AarogyaRaksha - AI Layer for Healthcare Affordability

55M Indians fall into poverty from medical bills each year. AarogyaRaksha prevents it - AI matches schemes, audits bills, and cuts medicine costs by up to 90%. Built on India’s health stack for scale.

Comment

Project Story: AarogyaRaksha - AI Layer for Healthcare Affordability

Inspiration

55 million Indians. That's how many families fall into poverty every single year because of medical expenses. I watched it happen to people around me - families selling land, taking loans at 18% interest, children dropping out of school because dad had a heart attack and the hospital bill was ₹18 lakhs.

The crazy part? The government built everything we need - PM-JAY gives ₹5 lakh coverage, Jan Aushadhi has 2,000+ generic medicines at 90% off, CGHS rate cards exist to benchmark fair prices. But nobody knows. The information doesn't reach the people who need it most.

That's when it hit me: we don't have a healthcare access problem in India. We have a healthcare awareness problem.

So I asked myself: what if we built the intelligence layer that sits on top of India's health DPI - the invisible bridge between government infrastructure and ordinary families? Not another telemedicine app. Not another appointment booker. Something that actually protects families from going broke.

AarogyaRaksha was born.

What It Does

AarogyaRaksha is an AI-powered financial shield for Indian families. We solve three massive problems with 5 core modules + 1 intelligence dashboard:

1. Scheme Matcher (AI Scheme Discovery)

Problem: 200+ government health schemes exist but citizens don't know what they qualify for.

What we do:

  • Answer 8 questions (name, age, state, income, category, occupation, medical conditions)
  • Our AI matches you to schemes worth up to ₹5 lakh coverage
  • Shows eligibility confidence, benefits, documents required, application steps
  • Direct links to official government portals

Features:

  • ABHA ID auto-link option for instant profile sync
  • Multi-state scheme database (Central + 15 major states)
  • Document checklist generator
  • Application step-by-step guide

2. Bill Auditor (Hospital Bill Analysis)

Problem: Hospitals markup consumables 1000-1500% - ₹800 for saline that costs ₹50.

What we do:

  • Upload hospital bill (photo/PDF)
  • Vision AI reads every line item
  • Compares against CGHS rate benchmarks
  • Flags overcharging with severity rating

Features:

  • Client-side OCR + Gemini Vision
  • Red/Yellow/Green flagging for each item
  • Consumer rights guidance
  • Potential savings calculation
  • Audit report generation

3. Medicine Optimizer (Generic Drug Finder)

Problem: Branded medicines cost 5-10x more than generic equivalents.

What we do:

  • Search any branded medicine
  • Find PMBJP (Jan Aushadhi) generic alternative
  • Show price comparison with savings %
  • Calculate monthly/yearly savings

Features:

  • 2,000+ medicine database
  • Real-time price comparison
  • Animated savings counter
  • Dosage information
  • Category grouping (diabetes, BP, cholesterol, antibiotics)

Demo searches that work: Telma 40, Lipitor, Augmentin, Metformin

4. Expense Planner (Medical Cost Estimator)

Problem: No tools help families plan for medical treatment costs proactively.

What we do:

  • Select condition/procedure
  • AI estimates total cost (consultation + tests + procedure + medicines + follow-ups)
  • Shows public vs private hospital cost comparison
  • Cross-references government scheme coverage

Features:

  • City-tier cost multipliers (Metro vs Tier-2 vs Tier-3)
  • Public vs private hospital comparison
  • Condition-based risk scoring
  • Scheme coverage gap analysis

5. Emergency Fund Calculator (Health Shield Target)

Problem: Zero financial preparedness for medical emergencies.

What we do:

  • Enter family profile (age, income, family size, pre-existing conditions, insurance status)
  • AI calculates recommended emergency health fund
  • Creates personalized micro-savings plan
  • Shows risk score with breakdown

Features:

  • Age/family/condition/insurance risk scoring
  • 36-month savings plan
  • Breakdown: hospitalization (50%), medicines (25%), diagnostics (15%), contingency (10%)
  • Risk level: Low/Medium/High

6. Analytics Dashboard (Savings Intelligence)

What we do:

  • Track savings over time
  • Expense breakdown visualization
  • Mutual fund automation for health shield

Features:

  • Monthly savings chart (area graph)
  • Expense category pie chart
  • "Automate Savings" button integration
  • Risk score visualization

Summary Table

Module Input Output Savings Potential
Scheme Matcher 8 questions Schemes worth ₹5L Unlock free coverage
Bill Auditor Upload bill photo Flagged overcharges ₹10,000-20,000/bill
Medicine Optimizer Medicine name Generic alternative ₹9,000+/year
Expense Planner Condition selected Cost estimate Plan ahead
Emergency Fund Family profile Fund target Financial preparedness
Analytics Usage data Savings tracking Behavioral change

How We Built It

Tech Stack:

  • Frontend: React 18 + Vite + Tailwind CSS + Framer Motion
  • Backend: Node.js + Express + MongoDB Atlas
  • AI Engine: Google Gemini 2.0 Flash (Vision + Structured Output)
  • OCR: Tesseract.js (client-side, privacy-preserving)

