dicom-mcp: A DICOM Model Context Protocol Server

This repository hosts a Model Context Protocol (MCP) server designed for interactions with DICOM (Digital Imaging and Communications in Medicine) systems. It enables Large Language Models (LLMs) to access and analyze medical imaging metadata by querying DICOM servers.

The dicom-mcp server supports querying patient information, studies, series, and instances using standard DICOM networking protocols. It can also extract text from encapsulated PDF documents stored in DICOM format, facilitating the analysis of clinical reports. Built on pynetdicom, it adheres to the Model Context Protocol specification. Key tools include functions for managing DICOM nodes, verifying connections, querying DICOM data, retrieving instances, and extracting PDF text.

The server is installed via pip and configured using a YAML file. It can be run from the command line or integrated with tools like Claude Desktop and Zed. Example queries are provided for common use cases. The project includes debugging tools and a development environment setup guide, along with instructions for running tests using Docker and pytest.