🧠 Disease Prediction System

An ML-powered web app that uses Bayes' Theorem and machine learning to estimate disease likelihood from symptoms — built for students, researchers, and developers.

🚀 Live Demo · 🐛 Report a Bug · 💡 Request a Feature

---

⚠️ Disclaimer: Educational Use Only This project is for learning and demonstration purposes only. It is not a medical tool and must not be used for real-world diagnosis or treatment. Always consult a qualified healthcare professional.

---

📋 Table of Contents

• About the Project
• Key Features
• Tech Stack
• Quick Start
• Detailed Setup
• How It Works
• Project Structure
• Dataset & Model
• Privacy & Data Handling
• Contributing
• License

---

🔍 About the Project

The Disease Prediction System makes medical reasoning transparent and interactive. It combines classical Bayesian probability with modern ML to show users not just what a prediction is, but why — step by step.

It's designed to bridge the gap between complex statistical concepts and intuitive understanding, with visual charts, AI-powered explanations, and multi-language support.

---

✨ Key Features

📘 Educational

Feature	Description
Bayes' Theorem Visualizer	Step-by-step breakdown of prior → posterior probability
Interactive Sliders	Experiment with probability values in real time
Built-in Glossary	Plain-English definitions of medical/statistical terms

🤖 ML & Prediction

Feature	Description
Symptom-Based Prediction	Select symptoms to get ML-generated probability scores
BMI Integration	Height/weight inputs contribute to risk estimation
Risk Categorization	Results classified as Low / Medium / High risk
Prediction History	Stored timeline of past predictions with analytics

🧠 AI Features

Feature	Description
AI Explanations	Gemini-powered interpretation of results
Next-Step Guidance	Suggests consultation, testing, or lifestyle review
Multi-Language Output	English 🇬🇧 · Hindi 🇮🇳 · Gujarati 🇮🇳 · Tamil 🇮🇳

📊 Analytics Dashboard

• Comparative probability trend charts
• Risk distribution graphs
• Historical prediction tables

---

🛠 Tech Stack

Layer	Technology
Backend	Python, Flask
Frontend	HTML, CSS, JavaScript
ML / Math	Bayesian Inference, ResNet CNN
AI	Google Gemini API
Dashboard	Streamlit
Data	CSV (hospital_data.csv)

---

🚀 Quick Start

Option 1 — View Online (instant, no setup) 👉 https://disease-prediction-nwnu.onrender.com

Option 2 — Run Locally (30 seconds)

git clone https://github.com/aliviahossain/Disease-prediction.git
cd Disease-prediction
pip install -r requirements.txt
python run.py

Then open your browser at: http://127.0.0.1:5001/

---

🛠️ Detailed Setup

1. Create a Virtual Environment (Recommended)

python -m venv venv
source venv/bin/activate      # macOS / Linux
venv\Scripts\activate         # Windows

2. Install Dependencies

pip install -r requirements.txt

3. (Optional) Enable AI Recommendations

Get a free API key from Google AI Studio, then set it:

# Option A: .env file (recommended)
echo "GEMINI_API_KEY=your_key_here" > .env

# Option B: Environment variable
export GEMINI_API_KEY=your_key_here    # macOS / Linux
set GEMINI_API_KEY=your_key_here       # Windows

4. Run the App

python run.py

---

📦 Deployment

The application can be deployed to Google Cloud Run.

See: docs/deployment/gcp-cloud-run.md

See docs/deployment/gcp-cloud-run.md for deployment instructions.

---

🧮 How It Works

Bayes' Theorem

The app updates disease probability after observing symptoms or test results using:

P(Disease | Evidence) = [ P(Evidence | Disease) × P(Disease) ]
                        ─────────────────────────────────────────────────────────
                        [ P(Evidence | Disease) × P(Disease) + P(Evidence | No Disease) × P(No Disease) ]

Term	Meaning
P(Disease)	Prior probability — baseline disease prevalence
P(Evidence\|Disease)	Sensitivity — how often evidence appears given disease
P(Evidence\|No Disease)	False positive rate
P(Disease\|Evidence)	Posterior — updated probability after new evidence

BMI Integration

The system accepts height (cm) and weight (kg) to compute BMI, which adjusts risk estimates:

BMI = weight (kg) / (height in meters)²

BMI Range	Category
Below 18.5	Underweight
18.5 – 24.9	Normal
25 – 29.9	Overweight
30+	Obese

---

🗂️ Project Structure

Disease-prediction/
├── run.py                         # App entry point
├── dashboard.py                   # Streamlit analytics dashboard
├── requirements.txt               # Python dependencies
├── hospital_data.csv              # Bayesian stats (generated by pipeline)
├── data/
│   ├── raw/                       # Original datasets (git-ignored)
│   ├── cleaned/
│   │   └── hospital_data_cleaned.csv
│   └── preprocess.py              # Data pipeline script
├── backend/
│   ├── routes/                    # Flask routes (auth, ML, calculator)
│   ├── models/                    # Database & ML models
│   ├── utils/                     # Bayesian calculator & AI helpers
│   ├── static/                    # JS & CSS
│   └── templates/                 # HTML templates
├── README.md
└── LICENSE

---

📊 Dataset & Model

Bayesian Calculator Datasets

Disease	Dataset	Source
Heart Disease	UCI Heart Disease Dataset	UCI Repository
Diabetes	Pima Indians Diabetes Dataset	Kaggle
Breast Cancer	Breast Cancer Wisconsin	UCI Repository

To regenerate hospital_data.csv from raw datasets:

python data/preprocess.py

Raw dataset files are not committed to this repo. Download them from the links above and place them in data/raw/.

Image Classification Datasets

Model	Dataset	Conditions
Eye Disease CNN	Eye Diseases Classification (Kaggle)	Glaucoma, Diabetic Retinopathy, Cataract, Normal
Skin Disease CNN	Skin Diseases Image Dataset (Kaggle)	Melanoma, Eczema, Psoriasis, and 7 others

The CNN architecture is based on ResNet, inspired by He et al. (2016) — Deep Residual Learning for Image Recognition.

---

🔒 Privacy & Data Handling

• ✅ All calculations run locally on your machine
• ✅ No symptoms, images, or personal data are uploaded or stored externally
• ✅ AI requests only transmit probability values — never user identity
• ✅ Prediction history is stored locally using JSON-based storage

---

🔧 Troubleshooting

AI recommendations not working?

• Make sure GEMINI_API_KEY is set correctly
• Restart the application after setting the key
• Check your internet connection
• Verify your API quota at Google AI Studio

App not starting?

• Confirm Python version is 3.9 or higher: python --version
• Activate your virtual environment before running
• Reinstall dependencies: pip install -r requirements.txt

---

🤝 Contributing

Contributions are warmly welcome! Here's how to get started:

1. Fork the repository
2. Create a new branch: git checkout -b feature/your-feature-name
3. Make your changes and commit: git commit -m "Add: your meaningful message"
4. Push to your fork: git push origin feature/your-feature-name
5. Open a Pull Request

Please read CONTRIBUTING.md and CODE_OF_CONDUCT.md before contributing.

---

📜 License

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

---

Created and maintained by Alivia Hossain

⭐ If you find this project useful, consider giving it a star!