Healthcare Research MCP - Simplified

A researcher-friendly Model Context Protocol (MCP) server for healthcare research. Ask questions in plain English - get comprehensive analysis, code, and visualizations.

🏥 Transform "Find studies about COVID vaccines" into complete research workflows

✨ What Makes This Special?

Just Ask Your Question!

// Instead of learning complex tools...
await mcp.call('analyze', {
  query: "Does vancomycin reduce sepsis mortality?"
});

// Get everything you need:
// ✓ Literature review
// ✓ Statistical analysis code
// ✓ Publication-ready figures
// ✓ Medical code lookups
// ✓ Study design recommendations

🚀 Quick Start (5 minutes)

# Clone the repository
git clone https://github.com/sajor2000/healthcare-research-mcp.git
cd healthcare-research-mcp

# Install dependencies
npm install

# Build the simplified server
npm run build:simplified

# Start the server
npm run start:simplified

That's it! You're ready to start researching.

🎯 Only 5 Simple Tools

1. analyze - Your Research Assistant

Ask anything in plain English:

await mcp.call('analyze', {
  query: "Find all studies about COVID vaccine effectiveness in elderly patients"
});

2. lookup_codes - Medical Code Search

await mcp.call('lookup_codes', {
  search_term: "diabetes"
});

3. generate_code - Statistical Analysis

await mcp.call('generate_code', {
  analysis_description: "compare survival between two groups",
  language: "R"  // or "Python"
});

4. create_visualization - Publication Figures

await mcp.call('create_visualization', {
  description: "forest plot of risk factors"
});

5. check_data - Data Validation

await mcp.call('check_data', {
  data_description: "OMOP patient data"
});

📚 Real-World Examples

Systematic Review

await mcp.call('analyze', {
  query: "Systematic review of ACE inhibitors vs ARBs for heart failure, focus on mortality outcomes"
});

Clinical Trial Analysis

await mcp.call('analyze', {
  query: "Compare 30-day readmission rates between treatment arms in my trial data, adjusting for baseline characteristics"
});

Cohort Study

await mcp.call('analyze', {
  query: "Create a cohort of diabetic patients with HbA1c > 9% and analyze their cardiovascular outcomes"
});

🔧 Configuration

Create a .env file:

# API Keys (optional - for enhanced features)
ANTHROPIC_API_KEY=your_key_here
OPENAI_API_KEY=your_key_here

# Server settings
PORT=3000
LOG_LEVEL=info

📊 What You Can Do

• Literature Reviews: Find and synthesize research papers
• Statistical Analysis: Generate R/Python code for any analysis
• Data Visualization: Create publication-ready figures
• Medical Coding: Look up ICD-10, RxNorm, LOINC codes
• Study Design: Get recommendations for your research
• Data Validation: Check OMOP/CLIF compliance

🏥 Supported Data Standards

• OMOP CDM v5.4: Full support with automatic query generation
• CLIF v2.0: Critical care data format
• Custom CSV/Parquet: Works with your existing data

📖 Documentation

• Researcher's Guide: Complete examples and tutorials
• Visual Overview: See how simple it is

🧪 Try It Now

Example 1: Find Medical Codes

await mcp.call('analyze', {
  query: "What are the ICD-10 codes for different types of diabetes?"
});

Example 2: Statistical Analysis

await mcp.call('analyze', {
  query: "Generate code to compare mortality rates between vaccinated and unvaccinated COVID patients"
});

Example 3: Create a Figure

await mcp.call('analyze', {
  query: "Create a Kaplan-Meier survival curve for my heart failure cohort"
});

🤝 Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

📄 License

MIT License - see LICENSE for details.

🙏 Acknowledgments

This project integrates medical standards including OMOP CDM, STROBE guidelines, and TRIPOD+AI principles. See CITATIONS.md for full attributions.

---

Built for researchers, by researchers. No technical expertise required! 🔬