PantelisGeorgiadis/dicomweb-mcp-server

What it does

The dicomweb-mcp-server is a Model Context Protocol (MCP) server that exposes DICOMweb-compliant DICOM archives to AI assistants. It enables seamless integration between AI tools and medical imaging systems by providing standardized access to DICOM data (medical imaging format used in healthcare). This allows AI assistants to search, retrieve, and analyze medical imaging studies, series, and instances without direct database access.

How to set up

Installation requires cloning the repository from GitHub and configuring the server with your DICOMweb-compliant DICOM archive endpoint. The server can be deployed locally or in cloud environments. Configuration involves setting authentication credentials and specifying the target DICOM archive URL. Once running, connect your MCP-compatible AI assistant (Claude, ChatGPT with MCP support) to the server endpoint.

Key features

• DICOMweb protocol compliance for standardized medical data access
• Study, series, and instance search capabilities
• Metadata inspection and retrieval
• Structured report (SR) reading functionality
• Cloud and on-premises deployment options
• AI assistant integration via MCP protocol
• DICOM instance encapsulation support