🏥 Health-Insight using machine learing and flask

End-to-End ML-Powered Disease Prediction Web Application

Health-Insight is a full-stack web application that integrates multiple machine learning models with a Flask backend to provide real-time disease risk predictions. The system is designed with production-style considerations such as feature consistency, model serialization, dynamic form generation, and robust error handling.

🚀 Features

Multi-disease prediction (Diabetes, Heart, Kidney, Liver, Cancer)

Individual ML model per disease

Dynamic input forms based on disease-specific features

Flask REST backend with real-time inference

Model persistence using Pickle

Input validation and categorical encoding

Scalable and modular project structure

🛠 Tech Stack

Backend: Python, Flask

Machine Learning: Scikit-learn, Pandas, NumPy

Models: RandomForestClassifier

Frontend: HTML, CSS, JavaScript

Serialization: Pickle

📂 Project Structure

 Health-Insight/ <br>
│ <br>
├── app.py  <br>
├── models/ <br>
│   ├── diabetes.pkl  <br>
│   ├── heart.pkl   <br>
│   ├── kidney.pkl  <br>
│   ├── liver.pkl   <br>
│   └── cancer.pkl  <br>
│  <br>
├── training/  <br>
│   ├── diabetes.py  <br>
│   ├── heart.py  <br>
│   ├── kidney.py  <br>
│   ├── liver.py  <br>
│   └── cancer.py  <br>
│  <br>
├── datasets/  <br>
│   ├── diabetes.csv  <br>
│   ├── heart.csv  <br>
│   ├── kidney.csv  <br>
│   ├── liver.csv  <br>
│   └── cancer.csv  <br>
│   <br>
├── templates/  <br>
│   ├── index.html  <br>
│   ├── form.html   <br>
│   └── result.html  <br>
│   <br>
├── static/   <br>
│   ├── css/  <br>
│   └── js/   <br>
│  <br>
└── README.md  <br>
<br>

⚙️ How It Works

- Each disease has a separately trained ML model.

- Models are serialized using Pickle.

- Flask dynamically loads the correct model and feature set.

- User inputs are validated and transformed before prediction.

- Predictions are returned in real time via the web interface.

▶️ Run Locally

pip install -r requirements.txt <br>
python app.py <br>

Open http://127.0.0.1:5000/ in your browser.

🧠 Key Learnings

- Maintaining feature consistency between training and inference

• Handling categorical encoding in production ML systems

• Debugging real-world ML deployment issues

• Designing modular and scalable backend architecture

📌 Future Improvements

• REST API endpoints

• Authentication & user history

• Model monitoring & retraining pipeline

• Dockerized deployment