Datasets › NCT-CRC-HE-100K

NCT-CRC-HE-100K

7 Apr 2018 archive 2025-07-28

The NCT-CRC-HE-100K dataset is a set of 100,000 non-overlapping image patches extracted from 86 H&E stained human cancer tissue slides and normal tissue from the NCT biobank (National Center for Tumor Diseases) and the UMM pathology archive (University Medical Center Mannheim). While the dataset Colorectal Cacner-Validation-Histology-7K (CRC-VAL-HE-7K) consist of 7180 images extracted from 50 patients with colorectal adenocarcinoma and were used to create a dataset that does not overlap with patients in the NCT-CRC-HE-100K dataset. It was created by pathologists by manually delineating tissue regions in whole slide images into the following nine tissue classes: Adipose (ADI), background (BACK), debris (DEB), lymphocytes (LYM), mucus (MUC), smooth muscle (MUS), normal colon mucosa (NORM), cancer-associated stroma (STR), colorectal adenocarcinoma epithelium (TUM).

Image source: https://www.cs.unc.edu/~mn/sites/default/files/macenko2009.pdf

Benchmarks archive 2025-07-28

All 2 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
Medical Image Classification NCT-CRC-HE-100K Efficientnet-b0 Accuracy (%) 95.59 EfficientNet: Rethinking Model Scaling for Convolutional... ultralytics/yolov5 +143 7 Compare
Image Classification NCT-CRC-HE-100K SAG-ViT F1 98.61 SAG-ViT: A Scale-Aware, High-Fidelity Patching Approach... shravan-18/SAG-ViT 1 Compare

Papers archive 2025-07-28

7 shown of 7 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 8. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
SAG-ViT: A Scale-Aware, High-Fidelity Patching Approach with Graph Attention for Vision Transformers 1 1 14 Nov 2024 not harvested
ResNet strikes back: An improved training procedure in timm 14 1 1 Oct 2021 ran 0 of 3 samples (3 unverified)
RegNet: Self-Regulated Network for Image Classification 14 1 3 Jan 2021 not harvested
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks 144 1 28 May 2019 ran 171 of 302 samples (131 unverified; 112 pointer-only for licence)
Res2Net: A New Multi-scale Backbone Architecture 34 1 2 Apr 2019 ran 3 of 9 samples (6 unverified; 9 pointer-only for licence)
Densely Connected Convolutional Networks 146 1 25 Aug 2016 ran 18 of 71 samples (53 unverified; 7 pointer-only for licence)
Deep Residual Learning for Image Recognition 484 2 10 Dec 2015 ran 230 of 377 samples (147 unverified; 187 pointer-only for licence)

Dataset loaders archive 2025-07-28

1 loader as listed in the archive; links are outbound and not re-checked here.

Tasks archive 2025-07-28

License archive 2025-07-28

No licence recorded in the archive. Absence here is not a statement about the dataset's terms.

Modalities archive 2025-07-28

Languages archive 2025-07-28

No language tagged.

Variants archive 2025-07-28

  • NCT-CRC-HE-100K

1 variant name, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections