Run Health_edu by ---> streamlit run ossp_pbl.py

🏥 HealthEdu - AI-Powered Health Education & Disease Prediction System

A comprehensive health education platform that leverages Random Forest Classification to predict diseases based on symptoms and provides personalized health recommendations including medications, dietary plans, precautions, and workout routines. Built with a strong emphasis on medical ethics and educational transparency.

🌟 Features

🤖 Machine Learning Core

• Random Forest Classifier: Advanced ensemble learning for accurate disease-symptom pattern matching
• Multi-Label Symptom Processing: Handles complex symptom combinations
• 30% Test Split Validation: Rigorous model testing for reliability
• Real-time Prediction Engine: Instant disease classification from user inputs

🔍 Symptom Information System

• Interactive symptom selection interface
• Pattern-based condition matching
• Educational disease information lookup
• Multi-symptom analysis capabilities

💊 Medication Education Tools

• Drug-condition association learning
• Integration with RxTerms API (NLM Clinical Tables)
• Links to Indian government portals (CDSCO, IPC)
• Access to verified databases (1mg, PharmEasy, Drugs.com)
• Intelligent medicine name normalization and fallback queries

🔍 Medication Verification System

• Educational medication-condition verification
• Cross-reference with medical databases
• Safe medication information lookup

📊 Health Data Visualization

• Interactive Charts: Condition frequency analysis using Streamlit charts
• Pie Charts: Top 10 conditions distribution with Matplotlib
• Scatter Plots: Symptom correlation analysis using Plotly
• Real-time data exploration tools

🛡️ Safety & Ethics

• Comprehensive medical disclaimers throughout the application
• Clear distinction between education and medical advice
• Professional resource directory (WHO, CDC, Mayo Clinic)
• Indian-specific medical resources (Emergency: 112, 102/108, 104)
• Emphasis on consulting healthcare professionals

🚀 Quick Start

Prerequisites

• Python 3.8 or higher
• pip package manager

Installation

1. Clone the repository:

git clone https://github.com/sobhan2204/HealthEdu.git
cd HealthEdu

2. Install required dependencies:

pip install -r req.txt

Required packages:

• streamlit
• pandas
• matplotlib
• plotly
• scikit-learn
• requests
• urllib3

Running the Application

Launch the Streamlit app:

streamlit run Health_edu.py

The application will open automatically in your default web browser at http://localhost:8501

📊 Dataset Architecture

The project utilizes 8 comprehensive medical datasets:

Dataset	Purpose	Contents
Training.csv	ML Model Training	Disease-symptom relationships for Random Forest
Symptom-severity.csv	Severity Analysis	Weighted severity scores for symptoms
symptoms_df.csv	Symptom Database	Comprehensive symptom catalog
description.csv	Medical Information	Detailed disease descriptions
medications.csv	Pharmacology	Medication recommendations per condition
diets.csv	Nutrition	Dietary guidelines for conditions
precautions_df.csv	Safety Guidelines	Precautionary measures
workout_df.csv	Fitness Plans	Condition-specific exercise routines

🤖 Machine Learning Pipeline

Architecture & Implementation

# Feature Engineering
symptom_columns = training_df.columns[1:-1]
X = training_df[symptom_columns]  # Multi-symptom features
y = training_df[disease_column]    # Disease labels

# Train-Test Split (70-30)
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.3, random_state=42
)

# Random Forest Classifier
model = RandomForestClassifier(random_state=42)
model.fit(X_train, y_train)

# Prediction with accuracy tracking
y_pred = model.predict(X_test)
accuracy = accuracy_score(y_test, y_pred)

Key ML Features:

• Algorithm: Random Forest (ensemble method with multiple decision trees)
• Validation: 70-30 train-test split for robust evaluation
• Input Processing: Binary symptom encoding (present=1, absent=0)
• Output: Multi-class disease classification
• Performance Metrics: Accuracy score and classification reports

API Integrations:

• RxTerms API: NLM Clinical Tables for medication information
• Caching: @st.cache_data for optimized API performance (1-hour TTL)
• Error Handling: Comprehensive fallback mechanisms

🛠️ Technology Stack

Machine Learning & Backend (Developed by sobhan2204)

• Python 3.8+: Core programming language
• Scikit-learn: Random Forest implementation, preprocessing, metrics
• Pandas: Data manipulation and CSV processing
• NumPy: Numerical computations (via sklearn dependencies)
• Requests: External API integration
• Urllib: URL encoding for API calls

Frontend & Visualization (Developed by my friend)

• Streamlit: Interactive web application framework
• Matplotlib: Static visualizations (pie charts, histograms)
• Plotly Express: Interactive 3D scatter plots and dynamic charts
• Custom CSS: Styled UI components with disclaimer boxes

