Skip to content

DEMO.md

Latest commit

 

History

History
223 lines (171 loc) · 5.87 KB

DEMO.md

File metadata and controls

223 lines (171 loc) · 5.87 KB

Healthcare Navigator Demo Guide

Quick Start Demo

Visit Demo for more information

https://www.loom.com/share/727548249bb940b6896b94da8b88fe17?sid=9013817d-c58a-4dba-ae98-c546cf7005c3

1. Setup (2 minutes)

# Clone and setup
git clone <repository>
cd intelligent-healthcare-navigator

# Install dependencies
pip install -r requirements.txt

# Configure environment
cp .env.template .env
# Edit .env with your API keys

2. CLI Demo (3 minutes)

# Start CLI
python cli.py

# Try these commands:
ask "What is diabetes?"
drug "aspirin"
symptom "headache and fever"
upload "sample_report.pdf"

3. Web Interface Demo (5 minutes)

# Start web app
streamlit run web_app.py

Demo Flow:

  1. Medical Term Lookup: "What is hypertension?"
  2. Drug Safety Check: "Tell me about ibuprofen safety"
  3. Symptom Analysis: "I have chest pain and shortness of breath"
  4. Document Upload: Upload a medical report
  5. User Preferences: Set age, allergies, medical history

Key Features to Demonstrate

🩺 Medical Term Explanations

  • Input: "What is atrial fibrillation?"
  • Shows: WHO ICD data + AI simplification
  • Highlights: Technical vs. simplified explanations

💊 Drug Information & Safety

  • Input: "Tell me about warfarin"
  • Shows: FDA recalls, adverse events, safety profile
  • Highlights: Allergy warnings, safety assessments

🔍 Symptom Analysis

  • Input: "I have severe chest pain and difficulty breathing"
  • Shows: Urgency detection, care recommendations
  • Highlights: Emergency detection, medical disclaimers

📄 Document Processing

  • Input: Upload medical report PDF
  • Shows: Summary, key findings, entity extraction
  • Highlights: Medical entity recognition

🧠 Context Awareness

  • Shows: Conversation memory, user preferences
  • Highlights: Personalized responses based on age/allergies

Demo Script (10 minutes)

Opening (1 min)

"Healthcare Navigator is an AI-powered system that helps users understand medical information safely and accurately."

Core Capabilities (6 mins)

1. Medical Knowledge (2 mins)

CLI: ask "What is Type 2 diabetes?"
  • Shows WHO ICD integration
  • Demonstrates AI simplification
  • Highlights medical disclaimers

2. Drug Safety (2 mins)

Web: Search "aspirin" with allergy to "NSAIDs"
  • Shows FDA data integration
  • Demonstrates allergy warnings
  • Highlights safety assessments

3. Symptom Intelligence (2 mins)

CLI: symptom "severe chest pain, shortness of breath, sweating"
  • Shows urgency detection
  • Demonstrates emergency recommendations
  • Highlights responsible AI responses

Advanced Features (2 mins)

4. Document Analysis

Web: Upload medical report
  • Shows document processing
  • Demonstrates entity extraction
  • Highlights key findings

Closing (1 min)

"Healthcare Navigator combines multiple medical APIs with responsible AI to provide accurate, safe healthcare information."

Technical Highlights

Architecture

  • ReAct Pattern: Reasoning → Acting → Observation
  • Multi-API Integration: WHO ICD, OpenFDA, Google Gemini
  • Safety-First Design: Disclaimers, validation, error handling

APIs Demonstrated

  • WHO ICD-11: Medical term definitions
  • OpenFDA: Drug recalls, adverse events
  • Google Gemini: AI reasoning and simplification

Safety Features

  • Input sanitization and validation
  • Medical disclaimers on all responses
  • Emergency detection and warnings
  • Allergy checking and alerts

Sample Queries for Demo

Medical Terms

  • "What is myocardial infarction?"
  • "Explain rheumatoid arthritis"
  • "What does hypertension mean?"

Drug Information

  • "Is acetaminophen safe during pregnancy?"
  • "Tell me about metformin side effects"
  • "What recalls exist for insulin?"

Symptoms (Use Carefully)

  • "I have a persistent cough and fever"
  • "What could cause severe headaches?"
  • "I'm experiencing dizziness and nausea"

Emergency Examples (Shows Urgency Detection)

  • "I have severe chest pain and can't breathe"
  • "Someone is unconscious and not responding"
  • "I think I'm having a stroke"

Demo Environment Setup

Required API Keys

# .env file
GEMINI_API_KEY=your_gemini_key
WHO_ICD_CLIENT_ID=your_who_client_id
WHO_ICD_CLIENT_SECRET=your_who_secret
OPENFDA_API_KEY=your_fda_key  # Optional but recommended

Sample Files for Upload

  • Create sample medical reports (PDF/TXT)
  • Include various medical entities
  • Test different document types

Troubleshooting Demo Issues

Common Issues

  1. API Rate Limits: Use demo keys with higher limits
  2. Slow Responses: Pre-warm APIs before demo
  3. Network Issues: Have offline fallback examples

Fallback Strategies

  • Use cached responses for key demos
  • Have screenshots ready
  • Prepare offline CLI examples

Demo Metrics to Highlight

Performance

  • Query processing: ~2-5 seconds
  • Document analysis: ~5-10 seconds
  • Multi-API coordination: Seamless

Accuracy

  • WHO ICD medical definitions
  • Real-time FDA safety data
  • AI-enhanced explanations

Safety

  • 100% response disclaimer coverage
  • Emergency detection capability
  • Input validation and sanitization

Post-Demo Q&A Preparation

Technical Questions

  • "How do you ensure medical accuracy?" → WHO ICD + FDA APIs
  • "What about liability?" → Disclaimers + educational purpose
  • "Can it diagnose?" → No, information only + professional referral

Business Questions

  • "Who is the target user?" → Patients, caregivers, health educators
  • "What's the competitive advantage?" → Multi-API integration + safety focus
  • "How does it scale?" → Cloud-ready architecture + caching

Future Roadmap

  • Additional medical APIs
  • Multilingual support
  • Healthcare provider integration
  • Mobile application