<<<<<<< HEAD

MEDICO - Medical Diagnosis System

A comprehensive medical diagnosis system with AI-powered symptom analysis.

Project Structure

MEDICO4/
├── frontend_service/     # Frontend application
├── backend_service/      # Backend API service
├── ai_service/           # AI/NLP service
├── data_service/         # Database service
├── tests/                # Test suite
├── docs/                 # Documentation
├── start-app.bat         # Windows startup script
└── start-app.sh          # Unix/Linux startup script

Features Implemented

Phase 1 - System Foundation

1. Project Environment Setup

   • Complete folder structure
   • Dependency management with requirements.txt
   • Environment configuration files

2. Authentication & Role Management

   • User registration and login endpoints
   • JWT token generation and verification
   • Role-based routing (Patient/Doctor)
   • Google OAuth2 login integration

3. Patient Dashboard

   • Chat interface for symptom description
   • Mock diagnosis responses
   • Session management

4. API Routing Framework

   • RESTful endpoints for authentication
   • User profile management
   • Diagnosis service (stubbed)
   • Health check endpoint

5. Service Interconnection Blueprint

   • Defined interfaces for future AI integration
   • Consistent response formats

Phase 2 - AI Consultation Workflow

1. Disease Database Integration

   • Dynamic JSON loading without system restart
   • Hot reload watcher for real-time updates
   • Symptom-to-disease mapping cache

2. NLP Symptom Extraction

   • Text cleaning and normalization
   • Symptom pattern matching with synonyms
   • Fallback mode for offline operation

3. Diagnosis Inference Engine

   • Symptom-disease matching using Jaccard similarity
   • Confidence scoring based on symptom overlap
   • Top-3 disease predictions with medications and tests

4. AI Chatbot Connection

   • Real-time symptom analysis
   • Formatted diagnosis results
   • User-friendly chat interface

Phase 3 - Architecture Restructure (Latest)

1. AI-Centric Architecture

   • AI as the primary conversation orchestrator
   • Backend as a context relay layer only
   • NLP as a data extraction service only
   • Clean separation of concerns

2. Enhanced Context Management

   • Complete session context sent to AI for every message
   • Improved context persistence and retrieval
   • Elimination of redundant questions

3. Dynamic Conversation Flow

   • AI generates follow-up questions dynamically
   • Context-aware responses without hardcoded branches
   • Natural, fluid conversation experience

Google OAuth2 Integration

The Medico app now supports Google login using OAuth2 authentication. This allows users to sign in with their Google accounts.

Backend Implementation

• Added new endpoint: POST /auth/google
• Uses Google's id_token.verify_oauth2_token() for token verification
• Automatically creates new users or fetches existing ones
• Returns a JWT token for Medico's internal auth system

Frontend Implementation

• Added Google Sign-In button to the login page
• Uses Google Identity Services for authentication
• Handles token exchange with the backend

Setup Instructions

1. Obtain Google OAuth2 Credentials:

   • Go to the Google Cloud Console
   • Create a new project or select an existing one
   • Navigate to "APIs & Services" > "Credentials"
   • Click "Create Credentials" > "OAuth client ID"
   • Select "Web application" as the application type
   • Add authorized JavaScript origins:
     • http://localhost:3000 (for local development)
   • Add authorized redirect URIs (if needed):
     • http://localhost:3000/login
   • Copy the generated Client ID

2. Configure Backend:

   • Open backend_service/.env
   • Replace YOUR_GOOGLE_CLIENT_ID_HERE with your actual Google Client ID:

     GOOGLE_CLIENT_ID=your_actual_google_client_id_here
     

3. Configure Frontend:

   • Open frontend_service/app.js
   • Replace YOUR_GOOGLE_CLIENT_ID_HERE with your actual Google Client ID:

     const clientId = "your_actual_google_client_id_here";

4. Install Dependencies:

   cd backend_service
   pip install google-auth google-auth-oauthlib google-auth-httplib2

5. Run the Application:

   # Start backend
   cd backend_service
   python run.py
   
   # Start frontend (in a new terminal)
   cd frontend_service
   python -m http.server 3000

6. Test the Integration:

   • Visit http://localhost:3000
   • Click on the "Login" link
   • Click the "Sign in with Google" button
   • Complete the Google authentication flow

Getting Started

Prerequisites

• Python 3.8+
• pip package manager

Installation

1. Clone the repository
2. Navigate to the project directory
3. Run the startup script:

Windows:

start-app.bat

Unix/Linux/Mac:

./start-app.sh

Manual Startup

Backend Service:

cd backend_service
pip install -r requirements.txt
python run.py

Frontend Service:

cd frontend_service
python -m http.server 3000

API Endpoints

• POST /auth/register - User registration
• POST /auth/login - User login
• POST /auth/google - Google OAuth2 login
• GET /api/user/profile - Get user profile
• POST /api/diagnosis - Get diagnosis
• POST /api/diagnosis/symptoms - Extract symptoms
• POST /api/data/reload - Reload disease data
• GET /api/data/stats - Get disease statistics
• GET /api/system/health - System health check

Technologies Used

• Backend: FastAPI, Python, SQLite
• Frontend: HTML, CSS, JavaScript
• Authentication: JWT, Google OAuth2
• Database: SQLite (will migrate to PostgreSQL later)
• AI Services: Custom NLP implementation with GPT4All integration
• Data Management: JSON-based disease database with hot reload

Documentation

• AI Workflow Summary
• API Reference
• Disease Database Guide
• Foundation Validation
• Architecture Restructure Summary
• New Architecture Diagram

Future Enhancements

• Integration with advanced AI/NLP models
• PostgreSQL database implementation
• Doctor dashboard and features
• Advanced medical record management
• Teleconsultation features =======

MEDCO

AI POWERED MEDICAL DIAGONIS APP

72dd9e799f19eceb81482d08c582173178d46436