📁 Project Structure

HealthEdu/
├── Health_edu.py              # Main Streamlit application (ML + Frontend)
│   ├── ML Components:
│   │   ├── Random Forest training
│   │   ├── Symptom preprocessing
│   │   ├── Disease prediction
│   │   └── API integrations
│   └── Frontend Components:
│       ├── Streamlit UI
│       ├── Interactive visualizations
│       └── User interface design
│
├── Training.csv               # ML training dataset (disease-symptom matrix)
├── Symptom-severity.csv       # Symptom severity weights
├── symptoms_df.csv            # Symptom reference database
├── description.csv            # Disease information
├── medications.csv            # Medication database
├── diets.csv                  # Dietary recommendations
├── precautions_df.csv         # Safety precautions
├── workout_df.csv             # Exercise routines
├── req.txt                    # Python dependencies
└── README.md                  # Documentation

💡 How It Works

1. User Input

Users select symptoms from an interactive multi-select interface with 5-column button layout

2. Data Preprocessing

# Binary encoding of symptoms
input_data = {symptom: (1 if symptom in selected else 0) 
              for symptom in symptom_columns}
input_df = pd.DataFrame([input_data])

3. ML Prediction

Random Forest classifier analyzes symptom patterns and predicts the most likely condition

4. Information Retrieval

System fetches relevant information from multiple CSV databases:

• Disease descriptions
• Medication recommendations (with API verification)
• Dietary guidelines
• Precautionary measures
• Workout suggestions

5. Educational Presentation

Results displayed with clear disclaimers, professional resource links, and next steps

🎯 Application Modules

🏠 Home

• Platform overview and mission
• Clear educational disclaimers
• Usage instructions
• Professional resource directory

📋 Symptom Information System

• Multi-symptom selection interface
• ML-based pattern matching
• Condition information display
• Doctor consultation guidance

💊 Medication Education

• Symptom-to-medication learning
• Interactive medication lookup
• RxTerms API integration
• Indian government portal links
• Never replaces prescriptions

🔍 Medication Verification

• Condition-medication association checker
• Educational database cross-reference
• Individual medicine lookup
• Safety information

📊 Data Visualization

• Condition frequency bar charts
• Distribution pie charts (Top 10)
• Symptom scatter plots with Plotly
• Interactive data exploration

🏥 Professional Resources

• Emergency contacts (India: 112, 102/108, 104)
• WHO, CDC, Mayo Clinic links
• Indian government health portals
• Verified medical databases

⚠️ Critical Disclaimers

This application is STRICTLY educational and NOT for medical diagnosis, treatment, or prescription.

• ❌ NOT a substitute for professional medical advice
• ❌ NOT a diagnostic tool
• ❌ NOT a prescription service
• ✅ IS an educational learning platform
• ✅ IS designed to facilitate informed conversations with doctors

Always consult qualified healthcare professionals for:

• Medical diagnosis
• Treatment decisions
• Medication prescriptions
• Health concerns

Emergency contacts are provided throughout the application.

🌍 Indian Healthcare Integration

Special features for Indian users:

• CDSCO (Central Drugs Standard Control Organization)
• IPC (Indian Pharmacopoeia Commission)
• Emergency numbers: 112 (National), 102/108 (Ambulance), 104 (Medical Helpline)
• Links to 1mg, PharmEasy, Apollo Pharmacy
• Doctor search via Practo, DocPrime, Lybrate

👥 Development Credits

🤖 Machine Learning & Backend Development

Developed by: sobhan2204

Responsibilities:

• Random Forest classifier implementation
• Data preprocessing and feature engineering
• API integrations (RxTerms, medical databases)
• Prediction engine development
• Model training and validation
• Backend logic and data flow

🎨 Frontend Development

Developed by: My talented friend

Responsibilities:

• Streamlit interface design
• Custom CSS styling
• Interactive visualizations (Matplotlib, Plotly)
• User experience optimization
• UI components and layouts

🤝 Contributing

Contributions are welcome! Please follow these steps:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

Areas for contribution:

• Additional ML algorithms (SVM, Neural Networks)
• Expanded datasets
• Multi-language support
• Mobile responsiveness
• Additional visualizations

📝 Future Enhancements

• Deep learning models (CNN, LSTM)
• Medical imaging analysis
• Chatbot integration with LLMs
• Multi-language support (Hindi, regional languages)
• Mobile app (React Native/Flutter)
• Wearable device integration
• Enhanced API integrations
• User health history tracking
• Telemedicine platform integration
• Expanded Indian regional medical resources

🐛 Known Issues

• CSV filename typo handling: symptoms_df.csv vs symtoms_df.csv
• RxTerms API may have rate limits (cached for 1 hour)
• Some medications may require name normalization for accurate lookup

📊 Model Performance

The Random Forest classifier is trained with:

• Test size: 30% of dataset
• Random state: 42 (reproducible results)
• Validation: Accuracy score and classification report
• Real-time inference: < 1 second prediction time

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

📧 Contact

Sobhan - @sobhan2204

Project Link: https://github.com/sobhan2204/HealthEdu

---

⭐ If you find this project helpful for learning about health and ML, please star the repository!

Made with ❤️ for health education, medical ethics, and responsible AI

---

🙏 Acknowledgments

• Medical datasets used for educational purposes
• NLM Clinical Tables (RxTerms API)
• Indian government health portals
• Streamlit community
• Scikit-learn documentation
• Healthcare professionals who inspire ethical health tech