Transforms whole-slide pathology images into structured data via pipelines.
npx skills add https://github.com/jasrajtulsi/GRAD-SCOPE --skill pathml-jasrajtulsi
Please help me install this Agent Skill. Skill: pathml Source: https://github.com/jasrajtulsi/GRAD-SCOPE/tree/main/.claude/skills/pathml Command: npx skills add https://github.com/jasrajtulsi/GRAD-SCOPE --skill pathml-jasrajtulsi
PathML removes the manual complexity of whole-slide pathology analysis by unifying slide loading, preprocessing, segmentation, spatial analysis, and model training in one workflow.
Use PathML to load your pathology slides, build a preprocessing pipeline, and run it on your dataset to generate normalized tiles, masks, features, and downstream analysis outputs.
High-intent search queries and answers about installing and using this skill.
Whole-slide imaging preprocessing involves loading slides, applying stain normalization, and detecting tissue and nuclei to prepare data for cell segmentation. You can build a pipeline to batch process slides and generate structured masks and normalized tiles.
Stain normalization standardizes color variations across pathology slides, while spatial graph construction maps the physical relationships between segmented cells. Together they enable reproducible spatial biology analysis and marker quantification for downstream modeling.
Yes, PyTorch-based inference and training integrations support HoVer-Net and HACTNet workflows, alongside ONNX inference. This allows you to train custom models, evaluate performance, and run inference on whole-slide imaging data.
Slide ingestion supports over 160 whole-slide imaging formats, including brightfield, DICOM, OME-TIFF, CODEX, and Vectra. This broad compatibility allows you to load diverse multiparametric imaging cohorts for unified analysis.
After performing nucleus and cell segmentation, you can construct spatial graphs and quantify markers from multiparametric imaging slides. The workflow exports these structured cell-level features directly to AnnData and HDF5 formats for downstream analysis.
The best way to handle tissue detection and artifact filtering in brightfield slides is through an automated preprocessing pipeline. This includes tissue detection, blur filtering, artifact removal, and mask cleanup to ensure high-quality tiles for model training.
silversteinConvert whole-slide pathology images into tiles, masks, graphs, and quantitative outputs.
KetomihineGuide Scanpy workflows for single-cell analysis on AnnData objects.
LeonChaoXLoad and preprocess whole-slide pathology images into ML-ready datasets.
ovachieverExtract nuclei and cellular features from whole-slide pathology images.
