🏥 Smart Public Health Warning System - Phase 3 Complete

AI-powered disease outbreak prediction and health monitoring system for Indian cities.

📋 Project Overview

The Smart Public Health Warning System is a comprehensive DBMS + Machine Learning project that monitors public health data across Indian cities and predicts potential disease outbreaks using historical patterns, seasonal trends, and environmental factors.

Key Features

✅ Phase 1 & 2 (DBMS)

• Comprehensive relational database with 8 tables
• Real-time health data dashboard
• Environmental monitoring integration
• Interactive SQL query interface
• Hospital capacity tracking
• Health alert management

✅ Phase 3 (Machine Learning)

• AI-powered outbreak prediction
• Seasonal pattern recognition
• Environmental correlation analysis
• City-wide risk assessment
• Disease-specific forecasting
• Realistic probability calculations

---

🗂️ Project Structure

SmartHealthWarningSystem/
├── app_phase3.py                 # Main Flask application with ML
├── ml_predictor.py               # ML prediction module
├── templates/                    # HTML templates
│   ├── dashboard.html            # Main dashboard
│   ├── predictions.html          # ML predictions interface
│   └── query.html               # SQL query interface
├── Data Files (CSV):
│   ├── cities.csv               # 20 Indian cities
│   ├── hospitals.csv            # 150 hospitals
│   ├── diseases.csv             # 15 diseases
│   ├── symptoms.csv             # 51 symptoms
│   ├── disease_symptoms.csv     # Disease-symptom mappings
│   ├── patient_reports.csv      # 2,500 patient cases
│   ├── environmental_data.csv   # 1,500 environmental readings
│   └── alerts.csv              # 400 health alerts
└── health_warning_system.db    # SQLite database (auto-generated)

---

🚀 Quick Start Guide

Prerequisites

• Python 3.8 or higher
• pip (Python package manager)
• VS Code (recommended)

Installation Steps

# 1. Navigate to project folder
cd SmartHealthWarningSystem

# 2. Create virtual environment
python -m venv health_env

# 3. Activate virtual environment
# Windows:
health_env\Scripts\activate
# Mac/Linux:
source health_env/bin/activate

# 4. Install required packages
pip install flask pandas numpy

# 5. Run the application
python app_phase3.py

# 6. Open browser
# Dashboard: http://localhost:5000
# ML Predictions: http://localhost:5000/predictions
# SQL Queries: http://localhost:5000/query

---

📊 Database Schema

Tables

1. cities - Master list of Indian cities
2. hospitals - Healthcare facilities
3. diseases - Disease catalog
4. symptoms - Medical symptoms
5. disease_symptoms - Disease-symptom mappings (junction table)
6. patient_reports - Patient case records (main fact table)
7. environmental_data - Air quality, water quality, temperature, etc.
8. alerts - Health warnings and notifications

Relationships

• Cities ← (One-to-Many) → Hospitals, Patient Reports, Environmental Data, Alerts
• Diseases ← (Many-to-Many) → Symptoms (via disease_symptoms)
• Diseases ← (One-to-Many) → Patient Reports
• Hospitals ← (One-to-Many) → Patient Reports

---

🧠 Machine Learning Model

How It Works

Our ML model predicts outbreak probability using:

1. Historical Pattern Analysis

   • Last 6 months of case data
   • Average case counts
   • Recent trends

2. Seasonal Factors

   • Monsoon diseases (Jun-Sep): Dengue, Malaria, Typhoid
   • Winter diseases (Nov-Feb): H1N1, Pneumonia, TB
   • Summer diseases (Mar-May): Chickenpox, Measles

3. Environmental Correlation

   • Air Quality Index (AQI) → Respiratory diseases
   • Water Quality Index (WQI) → Waterborne diseases
   • Temperature → Vector-borne diseases

4. Risk Classification

   • Low Risk: Probability 0-40%
   • Medium Risk: Probability 40-60%
   • High Risk: Probability 60-80%
   • Critical Risk: Probability 80-100%

Realistic Predictions

The model is designed to provide realistic predictions:

• Not every disease triggers an outbreak warning
• Accounts for off-season patterns (reduced risk)
• Considers environmental factors
• Includes random variation (±20%)
• Based on actual historical data patterns

---

🎯 Features Breakdown

1. Real-time Dashboard (/)

• Disease Outbreak Summary: Current outbreak patterns
• Environmental Risk: Cities with poor air/water quality
• Hospital Capacity: Bed utilization rates
• Active Alerts: Current health warnings

2. ML Predictions (/predictions)

• City Risk Overview: Overall risk level for all cities
• Detailed Predictions: Top disease risks per city
• Probability Scores: Likelihood of outbreak
• Predicted Case Counts: Expected number of cases
• Reasoning: Why the prediction was made

3. SQL Query Interface (/query)

• Execute custom SELECT queries
• Sample queries included
• Export results as CSV
• Learn database structure

---

📈 Sample Predictions Output

City: Mumbai
Top Disease: Dengue
Risk Level: High
Probability: 72%
Predicted Cases: 45
Reasoning: Peak season for this disease; Environmental conditions favor disease spread

---

🛠️ Troubleshooting

Issue: No data on dashboard

Solution:

del health_warning_system.db
python app_phase3.py

Issue: Module not found

Solution:

health_env\Scripts\activate
pip install flask pandas numpy

Issue: Port already in use

Solution: Change port in app_phase3.py:

app.run(debug=True, host='0.0.0.0', port=5001)

---

🎓 Educational Value

DBMS Concepts Demonstrated

• Database normalization (1NF, 2NF, 3NF)
• Foreign key constraints
• Complex SQL queries (JOINs, aggregates, subqueries)
• Indexing for performance
• Views for data abstraction
• Transaction management

ML Concepts Demonstrated

• Time series analysis
• Feature engineering
• Seasonal pattern recognition
• Environmental correlation
• Risk classification
• Predictive modeling

---

📝 Key Files Explained

app_phase3.py

Main Flask application that:

• Initializes database
• Imports CSV data
• Serves web interface
• Provides API endpoints
• Integrates ML predictions

ml_predictor.py

Machine learning module that:

• Analyzes historical patterns
• Calculates seasonal factors
• Correlates environmental data
• Predicts outbreak risk
• Provides reasoning for predictions

templates/predictions.html

ML predictions interface that:

• Displays city risk overview
• Shows detailed disease predictions
• Visualizes probability scores
• Explains prediction reasoning

---

🌟 Future Enhancements (Optional)

1. Real-time data integration via APIs
2. Geographic visualization with maps
3. Mobile app for field workers
4. Email/SMS alert notifications
5. Advanced ML models (LSTM, Random Forest)
6. User authentication and role management
7. Data export and reporting features

---

👥 Team Information

Project: Smart Public Health Warning System Phase: Phase 3 (Complete with ML) Team: Health Sentinel Course: DBMS Project

---

📜 License

This project is created for educational purposes as part of DBMS coursework.

---

🙏 Acknowledgments

• Historical disease data patterns based on actual Indian epidemiological studies
• Environmental data thresholds from WHO and CPCB guidelines
• Machine learning approach inspired by public health forecasting systems

---

📞 Support

For issues or questions:

1. Check troubleshooting section above
2. Verify all CSV files are present
3. Ensure virtual environment is activated
4. Check console output for error messages

---

Project Status: ✅ Complete and Production Ready

Last Updated: November 19, 2025