Healthcare RAG

The HealthcareRAGTools project is an agentic AI system designed to assist healthcare professionals by leveraging Retrieval-Augmented Generation (RAG) techniques. This project integrates a Model Context Protocol (MCP) server with a Chroma vector database to log patient symptoms, retrieve similar cases, and search medical documents. It supports interactive queries via a terminal-based client and can also be used within the Cursor IDE’s agent chat interface.

This project enables the following key functionalities:
- Log Patient Symptoms: Logs patient symptoms (e.g., "fever and cough") with severity levels (e.g., "Moderate") and retrieves similar cases from a database for comparison.
- Search Documents: Searches a collection of medical documents (e.g., flu or asthma symptom descriptions) to provide relevant information based on user queries. The system uses the Groq API for natural language processing, with the llama-3.3-70b-versatile model. It is built to support real-time healthcare data management and retrieval, making it a valuable tool for clinical decision support.

The project relies on the following tools and technologies:

Python 3.8+: The primary programming language for scripting and server logic. FastMCP: A modular compute platform framework for building and running agentic systems. LangChain: A library for building context-aware language models and agents. LangChain-Groq: Integration with Groq’s API for advanced language model capabilities. Chroma: An open-source vector database for storing and retrieving embeddings of medical documents and patient data. Sentence-Transformers: Used to generate embeddings for text data in the Chroma database. HTTpx: For handling HTTP requests within the system. Python-Dotenv: Manages environment variables, such as the Groq API key. LangChain-Community: Additional community-supported LangChain tools. Requests: For making HTTP requests to external services. UV: A package and virtual environment manager for dependency management. Cursor IDE: The development environment, with plans to enable agent chat functionality.

\Desktop\mcpserver\ragmcp
├── .env # Stores the GROQ_API_KEY environment variable ├── .gitignore # Excludes venv, chroma_db, and .env from version control ├── server.py # Main MCP server script with HealthcareRAGTools logic ├── setup_db.py # Script to initialize the Chroma vector database ├── healthcare_client.py # Terminal-based client for interactive queries ├── healthcare.json # Configuration file for the MCP server ├── requirements.txt # Lists project dependencies ├── documents\ # Directory for sample medical documents │ ├── doc1.txt # Example document: Flu symptoms │ ├── doc2.txt # Example document: Asthma symptoms ├── patient_records.json # JSON file storing patient symptom data ├── chroma_db\ # Directory for the Chroma vector database └── ragmcp\ # Virtual environment directory

To set up the project on your local machine, follow these steps:

Prerequisites Windows 10/11 with Command Prompt. Python 3.8+ installed. UV (Universal Virtual Environment) installed: pip install uv. A Groq API key from console.groq.com. Steps Create and Activate the Virtual Environment: uv venv ragmcp\Scripts\activate Confirm the prompt shows (ragmcp). Install Dependencies: Ensure requirements.txt exists with the listed dependencies: set UV_LINK_MODE=copy uv pip install -r requirements.txt Configure Environment Variables: Create or edit .env with your Groq API key: GROQ_API_KEY=your-groq-api-key Initialize the Database: Run the setup script to populate the Chroma database with sample documents: uv run python setup_db.py Expected output: Vector database initialized with sample medical documents. Running the Project Terminal-Based Client Start the Client: uv run python healthcare_client.py Expected Output: Loading environment variables... Loading config file: C:\Users\sniki\OneDrive\Desktop\mcpserver\ragmcp\healthcare.json Initializing HealthcareRAGTools chat... MCPClient initialized ChatGroq initialized

===== Interactive HealthcareRAGTools Chat ===== Type 'exit' or 'quit' to end the conversation Type 'clear' to clear conversation history Example queries:

- Log symptoms for patient P123: fever and cough, severity Moderate, show similar cases
- Search documents for flu symptoms ==================================

Test a Query: Type: Log symptoms for patient P123: fever and cough, severity Moderate, show similar cases Expected response: Assistant: Symptoms logged for Patient P123: 'fever and cough' (Moderate). Similar cases: None Exit: Type exit or quit.

Copy the json code into File>Preferences>Cursor Settings>MCP/MCP Tools and run the server as follows:

You can use the same queries here as well.

Vectorize MCP server for advanced retrieval, Private Deep Research, Anything-to-Markdown file extraction and text chunking.

Agentic RAG over your own documents on an embedded LanceDB, no database server to run. Hybrid search with reranking, Docling parsing for PDFs and 40+ formats, multimodal retrieval, and answers cited to page numbers and section headings.

Local RAG system for Claude Code with hybrid search (semantic + BM25), cross-encoder reranking, markdown-aware chunking, 9 file formats, file watcher, and 12 MCP tools. Zero external servers. pip install knowledge-rag

A local MCP server implementing Retrieval-Augmented Generation (RAG) with sentence window retrieval and support for multiple file types.

Rust-powered PDF toolkit over MCP: create, read, and analyze PDFs; extract text and entities for RAG; convert to Markdown; split/merge/rotate/reorder pages; manage form fields and annotations; encrypt documents. Runs locally via uvx oxidize-mcp.

Physics-based anomaly detection via MCP — send any data, get anomalies back using wave-equation dynamics. No training pipelines, no model files.

Fulcra Context MCP server for accessing your personal health, workouts, sleep, location, and more, all privately. Built around Context by Fulcra.

A healthcare MCP server for EMRs like Cerner and Epic, providing tools to interact with FHIR data and medical resources.

Read-only access to your Alma nutrition data, meals, goals, scores, workouts, weight history, pantry, and grocery list.

Bioinformatics data for AI agents — gene search, protein structures, clinical variants, PubMed literature, and DNA sequences via NCBI and UniProt. No API key required.