Back to datasets
Dataset assetOpen Source CommunityImage ClassificationColorectal Cancer

1aurent/Kather-texture-2016

The dataset represents a texture collection of human colorectal cancer histology images. It contains 5,000 RGB images of size 150 × 150 px (74 µm × 74 µm), each belonging to one of eight tissue classes. Images were digitized at 20× magnification (0.495 µm/pixel) using an Aperio ScanScope (Leica Biosystems). The samples are formalin‑fixed, paraffin‑embedded (FFPE) colorectal adenocarcinoma tissues from pathology archives, fully anonymized and approved by an ethics committee.

Source
hugging_face
Created
Nov 28, 2025
Updated
May 25, 2024
Signals
270 views
Availability
Linked source ready
Overview

Dataset description and usage context

Colorectal Cancer Histology Texture Dataset

Description

  • License: CC‑BY‑4.0
  • Size Category: 1K < n < 10K
  • Task Category: Image Classification
  • Tags:
    • Biology
    • Colorectal Cancer
    • Histopathology
    • Histology
    • Digital Pathology

Configuration

  • Default Split:
    • Split: train
    • Path: data/train-*

Information

  • Features:

    • Image: type image
    • Label: categorical with classes:
      • 0 ADIPOSE
      • 1 COMPLEX
      • 2 DEBRIS
      • 3 EMPTY
      • 4 LYMPHO
      • 5 MUCOSA
      • 6 STROMA
      • 7 TUMOR
  • Split: train – 5,000 samples, 329 215 083 bytes

  • Download Size: 293 441 024 bytes

  • Dataset Size: 329 215 083 bytes

Need downstream help?

Pair the dataset with AI analysis and content workflows.

Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.

Explore AI studio