OMOP

MCP server that maps clinical terminology to OMOP concepts using large language models for term standardization and OMOP vocabulary navigation.

other

GitHub Stars

other

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "ohnlp-omop_mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "<path-to-local-repo>",
        "run",
        "omop_mcp"
      ],
      "env": {
        "MODEL_NAME": "your-model-name",
        "OPENAI_API_KEY": "YOUR_OPENAI_API_KEY",
        "AZURE_API_VERSION": "2024-03-01",
        "AZURE_OPENAI_API_KEY": "YOUR_AZURE_API_KEY",
        "AZURE_OPENAI_ENDPOINT": "https://your-azure-endpoint.cognition.azure.com"
      }
    }
  }
}

The OMOP MCP Server lets you map clinical terminology to OMOP concepts using large language models. It provides a focused, callable service that you can integrate into clinical data workflows to standardize terms, validate mappings, and navigate the OMOP vocabulary with ease.

How to use

You run the MCP server locally and connect to it from your MCP client. Launch the server, point your prompts at the OMOP mapping capabilities, and guide the model with explicit target fields like measurement_concept_id and the source table (e.g., measurement). Prioritize vocabularies if you have preferences (for example, choose a preferred chain like "SNOMED" or a hierarchy such as "LOINC > SNOMED > RxNorm"). Use the provided tooling to map terms, validate mappings, search the vocabulary, and convert between coding systems.

How to install

Prerequisites you need before configuring the MCP server:

Ensure uv is installed on your system.
Clone the MCP project repository and enter the directory.
Prepare API credentials if you plan to call external AI services.

- Clone the repository and set up the environment

git clone https://github.com/OHNLP/omop_mcp.git
cd omop_mcp

- Configure API credentials (optional for API calls)

cp .env.template .env

Edit .env with your API credentials:

# Azure OpenAI Configuration
AZURE_OPENAI_ENDPOINT=
AZURE_OPENAI_API_KEY=
AZURE_API_VERSION=
MODEL_NAME=

# OpenAI Configuration (alternative)
OPENAI_API_KEY=

- Run the MCP server via the standard MCP command

The MCP server runs as a local process via the MCP runtime. Use the runtime to start the server from your cloned repository path.

uv --directory <path-to-local-repo> run omop_mcp

Available tools

find_omop_concept

Maps clinical terminology to OMOP concepts, validates terminology mappings, searches OMOP vocabulary, and converts between coding systems.

validate_mapping

Validates the consistency and correctness of terminology mappings against OMOP concepts.

search_vocabulary

Searches the OMOP vocabulary for concepts and related terms.

convert_between_systems

Converts between different clinical coding systems (e.g., SNOMED, LOINC, RxNorm) within the OMOP framework.

Built by

VeilStrat

AI signals for GTM teams