LiverExplore logo

LiverExplore

Quantitative cellular-level characterization of liver tissue microarchitecture and fibrosis from H&E biopsies for MASH biomarker discovery.

Solution by PathAI
Visit website

Overview

LiverExplore is an AI-powered digital pathology tool developed by PathAI, designed to enhance the discovery of novel biomarkers in metabolic dysfunction-associated steatohepatitis (MASH) and other liver diseases. By delivering quantitative, cellular-level characterization of tissue and fibrosis microarchitecture directly from H&E-stained liver biopsy whole slide images, LiverExplore is purpose-built for research teams working in MASH drug development and liver disease research. The platform is intended for research use only and is not for use in diagnostic procedures.

At its core, LiverExplore generates a panel of more than 1,000 quantitative Human Interpretable Features (HIFs) — metrics that capture areas, counts, densities, and proximities of cell types within each tissue region — enabling rapid, scalable, and standardized spatial analyses of the entire tissue from a single H&E slide, without the need for additional staining or slide scanning.

Key Capabilities

  • AI-Powered HIF Panel: More than 1,000 curated Human Interpretable Features provide comprehensive quantification of liver tissue microarchitecture, supporting both established and novel biomarker discovery.
  • Tissue and Cell Classification: Automated detection and AI-generated heatmaps visualize more than 20 distinct tissue regions and more than 10 cell types across the biopsy.
  • H&E Fibrosis Detection and Subtyping: AI-powered identification of collagen and fibrosis tissue is performed directly on the H&E image, eliminating the need for additional slide scanning or special stains.

Fibrosis Subtypes Detected

  • Portal fibrosis
  • Periportal fibrosis
  • Perivenular fibrosis
  • Perisinusoidal fibrosis
  • Incomplete septal fibrosis
  • Complete septal fibrosis
  • Nodular fibrosis

Anatomical Landmarks Identified

  • Portal tract
  • Central vein
  • Zone 1
  • Zone 2
  • Zone 3

Cell Types and Tissue Regions Classified

  • Immune cells
  • Normal hepatocytes
  • Ballooned hepatocytes
  • Steatotic hepatocytes
  • Bile ducts
  • Blood vessels
  • Portal inflammation
  • Interface hepatitis
  • Lobular inflammation

Research and Drug Development Applications

  • Improve: Quantification of established histologic biomarkers of MASH, such as patterns of fibrosis and counts of ballooned hepatocytes.
  • Identify: Novel histological biomarkers and drivers of therapeutic efficacy to support pipeline decision-making.
  • Generate: Comprehensive supporting evidence for regulatory submission in MASH clinical trials.

Product Specifications

  • Indication: Metabolic dysfunction-associated steatohepatitis (MASH)
  • Intended Use: Research Use Only
  • Stain Type: H&E
  • Compatible Scanners: Leica Aperio® AT2 (additional scanner compatibility may be available upon request)
  • Outputs: A panel (.csv) of more than 1,000 quantitative HIFs, plus heatmap classifications of tissue, cells, fibrosis subtypes, and liver landmarks

LiverExplore may also be compatible with other liver indications and scanner platforms beyond those listed. PathAI's research publications supporting the platform include work on AI-based measurement of NASH histology, analytical and clinical validation studies, and investigations into the histologic implications of novel MASH therapeutics. Teams interested in applying LiverExplore to their research programs are encouraged to contact PathAI's business development team for further information.

Meta

Domain
Digital Pathology & Imaging
Subdomain
Tissue Biomarker Quantification
Software type(s)
Analytical Platform
Deployment type(s)
Cloud / SaaS
Industry vertical(s)
Academic / ResearchBiotechCROPharma
Development stage(s)
ClinicalPreclinical / Pre-MarketResearch & Discovery
Target user(s)
Research ScientistBioinformatician / Computational ScientistClinical / Diagnostic Professional
Compliance standard(s)
ISO 27001ISO 13485GCLP
Tag(s)
Uses AI