DiabeticAI – Diabetes Risk Prediction System

DiabeticAI is a machine learning-powered web application that predicts the risk of diabetes using clinical health indicators. The system integrates a trained ML model with a Django REST API and a React frontend to provide real-time predictions.

The goal of this project is to demonstrate end-to-end ML system development, including data preprocessing, model training, API integration, and interactive prediction interfaces.

---

Live Demo

https://diabetes-model-mhzi.vercel.app/

---

Features

• Machine learning model for diabetes risk prediction
• Data preprocessing and feature engineering pipeline
• Real-time predictions through REST API
• React frontend for user input and result visualization
• Modular backend architecture using Django REST Framework

---

Tech Stack

Backend

• Python
• Django
• Django REST Framework
• scikit-learn
• Pandas
• NumPy

Frontend

• React
• Vite
• TailwindCSS

Tools

• Git
• REST APIs
• Model Serialization (Pickle / Joblib)

---

Machine Learning Pipeline

The ML workflow includes:

1. Data preprocessing
2. Handling missing values
3. Feature scaling and normalization
4. Model training using supervised classification
5. Model evaluation using accuracy and ROC metrics
6. Model serialization and deployment via API

Example pipeline:

Raw Data
   ↓
Preprocessing
   ↓
Feature Engineering
   ↓
Model Training
   ↓
Evaluation
   ↓
Django REST API
   ↓
React Frontend Prediction

---

Project Structure

project/
│
├── backend/
│   └── diabetes/
│       ├── models
│       ├── views
│       ├── serializers
│       └── settings
│
├── src/            # React frontend
├── public/
│
├── analysis.csv
├── offline-analysis.ipynb
└── README.md

---

How Predictions Work

1. User enters health parameters in the frontend.
2. Data is sent to the Django API endpoint.
3. Backend preprocesses the input data.
4. The trained ML model predicts diabetes risk.
5. Prediction is returned to the frontend and displayed to the user.

---

Local Setup

Clone the repository

git clone https://github.com/deepmhatre13/diabetes_model
cd diabetes_model

Create virtual environment

python -m venv venv
source venv/bin/activate

Install dependencies

pip install -r requirements.txt

Run backend

python manage.py runserver

Run frontend

npm install
npm run dev

---

Key Engineering Highlights

• Designed end-to-end ML prediction pipeline from preprocessing to deployment
• Integrated trained ML model into Django REST APIs for real-time inference
• Built React frontend to interact with ML prediction service
• Structured project for modular backend and scalable API development

---