# 🚀 Quick Start Checklist
Follow this checklist to get your Cardiology Knowledge Graph MCP system up and running in under 30 minutes!
## ✅ Prerequisites (5 minutes)
### 1. Install Neo4j Desktop
- [ ] Download from [neo4j.com/download](https://neo4j.com/download/)
- [ ] Install and open Neo4j Desktop
- [ ] Create a new project called "Cardiology KG"
- [ ] Create a new database:
- Name: `cardiology-kg`
- Password: `password` (or your choice)
- [ ] **Start the database** (important!)
### 2. Get OpenAI API Key
- [ ] Visit [platform.openai.com/api-keys](https://platform.openai.com/api-keys)
- [ ] Create a new API key
- [ ] Save it securely (you'll need it in step 3)
## ⚙️ Installation (10 minutes)
### Option A: Automated Installation (Recommended)
```bash
chmod +x install.sh
./install.sh
```
### Option B: Manual Installation
```bash
# 1. Install Python dependencies
pip install -r requirements.txt
# 2. Set up environment
cp env.example .env
# Edit .env with your API key and Neo4j credentials
# 3. Test installation
python3 quick_start.py
```
## 🗄️ Database Setup (5 minutes)
### Optional: Add Sample Data
```bash
python3 setup_neo4j.py
```
This creates basic cardiology entities (heart, atrial fibrillation, medications, etc.)
## 🔧 Claude Desktop Integration (5 minutes)
### 1. Find Your Config File
- **Mac**: `~/Library/Application Support/Claude/claude_desktop_config.json`
- Create the file if it doesn't exist
### 2. Add MCP Server Configuration
```json
{
"mcpServers": {
"cardiology-kg": {
"command": "python",
"args": ["/full/path/to/your/mcp_server.py"],
"env": {
"NEO4J_URI": "bolt://localhost:7687",
"NEO4J_USERNAME": "neo4j",
"NEO4J_PASSWORD": "your-password",
"OPENAI_API_KEY": "your-openai-key"
}
}
}
}
```
**Important**: Replace `/full/path/to/your/mcp_server.py` with the actual path!
### 3. Restart Claude Desktop
- Completely quit Claude Desktop
- Restart it
- Look for the 🔨 hammer icon
## 🧪 Test Everything (5 minutes)
### 1. Verify MCP Connection
- Look for the 🔨 hammer icon in Claude Desktop
- If you don't see it, check Claude Desktop logs (Help → Show Logs)
### 2. Try Basic Commands
In Claude Desktop, try:
**Test 1: Simple ingestion**
```
Ingest this cardiology note: "Patient has atrial fibrillation. Started on metoprolol for rate control."
```
**Test 2: Add to graph**
```
Add this to my knowledge graph: [paste the JSON from step 1]
```
**Test 3: Query the graph**
```
What medications treat atrial fibrillation?
```
**Test 4: Get statistics**
```
Show me statistics about my cardiology knowledge graph
```
## 🎯 Success Indicators
You know everything is working when:
- [ ] 🔨 Hammer icon appears in Claude Desktop
- [ ] Document ingestion returns structured JSON
- [ ] Graph queries return relevant results
- [ ] Graph statistics show your data
## 🚨 Troubleshooting
### Common Issues
**No hammer icon in Claude**
- Check the full path in your config file
- Verify environment variables are set correctly
- Restart Claude Desktop completely
**"Connection refused" error**
- Make sure Neo4j database is **started** in Neo4j Desktop
- Check database credentials in your `.env` file
**"API key invalid"**
- Verify your OpenAI API key
- Check you have sufficient credits/quota
**Import errors**
- Run `pip install -r requirements.txt` again
- Check you're using Python 3.8+
### Get Help
- Run `python3 quick_start.py` for detailed diagnostics
- Check the full `README.md` for detailed documentation
- Look at sample files: `sample_cardiology_text.txt`
## 📚 Next Steps
Once everything is working:
1. **Try with real documents**: Ingest PDF textbooks or clinical notes
2. **Expand the ontology**: Edit extraction prompts for specific domains
3. **Use alternative LLMs**: Try BioGPT or local models
4. **Build specialized tools**: Add quiz generation or clinical decision support
## 🫀 Example Workflow
1. **Start with sample text**:
```
Ingest this file: sample_cardiology_text.txt
```
2. **Review and edit the extraction**
3. **Add to your knowledge graph**
4. **Ask questions**:
- "How does atrial fibrillation cause heart failure?"
- "What medications are used for heart failure?"
- "Show me the relationship between hypertension and left ventricular hypertrophy"
5. **Build your cardiology knowledge base** by adding more documents!
---
**Total time**: ~25 minutes
**Result**: A working AI-powered cardiology knowledge graph system! 🎉
