Dicom MCP
D

Dicom MCP

dicom-mcp是一个基于DICOM标准的模型上下文协议服务器,为大型语言模型提供查询和交互医学影像元数据的工具,支持患者信息、检查、序列和实例的检索,以及DICOM封装PDF文本提取功能。
2.5分
10.0K

What is dicom-mcp?

dicom-mcp is a specialized server that allows AI systems to securely access and analyze medical imaging data stored in DICOM format. It acts as a bridge between AI assistants and hospital PACS systems, enabling querying of patient records, imaging studies, and clinical reports while maintaining DICOM security standards.

How to use dicom-mcp?

After installation and configuration, you can interact with the server through supported AI platforms (like Claude or Zed) using natural language commands. The server handles the technical DICOM communications while you focus on the medical questions.

Use Cases

Ideal for radiologists needing AI assistance, clinical researchers analyzing imaging databases, or healthcare AI developers building medical imaging applications. Common scenarios include patient record lookup, study comparison, and automated report analysis.

Key Features

Comprehensive DICOM Querying
Search across patient records, imaging studies, series and individual DICOM instances with flexible filtering options
Clinical Report Extraction
Automatically extract text from PDF reports stored in DICOM format for AI analysis
Multiple PACS Support
Configure and switch between different hospital PACS systems with different security requirements
Smart Attribute Presets
Predefined query templates (minimal/standard/extended) simplify accessing the most relevant DICOM metadata
Advantages
Standardized access to medical imaging data through DICOM protocols
No need to learn complex DICOM query syntax - use natural language
Secure connection following hospital PACS security requirements
Lightweight server that integrates with existing AI platforms
Limitations
Requires configuration by IT staff for hospital PACS connections
Limited to metadata and PDF extraction - doesn't process image pixels
Performance depends on PACS server response times
Currently supports only DICOM (not other medical data formats)

Getting Started

Installation
Install the server using pip package manager
Configuration
Create a YAML configuration file with your PACS connection details (see example in Resources)
Launch Server
Start the server pointing to your configuration file
Connect AI Platform
Configure your AI platform (Claude/Zed) to connect to the running server

Example Scenarios

Patient Record Lookup
Find all available information for a patient who may have multiple imaging studies
Report Analysis
Extract and analyze text from radiology reports
Study Comparison
Compare multiple studies for the same patient

Frequently Asked Questions

What DICOM servers are supported?
Is patient data stored by the server?
Can I view actual medical images?
How is patient privacy maintained?

Additional Resources

Official Documentation
Model Context Protocol specification
Sample Configuration
Example YAML configuration file
pynetdicom Library
Underlying DICOM library used by the server
DICOM Standard Overview
Introduction to DICOM medical imaging standard

安装

复制以下命令到你的Client进行配置
"mcpServers": {
  "dicom": {
    "command": "uv",
    "args": ["--directory", "/path/to/dicom-mcp", "run", "dicom-mcp", "/path/to/configuration.yaml"]
  }
}
注意:您的密钥属于敏感信息,请勿与任何人分享。

替代品

V
Vestige
Vestige是一个基于认知科学的AI记忆引擎,通过实现预测误差门控、FSRS-6间隔重复、记忆梦境等29个神经科学模块,为AI提供长期记忆能力。包含3D可视化仪表板和21个MCP工具,完全本地运行,无需云端。
Rust
4.5K
4.5分
M
Moltbrain
MoltBrain是一个为OpenClaw、MoltBook和Claude Code设计的长期记忆层插件,能够自动学习和回忆项目上下文,提供智能搜索、观察记录、分析统计和持久化存储功能。
TypeScript
4.1K
4.5分
B
Bm.md
一个功能丰富的Markdown排版工具,支持多种样式主题和平台适配,提供实时编辑预览、图片导出和API集成能力
TypeScript
4.4K
5分
S
Security Detections MCP
Security Detections MCP 是一个基于Model Context Protocol的服务器,允许LLM查询统一的安全检测规则数据库,涵盖Sigma、Splunk ESCU、Elastic和KQL格式。最新3.0版本升级为自主检测工程平台,可自动从威胁情报中提取TTPs、分析覆盖差距、生成SIEM原生格式检测规则、运行测试并验证。项目包含71+工具、11个预构建工作流提示和知识图谱系统,支持多SIEM平台。
TypeScript
6.0K
4分
P
Paperbanana
PaperBanana是一个自动化生成学术图表和统计图的智能框架,支持从文本描述生成高质量的论文插图,采用多智能体管道和迭代优化,提供CLI、Python API和MCP服务器等多种使用方式。
Python
7.7K
5分
B
Better Icons
一个提供超过20万图标搜索和检索的MCP服务器和CLI工具,支持150多个图标库,帮助AI助手和开发者快速获取和使用图标。
TypeScript
6.7K
4.5分
A
Assistant Ui
assistant-ui是一个开源TypeScript/React库,用于快速构建生产级AI聊天界面,提供可组合的UI组件、流式响应、无障碍访问等功能,支持多种AI后端和模型。
TypeScript
7.3K
5分
A
Apify MCP Server
Apify MCP服务器是一个基于模型上下文协议(MCP)的工具,允许AI助手通过数千个现成的爬虫、抓取器和自动化工具(Apify Actor)从社交媒体、搜索引擎、电商等网站提取数据。它支持OAuth和Skyfire代理支付,可通过HTTPS端点或本地stdio方式集成到Claude、VS Code等MCP客户端中。
TypeScript
7.5K
5分
D
Duckduckgo MCP Server
已认证
DuckDuckGo搜索MCP服务器,为Claude等LLM提供网页搜索和内容抓取服务
Python
85.7K
4.3分
F
Firecrawl MCP Server
Firecrawl MCP Server是一个集成Firecrawl网页抓取能力的模型上下文协议服务器,提供丰富的网页抓取、搜索和内容提取功能。
TypeScript
149.8K
5分
F
Figma Context MCP
Framelink Figma MCP Server是一个为AI编程工具(如Cursor)提供Figma设计数据访问的服务器,通过简化Figma API响应,帮助AI更准确地实现设计到代码的一键转换。
TypeScript
75.8K
4.5分
B
Baidu Map
已认证
百度地图MCP Server是国内首个兼容MCP协议的地图服务,提供地理编码、路线规划等10个标准化API接口,支持Python和Typescript快速接入,赋能智能体实现地图相关功能。
Python
50.1K
4.5分
E
Exa Web Search
已认证
Exa MCP Server是一个为AI助手(如Claude)提供网络搜索功能的服务器,通过Exa AI搜索API实现实时、安全的网络信息获取。
TypeScript
57.3K
5分
C
Context7
Context7 MCP是一个为AI编程助手提供实时、版本特定文档和代码示例的服务,通过Model Context Protocol直接集成到提示中,解决LLM使用过时信息的问题。
TypeScript
106.0K
4.7分
M
Minimax MCP Server
MiniMax Model Context Protocol (MCP) 是一个官方服务器,支持与强大的文本转语音、视频/图像生成API交互,适用于多种客户端工具如Claude Desktop、Cursor等。
Python
63.6K
4.8分
E
Edgeone Pages MCP Server
EdgeOne Pages MCP是一个通过MCP协议快速部署HTML内容到EdgeOne Pages并获取公开URL的服务
TypeScript
32.0K
4.8分
AIBase
智启未来,您的人工智能解决方案智库