• Version
• Download 15
• File Size 515.73 KB
• File Count 1
• Create Date 19 March 2026
• Last Updated 19 March 2026

Predictive Analytics Model for Healthcare Readmission Risk

P.BALAKISHAN1, G.SHASHWITHA2, K.SAI PREETHAM3 , M.SAMYUKTHA4 ,MD RAIYAN ALI5
1Associate Professor, Department of CSE(AI&ML), Jyothishmathi Institute of Technology and Science, Telangana, India,
2UG Student, Department of CSE(AI&ML), Jyothishmathi Institute of Technology and Science, Telangana, India
3UG Student, Department of CSE(AI&ML), Jyothishmathi Institute of Technology and Science, Telangana, India,
4UG Student, Department of CSE(AI&ML), Jyothishmathi Institute of Technology and Science, Telangana, India,
5UG Student, Department of CSE(AI&ML), Jyothishmathi Institute of Technology and Science, Telangana, India

Abstract -This paper presents a predictive analytics model for identifying the risk of hospital readmission within 30 days of discharge using machine learning and deep learning techniques. Hospital readmissions are a major challenge in healthcare systems, leading to increased costs and reduced quality of patient care. The proposed system utilizes structured Electronic Health Record (EHR) data, including patient demographics, medical history, laboratory results, and admission details. in hospitals can be a result of deficiency in the care quality and can also cause problems in follow-up plans and instructions for patients regarding their health conditions. The methodology involves data preprocessing, feature selection, and training multiple machine learning models such as Logistic Regression, Decision Tree, Random Forest, and XGBoost, along with deep learning models.The models are evaluated using performance metrics such  as accuracy, precision, recall, and F1-score. Among all models, XGBoost achieved the best performance with an accuracy of approximately [75]%, outperforming traditional approaches. The developed system provides a risk score indicating the likelihood of patient readmission, enabling healthcare professionals to take preventive actions. This approach improves early decision-making, reduces unnecessary readmissions, and enhances overall patient outcomes.
Key Words: Hospital Readmission, Readmission Risk Prediction, Healthcare Analytics, Predictive Modeling, Clinical Decision Support, Patient Outcome Analysis, 30 Day Readmission.