pathml

Analyze whole-slide images with nucleus segmentation and stain normalization.

Updated Mar 10, 2026
One-click install
npx skills add https://github.com/Yezez9/Research-Agent --skill pathml-yezez9
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pathml
Source: https://github.com/Yezez9/Research-Agent/tree/main/scientific-skills/pathml
Command: npx skills add https://github.com/Yezez9/Research-Agent --skill pathml-yezez9

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a comprehensive toolkit for advanced whole-slide image analysis in computational pathology, enabling complex machine learning workflows and detailed spatial analysis.

Core Features & Use Cases

  • Image Loading: Supports 160+ WSI formats (Aperio, Hamamatsu, Leica, Zeiss, DICOM, OME-TIFF).
  • Preprocessing: Stain normalization, nucleus segmentation, artifact detection.
  • Machine Learning: Integrated models (HoVer-Net) for nucleus detection and classification.
  • Spatial Analysis: Graph construction for cell-cell interactions and neighborhood analysis.
  • Multiparametric Imaging: Specialized support for CODEX, Vectra, MERFISH.
  • Use Case: Analyze multiplex immunofluorescence images to identify spatial relationships between tumor cells and immune cells, quantifying marker expression and cell type interactions within tumor microenvironments.

Quick Start

Use the pathml skill to load a whole-slide image and perform nucleus segmentation.

Frequently Asked Questions about pathml

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I perform nucleus segmentation on whole-slide images for computational pathology?

Nucleus segmentation on whole-slide images is performed using integrated machine learning models like HoVer-Net. This computational pathology toolkit enables automatic detection and classification of nuclei within H&E stained tissue images.

What is the best way to analyze spatial relationships in multiplexed immunofluorescence images?

Analyzing spatial relationships in multiplexed immunofluorescence images is done by constructing spatial graphs of cell-cell interactions. This allows quantification of marker expression and neighborhood analysis within the tumor microenvironment.

Does this computational pathology toolkit support DICOM and OME-TIFF slide formats?

Yes, the computational pathology toolkit supports DICOM and OME-TIFF formats. It can load over 160 different whole-slide image formats, including Aperio, Hamamatsu, Leica, and Zeiss files.

Can I use DeepCell Mesmer for cell segmentation on H&E stained tissue images?

Yes, you can use DeepCell Mesmer for cell segmentation on H&E stained tissue images. The toolkit integrates directly with DeepCell Mesmer to provide advanced cell segmentation capabilities.

How do I apply stain normalization to whole-slide images before machine learning training?

Stain normalization is applied to whole-slide images during the preprocessing phase. This standardizes stain appearance before training machine learning models and includes artifact detection to clean the data.

Does the spatial omics analysis support CODEX, Vectra, and MERFISH multiparametric imaging?

Yes, the spatial omics analysis supports CODEX, Vectra, and MERFISH multiparametric imaging. It provides specialized preprocessing and graph construction for these specific multiplexed imaging platforms.