🏥 Hospital Readmission Prediction System

🔥 Live Demo

🚀 Live Application: https://delightful-flower-0e8ce2510.1.azurestaticapps.net

📹 Video Demo: Google Drive

🎓 RMD Engineering College - Team SuperNexis

Domain: Life Sciences
Team Name: SuperNexis Team No: 5 Problem: Predicting Hospit
Solution: Meta-Classifier Ensemble with Explainable AI

🎯 Problem & Solution

The Problem Statement

Hospital readmissions are one of the costliest challenges facing healthcare systems, but conventional models fail to predict readmissions well. Many existing models use exclusively manually-engineered features, which are dataset specific.

Our Solution

A meta-classifier ensemble system that:

• Predicts readmission risk with >78% accuracy
• Provides explainable AI insights using SHAP
• Offers dual-mode interface (manual + Excel batch processing)
• Deployed on Microsoft Azure with enterprise security

⚡ Key Features

AI-Powered Intelligence

• Meta-Classifier Ensemble: Advanced two-level architecture
• Real-time Predictions: Sub-second response times
• SHAP Explainable AI: Transparent feature importance with clinical reasoning
• Risk Stratification: Automated patient categorization (Low/Medium/High risk)

User Experience

• Responsive Design: Works on desktop, tablet, and mobile devices
• Dual Input Modes: Manual entry or Excel batch processing
• Interactive Dashboard: Real-time visualizations with exportable reports
• Enterprise Authentication: Firebase-powered secure user management

Cloud Architecture

• Azure Static Web Apps: Global CDN with edge optimization
• Auto-scaling Backend: Azure App Service with intelligent scaling
• Enterprise Security: HTTPS, CORS, and HIPAA-ready compliance
• Real-time Monitoring: Application insights and health checks

🛠️ Technology Stack

Frontend

• React 19 - UI Framework
• Vite - Build Tool
• Tailwind CSS - Styling
• Recharts - Data Visualization
• Firebase Auth - Authentication

Backend

• FastAPI - Web Framework
• Python 3.12 - Programming Language
• XGBoost - Gradient Boosting
• Logistic Regression - Meta-Classifier
• Random Forest - Ensemble Learning
• SHAP - Explainable AI
• pandas - Data Processing

Infrastructure

• Azure Static Web Apps - Frontend Hosting
• Azure App Service - Backend API
• GitHub Actions - CI/CD Pipeline

📊 Machine Learning Model

Meta-Classifier Ensemble Architecture

Our system uses a two-level ensemble approach:

Level 1: Base Classifiers

• Random Forest - Handles feature interactions, robust to outliers
• XGBoost - Gradient boosting with high accuracy
• Logistic Regression - Linear classification with probabilistic output

Level 2: Meta-Classifier

• Logistic Regression - Combines predictions from base classifiers optimally

User Interface

• Dashboard

Features

• Individual Patient Assessment: Enter patient data manually
• Batch Processing: Upload Excel files with multiple patients
• Real-time Visualizations: Interactive charts and risk breakdowns
• SHAP Explanations: See exactly why the AI made its prediction
• Mobile-Friendly: Works on all devices

🚀 Quick Start

Prerequisites

• Node.js 18+ and npm
• Python 3.12
• Git

Setup

# 1. Clone Repository
git clone https://github.com/Mohan-Balaji/hospital-readmission-prediction.git
cd hospital-readmission-prediction

# 2. Backend Setup
cd backend
python -m venv venv
venv\Scripts\activate  # Windows
pip install -r requirements.txt
uvicorn app:app --reload --host 0.0.0.0 --port 8000

# 3. Frontend Setup (new terminal)
cd frontend
npm install
npm run dev

Access Points

• Frontend: http://localhost:5173
• Backend API: http://localhost:8000
• API Docs: http://localhost:8000/docs

📚 API Documentation

Key Endpoints

Health Check

GET /health

Predict Readmission Risk

POST /predict
Content-Type: application/json

{
  "age": "[50-60)",
  "time_in_hospital": 3,
  "n_lab_procedures": 39,
  "n_procedures": 10,
  "n_medications": 79,
  "n_outpatient": 0,
  "n_inpatient": 10,
  "n_emergency": 9,
  "medical_specialty": "Other",
  "diag_1": "Respiratory",
  "diag_2": "Other",
  "diag_3": "Circulatory"
}

Get SHAP Explanation

POST /explain
Content-Type: application/json

{
  "age": "[50-60)",
  "time_in_hospital": 3,
  "n_lab_procedures": 39,
  "n_procedures": 10,
  "n_medications": 79,
  "n_outpatient": 0,
  "n_inpatient": 10,
  "n_emergency": 9,
  "medical_specialty": "Other",
  "diag_1": "Respiratory",
  "diag_2": "Other",
  "diag_3": "Circulatory"
}

👥 Team

Team SuperNexis - RMD Engineering College

• GitHub: @Mohan-Balaji

Academic Institution: RMD Engineering College
Department: Artificial Intelligence and Machine Learning

📞 Contact

📧 Email: bmohanbalaji1976@gmail.com
🔗 GitHub: hospital-readmission-prediction
🏫 Institution: RMD Engineering College, Chennai
📚 Department: Artificial Intelligence and Machine Learning

Team SuperNexis