Pathml

If you're doing serious computational pathology work, this is the full toolkit. It handles 160+ slide formats, builds preprocessing pipelines for H&E stain normalization, runs nucleus segmentation models like HoVer-Net, and constructs spatial graphs for tissue analysis. The real power is in the multiparametric imaging support for CODEX and Vectra, where you can segment cells with Mesmer and quantify marker expression. Fair warning, this is GPL-2.0 licensed and heavyweight. If you just need to extract tiles from H&E slides, the docs suggest histolab instead. But for ML model training on pathology data or spatial proteomics analysis, this covers the complete workflow from loading proprietary formats to exporting AnnData for downstream analysis.

skills.sh installs: 2K
GitHub stars: 56.6K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add k-dense-ai/scientific-agent-skills --skill pathml --agent claude-code

Installs into .claude/skills of the current project.

Files

PathML

Scope and safety boundary

Use PathML for local computational pathology research. It is beta research software, not a validated medical device, diagnostic system, clinical decision support tool, or substitute for a pathologist. Do not use outputs to diagnose, grade, stage, or treat a patient.

Pathology files may contain faces, labels, accession numbers, patient identifiers, DICOM tags, filenames, or linked clinical data. Before processing:

Confirm authorization, consent/waiver, data-use terms, and institutional policy.
De-identify pixels and metadata; keep the re-identification key outside the analysis workspace.
Use pseudonymous patient_id, slide_id, and specimen_id values. Do not put direct identifiers in filenames, logs, .h5path labels, model cards, or reports.
Keep inputs, intermediates, and outputs on approved local encrypted storage.
Split by patient (then slide) before tiling or fitting any preprocessing step.