Skill UI

PathML is a Python toolkit for computational pathology workflows, loading 160+ slide formats, preprocessing H&E images, constructing spatial graphs, training HoVer-Net/HACTNet models, and analyzing multiplex imaging for ML-ready pathology data.

Python library for digital pathology whole slide images: tissue detection, tile extraction, filters, and stain normalization. Ideal for preparing H&E datasets for deep learning and quick tile-based analysis.

Build local, research-only computational pathology workflows with PathML: load and tile whole-slide images, run preprocessing and QC pipelines, manage h5path data, quantify multiplex images, construct spatial graphs, and plan bounded model inference.

This structured tutor is designed for medical and health sciences students, covering core areas like anatomy, physiology, pharmacology, and pathology. It moves beyond rote memorization by focusing on case-based learning, systems integration, and understanding underlying mechanisms. By practicing clinical reasoning and differential diagnosis, users can bridge the gap between basic science knowledge and complex clinical application, preparing them for major medical licensing exams.

Query and download public cancer imaging data (CT, MR, PET, pathology) from the NCI Imaging Data Commons. Supports metadata SQL lookups, DICOM access via idc-index, REST or MCP, cohort selection, browser visualization and license checks without authentication.

DeepSpot-M is a multimodal foundation model that generates virtual spatial transcriptomics from H&E histology images. It maps 224x224 tiles to spatial gene expression using deep learning, supporting over 19k protein-coding genes. Ideal for research in pathology and cancer genomics, requiring gated model access and Python environment.