Swasthya: AI-Driven Wellness Monitoring Platform

Overview

Swasthya is an AI-driven health monitoring platform designed to detect early signs of fatigue, stress, and burnout using non-invasive ambient sensing and voice analytics.

Unlike traditional health systems that rely on wearables or intrusive sensors, Swasthya uses environmental signals, short voice interactions, and edge AI to generate a real-time Wellness Index (0–100) — all while keeping user data private and locally processed.

What Users Can Do

• View Real-Time Wellness Index: Instantly understand fatigue and stress levels.
• Monitor Sensor Data: Track heart rate, temperature, ambient light, and sound levels.
• AI Breathing Exercises: Receive personalized breathing routines based on wellness score.
• Hydration Tracking: Monitor daily water intake with smart email reminders.
• AI Wellness Companion: Talk to an AI friend over phone calls for emotional support.
• Stress Detection: Analyze voice tone and facial expressions for mood and fatigue.
• Therapy Scheduling: Discover nearby mental health clinics and book appointments.

Why We Built This

Early fatigue and stress are often ignored until they turn serious, especially in:

• Workplaces
• Classrooms
• Rural or resource-constrained healthcare settings

Wearables aren’t always practical, affordable, or comfortable.

Swasthya was built to explore how AI + ambient sensing can act as a silent health companion, providing early warnings without invading privacy or requiring constant user effort.

Tech Stack

Frontend

• React + TypeScript – Interactive dashboard
• Tailwind CSS – Clean, accessible UI
• Recharts – Sensor data visualization
• Google Maps API – Therapy clinic discovery
• Clerk – Authentication

Backend

• Node.js + Express – API layer
• Prisma ORM – Database access
• PostgreSQL – Persistent storage
• Nodemailer – Email reminders

ML & AI

• Python ML Backend – Sensor & AI processing
• Lightweight AI Models – Voice & mood analysis
• Google Gemini AI – Personalized breathing exercises
• Vapi.ai – AI-powered phone calls
• Edge AI – Local inference without cloud dependency

Hardware

• Heart Rate Sensor
• Body Temperature Sensor
• Ambient Light Sensor
• Sound Sensor
  (All connected via WiFi)

Key Technical Highlights

• Non-invasive health monitoring system
• Edge AI inference for privacy-first processing
• Multi-modal data fusion (sensor + voice + environment)
• Wellness Index generation (0–100 scale)
• Real-time dashboard with historical trends
• Hardware → Python → Node.js → React data pipeline

After Completion & Impact

• Built a working end-to-end prototype integrating hardware, AI, and frontend
• Demonstrated edge AI feasibility for healthcare use cases
• Designed a privacy-first health monitoring architecture
• Showcased at a Global Hackathon
• Received strong feedback for innovation and real-world applicability

Future Plans

• Improve fatigue detection accuracy with more signals
• Add long-term wellness trend analytics
• Deploy to low-resource healthcare environments
• Expand language and accessibility support
• Explore mobile-first edge deployment