DICOM
Enables AI assistants to query, read, download, and move medical imaging data on DICOM servers (PACS, VNA) including patient searches, study retrieval, PDF report extraction, and image transfer to AI endpoints for analysis.
What is DICOM?
DICOM is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables ai assistants to query, read, download, and move medical imaging data on dicom servers (pacs, vna) including patient searches, study retrieval, pdf report extraction, and image transfer to ai ...
Enables AI assistants to query, read, download, and move medical imaging data on DICOM servers (PACS, VNA) including patient searches, study retrieval, PDF report extraction, and image transfer to AI endpoints for analysis.
This server falls under the Business Applications category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Enables AI assistants to query, read, download, and move med
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx dicom-mcp-serverConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use DICOM
The DICOM MCP Server (dicom-mcp) enables AI assistants to query, read, and move medical imaging data on PACS (Picture Archiving and Communication System) and VNA servers using standard DICOM network protocols. It supports patient and study searches, series and instance queries, PDF clinical report text extraction from DICOM objects, and C-MOVE transfers to send images to AI analysis endpoints — making it possible to ask an AI assistant questions like 'what were the findings in John Doe's last CT report?' and get answers pulled directly from a DICOM server.
Prerequisites
- Python 3.12 or later
- uv or pip for installation
- Access to a DICOM server (PACS or VNA) with known host, port, and AE title — or a local Orthanc server via Docker for testing
- A YAML configuration file defining your DICOM nodes and calling AE title
- An MCP client such as Claude Desktop, Cursor, or Cline
Install dicom-mcp
Install the package using uv tool install (recommended) or pip. uv manages the Python environment automatically.
uv tool install dicom-mcp
# or with pip:
pip install dicom-mcpCreate a DICOM node configuration file
Create a YAML config file defining your DICOM server nodes. The calling_aet is the AE title this client will use when making DICOM network calls. Replace the host, port, and ae_title with your PACS or VNA details.
# config.yaml
nodes:
main:
host: "localhost"
port: 4242
ae_title: "ORTHANC"
description: "Local Orthanc DICOM server"
current_node: "main"
calling_aet: "MCPSCU"(Optional) Start a local Orthanc test server
If you don't have a PACS available, spin up a local Orthanc DICOM server using Docker for testing. The web UI is available at http://localhost:8042.
# From the tests/ directory of the cloned repo:
docker-compose up -dConfigure your MCP client
Add dicom-mcp to your MCP client config, passing the absolute path to your YAML config file as the first argument.
{
"mcpServers": {
"dicom": {
"command": "uvx",
"args": ["dicom-mcp", "/absolute/path/to/config.yaml"]
}
}
}Verify the DICOM connection
After connecting your MCP client, ask the AI assistant to verify the DICOM connection using the verify_connection tool. This performs a C-ECHO to confirm network connectivity to your DICOM node.
DICOM Examples
Client configuration
Claude Desktop configuration for dicom-mcp using uvx and a local config file.
{
"mcpServers": {
"dicom": {
"command": "uvx",
"args": ["dicom-mcp", "/Users/yourname/dicom-config.yaml"]
}
}
}Prompts to try
Ask your AI assistant questions about medical imaging data stored on your DICOM server.
- "Search for all studies for patient John Doe and list them with dates and modalities"
- "Find the most recent CT study for patient ID 12345 and extract any PDF report text"
- "List all series in study UID 1.2.840.10008.5.1.4.1.1.2 and show their modality and description"
- "Send the CT series from yesterday's study to the AI segmentation node configured as AI_NODE"
- "Verify the DICOM connection to the main node and show which nodes are configured"Troubleshooting DICOM
Connection refused or C-ECHO fails
Check that the host, port, and ae_title in your config.yaml match your PACS/VNA exactly. DICOM AE titles are case-sensitive. Test network connectivity first: nc -zv <host> <port>. Ensure your calling_aet is registered as an allowed SCU on the PACS side.
No results returned from query_patients or query_studies
DICOM query matching is exact or wildcard-based. Use a wildcard suffix for partial name matches (e.g. 'DOE*' instead of 'DOE'). Also check that the query level (PATIENT, STUDY, SERIES) matches what your PACS supports. Use get_attribute_presets to understand what attributes are available.
move_series fails or destination node is not found
C-MOVE requires the destination AE title to be registered on the PACS as a known storage destination. Add the destination node to your config.yaml nodes section and ensure the PACS's AE title list includes the destination AE with a routable IP and port.
Frequently Asked Questions about DICOM
What is DICOM?
DICOM is a Model Context Protocol (MCP) server that enables ai assistants to query, read, download, and move medical imaging data on dicom servers (pacs, vna) including patient searches, study retrieval, pdf report extraction, and image transfer to ai endpoints for analysis. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install DICOM?
Follow the installation instructions on the DICOM GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with DICOM?
DICOM works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is DICOM free to use?
Yes, DICOM is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.
DICOM Alternatives — Similar Business Applications Servers
Looking for alternatives to DICOM? Here are other popular business applications servers you can use with Claude, Cursor, and VS Code.
n8n
★ 189.1kA comprehensive MCP server that provides full control over n8n automation workflows through natural language. It offers 43 tools for managing workflows, executions, credentials, and data tables, with safety features like write-mode protection and dou
LobeHub
★ 77.5k🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
Jeecgboot
★ 46.4kAI 低代码平台，「低代码 + 零代码」双模式驱动：低代码一键生成前后端代码，零代码 5 分钟搭建系统，AI Skills 一句话画流程、设计表单、生成整套系统。内置 AI聊天、知识库、流程编排、MCP插件等，兼容主流大模型。引领「AI 生成 → 在线配置 → 代码生成 → 手工合并->AI修改」开发模式，消除 Java 项目 80% 的重复工作，提效而不失灵活。
CowAgent
★ 44.7kCowAgent (chatgpt-on-wechat) 是基于大模型的超级AI助理，能主动思考和任务规划、访问操作系统和外部资源、创造和执行Skills、通过长期记忆和知识库不断成长，比OpenClaw更轻量和便捷。同时支持微信、飞书、钉钉、企微、QQ、公众号、网页等接入，可选择DeepSeek/OpenAI/Claude/Gemini/ MiniMax/Qwen/GLM/LinkAI，能处理文本、语音、图片和文件，可快速搭建个人AI助理和企业数字员工。
Minds Platform
★ 39.2kPlatform dedicated to building an open foundation for applied Artificial Intelligence, designed for people seeking production-ready AI systems they can truly control, extend and deploy anywhere.
Astrbot
★ 32.8kAI Agent Assistant & development framework that integrates lots of IM platforms, LLMs, plugins and AI feature, and can be your openclaw alternative. ✨
Browse More Business Applications MCP Servers
Explore all business applications servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up DICOM in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use DICOM?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.