STAMP

概述

STAMP 是一个端到端的弱监督深度学习管道，专为组织病理学图像分析设计。它提供从全幻灯片图像处理到生物标志物预测的完整工作流，不需要像素级标注。该工具支持多种基础模型，可处理分类、回归和生存分析任务。通过 MCP 支持，STAMP 可作为智能工具集成到 AI 工作流中，用于计算病理学研究。

试试问 AI

装完之后，这里有 7 个你可以让 AI 做的事：

• init

  Create a new STAMP configuration file

  试试:Initialize a new STAMP project configuration试试:Create a config file for my pathology study
• preprocess

  Preprocess whole-slide images into feature vectors

  试试:Preprocess histopathology slides for feature extraction试试:Convert WSI images to processed tiles
• encode_slides

  Encode patch-level features into slide-level embeddings

  试试:Extract slide-level features from patch embeddings试试:Create slide representations using patch features
• encode_patients

  Encode features into patient-level embeddings

  试试:Aggregate slide features into patient-level representations试试:Create patient embeddings from slide data
• train

  Train a Vision Transformer model

  试试:Train a deep learning model on my pathology data试试:Build a biomarker prediction model from slides
• crossval

  Train a Vision Transformer model with cross validation

  试试:Perform cross-validation on my model试试:Validate model performance with multiple folds
• deploy

  Deploy a trained Vision Transformer model

  试试:Deploy my trained model for predictions试试:Make the model available for inference
• statistics

  Generate AUROCs and AUPRCs with 95%CI for a trained model

  试试:Calculate model performance metrics试试:Generate ROC curves for my biomarker predictor
• heatmaps

  Generate heatmaps for a trained model

  试试:Create heatmaps showing model attention regions试试:Visualize which areas of the slide are most important

说明：Tool names extracted from CLI documentation but no MCP-specific tool signatures provided. MCP support is mentioned but not detailed in this README.

可对比工具

安装

# 安装 uv
curl -LsSf https://astral.sh/uv/install.sh | sh

# 克隆并设置 STAMP
git clone https://github.com/KatherLab/STAMP.git
cd STAMP

# GPU 安装
uv sync --extra gpu
source .venv/bin/activate

# CPU 安装
uv sync --extra cpu
source .venv/bin/activate

{
  "mcpServers": {
    "stamp": {
      "command": "uv",
      "args": ["run", "--package", "stamp", "--", "mcp"],
      "env": {
        "PYTHONPATH": "/path/to/STAMP"
      }
    }
  }
}

FAQ

运行 STAMP 需要什么计算资源？

STAMP 可以在 CPU 和 GPU 系统上运行。GPU 安装需要 CUDA 13.0 和兼容的 NVIDIA 驱动程序（版本 580 或更高）。该工具旨在处理大型数据集，并可在配备 SLURM 的 HPC 系统上部署。

STAMP 的 MCP 集成如何工作？

STAMP 包含一个 'mcp/' 模块，能够与模型上下文协议 (MCP) 集成。这使得该工具可以作为 MCP 服务器运行，将其计算病理学功能暴露给 AI 工作流。

STAMP 对比