🏥 MedNav — AI-Powered Clinical Healthcare Navigator

Bringing transparency, AI precision, and genuine empathy back into the hands of the patient.

---

🚨 The Problem

Navigating the Indian healthcare system is broken for patients:

1. Medical Jargon Maze — Patients experience symptoms but don't know clinical terminologies or required procedures.
2. Zero Price Transparency — Hospital pricing is a black box with hidden fees (diagnostics, contingencies, surgeon fees). Patients have no idea what the fair market rate or the officially regulated government baseline is.
3. Information Overload — Generic 5-star Google ratings mean nothing when choosing a hospital for a life-saving procedure.

---

💡 The Solution

MedNav is an intelligent, end-to-end healthcare concierge that acts as an Intelligent Decision Layer on top of the healthcare infrastructure. By simply typing symptoms in plain English, patients receive an instant, transparent, and actionable clinical pathway.

MedNav is NOT a chatbot. It is a structured clinical and financial engine that cross-references government pricing schedules, real-time geospatial data, and live hospital sentiment to produce a mathematically sound recommendation.

---

✨ Core Features

🧠 AI Clinical Intent Extractor

Users type symptoms in conversational language (e.g., "My dad's knee has been hurting for months"). Our AI triage system translates this into standardized ICD-10 clinical diagnostic codes and maps out the exact required procedure (e.g., Osteoarthritis → Total Knee Replacement).

⚖️ Dual-Track Cost Analysis (Anchored in PMJAY)

MedNav's pricing engine is directly integrated with the official PMJAY (Pradhan Mantri Jan Arogya Yojana) Health Benefit Package 2024 Schedule. We provide a side-by-side breakdown:

• Government Baseline: Official PMJAY-regulated costs
• Private Market Average: Estimated costs adjusted via city-tier multipliers and facility-specific variance

Hidden fees (Room Stay, ICU, Doctor Fees, Pharmacy, Contingency) are mathematically exposed.

📊 Decoding Ratings into Real-World Clinical Facts

Instead of showing a meaningless 4.5-star average, MedNav dynamically scrapes live Google Places reviews and runs NLP sentiment analysis to extract clinical operational realities:

• ✅ "Fast Cardiac Admission"
• ⚠️ "Long OPD Wait Times"

📍 Intelligent Hospital Ranking Engine

Computes a localized "Fit Score" for nearby hospitals by combining:

• Haversine GPS distance tracking
• PMJAY-empanelment verification
• Live sentiment scores
• Budget constraints

🚨 'Red SOS' Emergency Override Protocol

If life-threatening symptoms (like a stroke or cardiac event) are detected, MedNav forces a halt on the standard workflow. It bypasses cost estimates and triggers a high-contrast Red SOS screen directing the patient to immediate emergency care.

---

🏗️ Architecture

┌─────────────────────────────────────────────────────────────┐
│                     React.js Frontend                       │
│            (Tailwind CSS + Framer Motion)                   │
└──────────────────────┬──────────────────────────────────────┘
                       │ REST API
┌──────────────────────▼──────────────────────────────────────┐
│                   FastAPI Backend                            │
│              (Async + High Concurrency)                     │
├─────────────┬───────────────┬───────────────┬───────────────┤
│   Intent    │     Cost      │   Hospital    │   Review      │
│  Extractor  │  Calculator   │   Finder      │  Analyzer     │
│  (Gemini)   │  (PMJAY HBP)  │ (Google Maps) │  (Gemini NLP) │
└─────────────┴───────────────┴───────────────┴───────────────┘
                       │
        ┌──────────────┼──────────────┐
        ▼              ▼              ▼
   Google Gemini   Google Places   PMJAY HBP 2024
   2.0 Flash API      API          Schedule Data

---

🛠️ Tech Stack

Layer	Technology
Frontend	React.js, Tailwind CSS, Framer Motion
Backend	FastAPI (Python), Async I/O
AI Engine	Google Gemini 2.0 Flash API
Geospatial & Reviews	Google Places API
Government Data	PMJAY Health Benefit Package 2024
Concurrency	asyncio.gather with API jitter limits

---

🚀 Getting Started

Prerequisites

• Python 3.11+
• Node.js 18+
• Google Gemini API Key
• Google Places API Key

Backend Setup

cd backend
pip install -r requirements.txt
cp .env.example .env
# Add your API keys to .env
uvicorn main:app --reload

Frontend Setup

cd mednav-ui
npm install
cp .env.example .env
# Set VITE_API_URL in .env
npm run dev

---

📂 Project Structure

├── backend/
│   ├── main.py                  # FastAPI entry point
│   ├── services/
│   │   ├── intent_extractor.py  # Gemini-powered symptom → ICD-10 mapping
│   │   ├── cost_calculator.py   # PMJAY + Private cost engine
│   │   ├── hospital_finder.py   # Google Places + Haversine ranking
│   │   ├── review_analyzer.py   # Live sentiment analysis
│   │   └── ranker.py            # Fit Score computation
│   └── requirements.txt
├── mednav-ui/
│   ├── src/
│   │   ├── components/          # React UI components
│   │   ├── api/                 # API client
│   │   └── App.jsx              # Main application
│   └── package.json
├── data/
│   └── pmjay_hbp_2024/          # Official PMJAY pricing data
└── procedures.json              # Procedure mapping database

---

👥 Team Arjuna

Name	Role
Atharva Joshi	Full-Stack Development, AI Integration
Siddhant Sonawane	Backend & Data Engineering
Swanand Nalawade	Frontend & UI/UX Design

---

🏆 Built at TenzorX Hackathon

This project was built for the TenzorX Hackathon organized by the Poonawalla Group on Unstop.

---

📄 License

This project is licensed under the MIT License — see the LICENSE file for details.

---

Made with ❤️ by Team Arjuna

If this project helped or inspired you, consider giving it a ⭐!