How does PathML handle large-scale image datasets?

It uses an HDF5-based storage system to efficiently manage image tiles, masks, and metadata, optimized for high-throughput machine learning training and inference.

Which whole-slide image formats are supported by PathML?

PathML supports over 160 proprietary and open-source formats, including Aperio SVS, Hamamatsu NDPI, Leica SCN, Zeiss ZVI, DICOM, and OME-TIFF.

Does this skill support automated stain normalization?

Yes, PathML provides preprocessing transforms for H&E images, including Macenko and Vahadane stain normalization methods.

What machine learning models are integrated with PathML?

PathML features built-in support for specialized pathology models like HoVer-Net for nucleus segmentation and HACTNet for hierarchical cell-type classification.

Can PathML be used for spatial proteomics?

Yes, it includes specialized capabilities for analyzing multiparametric imaging data from platforms like CODEX, Vectra, and MERFISH, including cell segmentation with Mesmer.

PathML Pathology Analysis

Name: PathML Pathology Analysis
Author: BbgnsurfTech

byBbgnsurfTech

•

Ciencia de Datos y ML

Analyzes whole-slide images and multiparametric data using specialized computational pathology workflows and machine learning models.

Acerca de

PathML is a specialized toolkit designed to streamline computational pathology and spatial biology workflows within Claude. It provides a modular framework for processing high-resolution whole-slide images (WSI), supporting over 160 formats like SVS and DICOM. The skill empowers users to build robust preprocessing pipelines, perform automated nucleus detection, and conduct spatial proteomics analysis on multiplexed imaging data. By integrating deep learning models such as HoVer-Net and offering graph-based spatial analysis, PathML transforms raw pathology slides into actionable data for research and clinical machine learning applications.

Características Principales

Spatial graph construction for advanced tissue architecture analysis
Multiparametric imaging support for CODEX, Vectra, and MERFISH data
Deep learning integration for nucleus segmentation and cell classification
Unified WSI loading for 160+ formats including SVS, NDPI, and DICOM
Composable preprocessing pipelines for stain normalization and tissue detection
3 GitHub stars

Casos de Uso

Quantifying spatial proteomics and cell-cell interactions in multiplex imaging
Automating histopathology image analysis and stain normalization workflows
Developing and deploying deep learning models for clinical pathology research

Acerca de

Características Principales

Spatial graph construction for advanced tissue architecture analysis
Multiparametric imaging support for CODEX, Vectra, and MERFISH data
Deep learning integration for nucleus segmentation and cell classification
Unified WSI loading for 160+ formats including SVS, NDPI, and DICOM
Composable preprocessing pipelines for stain normalization and tissue detection
3 GitHub stars

Casos de Uso

Quantifying spatial proteomics and cell-cell interactions in multiplex imaging
Automating histopathology image analysis and stain normalization workflows
Developing and deploying deep learning models for clinical pathology research