How I Built Swasthya Setu — An AI-Powered Health App Using the MERN Stack
By Subham Khandual | MERN Stack Developer | Final Year CSE Student, Bhubaneswar, India
👋 Introduction
Hi, I'm Subham Khandual, a Final Year B.Tech Computer Science student from Bhubaneswar, Odisha, India. I'm a MERN Stack Developer passionate about using technology to solve real-world problems.
In this post, I want to share the story behind Swasthya Setu — an AI-powered health monitoring web application I built as my final year project. I'll walk you through the problem I was trying to solve, the tech stack I chose, the challenges I faced, and what I learned along the way.
👉 Live Demo: https://swasthyasetu-tsbt.vercel.app
👉 GitHub: https://github.com/SubhamKhandual007
🏥 What is Swasthya Setu?
Swasthya Setu (meaning "Health Companion" in Hindi/Odia) is an AI-powered health app designed to help patients and healthcare providers monitor health data intelligently.
The core idea was simple: millions of people in India, especially in Tier 2 and Tier 3 cities, don't have easy access to regular health check-ins. What if a web app could act as a smart health companion — tracking vitals, flagging risks, and giving users actionable insights?
Key Features:
- 🤖 AI-based health risk analysis
- 📊 Patient dashboard with real-time health metrics
- 🔒 Secure authentication and user data privacy
- 📱 Fully responsive design for mobile users
- 🏥 Healthcare provider interface
🛠️ Tech Stack
Here's everything I used to build Swasthya Setu:
| Layer | Technology |
|---|---|
| Frontend | React.js, Tailwind CSS, Vite |
| Backend | Node.js, Express.js |
| Database | MongoDB (Atlas) |
| Authentication | JWT, Clerk |
| AI/ML | Python-based prediction model |
| Deployment | Vercel (Frontend), Render (Backend) |
I chose the MERN stack because it allows full JavaScript development across the entire application — from the React frontend to the Node.js backend to MongoDB. This consistency made it much easier to develop and debug as a solo developer.
💡 Why I Built This
During my third year, I noticed how underserved rural healthcare really is in India. A close family experience made me think: what if patients had a smart app that could monitor their health data and alert them before things got serious?
That thought became Swasthya Setu.
I also wanted to challenge myself technically — integrating AI/ML into a MERN app was something I hadn't done before, and I knew it would push my skills significantly.
🚧 Challenges I Faced
1. Integrating Python ML Model with Node.js Backend
The biggest challenge was connecting a Python-based machine learning prediction model with my Node.js backend. I solved this by running the ML model as a microservice and communicating via REST API calls.
2. Handling Real-Time Data
Displaying health metrics in real time on the patient dashboard required careful state management in React. I used useEffect hooks and polling to keep the dashboard updated without overloading the server.
3. Designing for Non-Tech Users
Healthcare users — especially older patients — aren't always tech-savvy. I had to keep the UI extremely simple and intuitive, which meant several redesigns of the dashboard.
4. Deployment Configuration
Getting the frontend (Vercel) and backend (Render) to communicate correctly in production, with proper CORS settings and environment variables, took longer than expected. But it was a valuable DevOps lesson.
📚 What I Learned
Building Swasthya Sathi as a solo MERN Stack Developer taught me:
- Full Stack Architecture — designing a system end-to-end, from database schema to UI
- API Design — creating clean, RESTful APIs with proper error handling
- AI Integration — connecting ML models to web applications
- React State Management — handling complex data flows in a real app
- Deployment & DevOps — managing production environments across multiple platforms
- User-Centric Design — building for real users, not just for developers
🔗 Other Projects I've Built
Alongside Swasthya Sathi, I've been working on several other projects:
- 🩸 Blood Cancer Prediction — ML model using Jupyter Notebook to predict blood cancer risk
- 🌾 Crop Recommendation System — ML-based crop suggestion app for farmers
- 🛡️ Suraksha Setu — Another web app focused on safety (Live Demo)
- 🤖 AI Assistant — A conversational AI assistant built with Python
All projects are open source on my GitHub: github.com/SubhamKhandual007
🚀 What's Next for Swasthya Setu
I'm planning to add:
- Real-time notifications via WebSockets
- Wearable device data integration
- Multilingual support (Odia, Hindi, English)
- Doctor appointment booking module
🤝 Let's Connect
I'm Subham Khandual, a MERN Stack Developer actively looking for opportunities — internships, collaborations, or full-time roles after graduation.
If you're working on something interesting in web development, AI, or healthtech, I'd love to connect!
- 🌐 Portfolio: subhamportfoliodemo.vercel.app
- 💼 LinkedIn: linkedin.com/in/subham-khandual
- 🐙 GitHub: github.com/SubhamKhandual007
- 📧 Email: subhamkhandual215@gmail.com
If you found this post helpful or interesting, drop a ❤️ and follow me for more content on MERN Stack development, AI integration, and real-world project builds!
Tags: #mernstack #react #nodejs #mongodb #webdev #healthtech #ai #javascript #beginners #showdev
Top comments (0)