Machine Learning

Hospital Readmission Risk Prediction Using Machine Learning

The Hospital Readmission Risk Prediction Using Machine Learning system is a comprehensive machine learning solution that predicts 30-day hospital readmission risk for diabetic patients. This project is ideal for B.Tech, MCA, CSE, and engineering students as their final year project. It implements Random Forest and XGBoost with balanced sample weighting, achieving 68.4% accuracy and 0.766 ROC-AUC.

The system utilizes the Diabetes 130-US hospitals dataset from the UCI Machine Learning Repository, containing over 100,000 patient records with 50+ clinical features including patient demographics, diagnoses, medications, and hospital visit history. It provides a web-based interface for real-time risk prediction, enabling healthcare providers to identify high-risk patients and implement targeted interventions, potentially reducing readmission rates and improving patient outcomes.

• Multi-Model Readmission Prediction
• XGBoost (68.4% Accuracy)
• Random Forest (67.2% Accuracy)
• Feature Engineering (composite risk scores)
• Balanced Sample Weighting for Class Imbalance
• Interactive Visualizations
• Real-time Risk Predictions
• Model Performance Comparison
• Feature Importance Analysis
• Flask Web Application