Architecture:

┌─────────────────────────────────────────────────────────────┐
│                    Frontend (React)                        │
│  ┌──────────┐  ┌──────────┐  ┌──────────┐  ┌──────────┐ │
│  │ Scheme   │  │  Bill    │  │ Medicine │  │ Expense  │  │ │
│  │ Matcher  │  │ Auditor  │  │ Optimizer│  │ Planner  │  │ │
│  └────┬─────┘  └────┬─────┘  └────┬─────┘  └────┬─────┘ │
└───────┼─────────────┼─────────────┼─────────────┼────────┘
        │             │             │             │
        ▼             ▼             ▼             ▼
┌─────────────────────────────────────────────────────────────┐
│              Backend API (Express + Gemini AI)              │
│  ┌─────────────────────────────────────────────────────┐  │
│  │  Gemini Service (Structured Output + Vision)        │  │
│  └──────────────────────────────────────────────────────┘  │
│  ┌──────────┐  ┌──────────┐  ┌──────────┐  ┌──────────┐ │
│  │ Schemes  │  │ CGHS     │  │ Medicines│  │ Disease  │  │
│  │ (200+)   │  │ Benchmarks│ │ (2,000+) │  │ Costs    │  │
│  └──────────┘  └──────────┘  └──────────┘  └──────────┘ │
└─────────────────────────────────────────────────────────────┘

Data Curation (The Moat):

  • 200+ government health schemes (Central + 15 major states)
  • 2,100+ Jan Aushadhi medicine prices
  • CGHS rate benchmarks for 1,500+ procedures
  • Disease cost profiles (cancer, diabetes, heart, cataract)

Challenges We Ran Into

Challenge 1: AI Output Consistency Gemini is powerful but sometimes creative with JSON structure. We needed strict schemas for bill analysis results. Solution: used Zod validation + retry logic with fallback prompts.

Challenge 2: Data Gap The schemes database had 2,000+ entries but missing state-specific schemes for tier-2/3 states. Solution: curated additional 50+ schemes manually from government portals and validated with NHA guidelines.

Challenge 3: Client-Side OCR Hospital bills are messy - crumpled paper, handwriting, different formats. Tesseract.js alone wasn't enough. Solution: combined Tesseract extraction with Gemini Vision for superior accuracy.

Challenge 4: The "Missing Middle" Problem PM-JAY targets the poorest. The middle class (₹5-25 lakh income) falls through the cracks - too rich for PM-JAY, too poor for private insurance. Solution: focused on uncovering state-specific schemes and medicine cost optimization as the primary value props for this segment.

Accomplishments That We're Proud Of

Achievement Impact
5 working AI modules Complete end-to-end user flows
Curated 2,000+ scheme database Covers Central + 15 major states
Vision-based bill analysis Real-world overcharging detection
Medicine optimizer live Showing 70-90% savings on common drugs
Hindi/English i18n Accessible to 140 crore Indians
Premium UI design Intercom-inspired trust-building aesthetic
Working demo Deployable TODAY, not just a pitch

The Demo Moments:

  • Search "Lipitor" → See ₹450 vs ₹34 → Watch savings counter animate to ₹9,000+/year
  • Upload hospital bill → AI flags 5 red-flagged items → "Your bill is inflated by ₹15,420"
  • Answer 8 questions → "You're eligible for 4 schemes worth ₹7.5 lakhs"

We've seen similar platforms in the healthcare space - everything solves for access (booking appointments, teleconsultation). We solve for affordability. That's the gap.

What We Learned

Lesson 1: The DPI Opportunity India has built incredible DPI infrastructure (UPI, ONDC, ABDM) but nobody built the citizen-facing intelligence layer on top. That's where the real value is.

Lesson 2: Information is Currency Families don't need more hospitals. They need to know what's already free. Scheme awareness alone can unlock ₹5 lakh coverage. Medicine switching can save ₹9,000/year. Bill auditing can recover ₹15,000+ in markups.

Lesson 3: Boring Wins Price comparison, bill verification, scheme eligibility. But they affect 140 crore Indians daily. The most sustainable startup ideas solve boring structural problems.

What's Next for AarogyaRaksha

Phase 1: Launch (BluePrint 2026)

  • Deploy frontend to Vercel
  • Seed 200+ schemes database
  • Live bill auditor demo to showcase

Phase 2: Expand Intelligence

  • Add ABHA ID integration for auto-eligibility
  • Build disease-specific cost calculators
  • Push notification alerts for scheme deadlines

Phase 3: Network Effects

  • Partner with Jan Aushadhi Kendras for store locator
  • Integrate with NHA for PM-JAY verification
  • Build employer wellness plans (B2B2C)

The Vision: Every Indian family has a financial shield against medical bankruptcy, not through insurance, but through information.

The Mission: Turn India's health DPI into actual protection for 140 crore people.

Built With

Share this project:

Updates