🩺 Smart Health Risk Prediction System

An AI-based Smart Health Risk Prediction System developed using Machine Learning and Streamlit to predict possible diseases based on user symptoms and health parameters. The system helps in early risk assessment by analyzing symptom patterns, estimating risk levels, and recommending suitable OPD consultations.

📌 Project Overview

The Smart Health Risk Prediction System is designed to assist users in identifying potential health risks using machine learning techniques. The system takes patient details and symptoms as input, predicts the most probable disease, and classifies the risk level as Low, Medium, or High.

In addition to prediction, the system provides hospital and specialist consultation suggestions and generates a downloadable medical report in PDF format. Patient records are stored locally for future analysis and visualization of disease trends.

This project demonstrates the practical application of Artificial Intelligence and Machine Learning in healthcare decision support systems.

✨ Features

• Disease prediction using Machine Learning (Random Forest)

• Risk level classification based on prediction confidence

• Patient information management

• OPD consultation suggestions based on disease and city

• Automated medical report generation (PDF)

• Patient history storage using SQLite database

• Disease trend visualization

• Interactive Streamlit web interface

• Real-time model accuracy display

🧠 System Workflow

• User enters patient information (age, gender, height, weight, city).

• Symptoms are selected as input features.

• Data is processed and passed to the trained ML model.

• Model predicts the most probable disease.

• Prediction confidence determines risk level.

• System suggests suitable OPD consultations.

• Patient details are stored in database.

• Medical report is generated and downloadable as PDF.

• Disease trends are visualized from stored records.

🛠 Technologies Used

• Python

• Machine Learning (Scikit-learn)

• Streamlit – User Interface

• Pandas & NumPy – Data Processing

• Matplotlib – Visualization

• SQLite – Database Management

• FPDF – PDF Report Generation

⚙️ Installation & Setup

1️⃣ Clone Repository

git clone https://github.com/khushikumari-2003/Health-Risk-Prediction-System.git

cd Health-Risk-Prediction-System

2️⃣ Install Dependencies

pip install -r requirements.txt

3️⃣ Run Application

streamlit run app.py

📊 Machine Learning Model

• Algorithm Used: Random Forest Classifier

• Feature Scaling: StandardScaler

• Label Encoding for disease classification

• Model trained using symptom-based dataset

• Accuracy displayed dynamically in application

📄 Report Generation

• After prediction, the system generates a medical report containing:

• Patient details

• Predicted disease

• Risk level

• Suggested OPD consultations

• Generated timestamp

The report can be downloaded directly from the application.

📈 Visualization

• The system stores patient history and displays:

• Disease occurrence trends

• Historical patient records

• Frequency-based disease analysis

🎯 Use Cases

• Early health risk assessment

• AI-assisted healthcare support

• Educational healthcare AI projects

• Medical data analysis demonstrations

🚀 Future Improvements

• Integration with real hospital APIs

• Deep Learning-based prediction models

• Cloud database integration

• User authentication system

• Deployment on cloud platforms

👩‍💻 Author

Khushi Kumari

B.Tech CSE (Ai & Ml)