includes important features such as patient age, medical history, diagnosis information, length of hospital stay, prior admissions, and treatment details, enabling a comprehensive assessment of patient health status. To build an effective prediction system, multiple machine learning algorithms— including Logistic Regression, Random Forest, and Gradient Boosting—are implemented and compared. Data preprocessing techniques such as handling missing values, feature encoding, and normalization are applied to enhance data quality and model performance. The models are evaluated using performance metrics such as accuracy, precision, recall, and F1-score to ensure reliable and robust predictions. A key contribution of this project is the integration of explainable artificial intelligence using SHAP (SHapley Additive exPlanations). This approach provides clear insights into how different features influence the prediction of readmission risk, allowing healthcare professionals to understand and trust the model’s decisions. By highlighting the most impactful factors, SHAP enhances transparency and supports informed clinical judgment.