CRC-HGD-v1 (v5): A Histopathological Image Dataset for Grading Colorectal Cancer
Elham Amjadi and Amin Bahreini and Sayyed Mohammadreza Hakimian and Mohammad Hasan Emami and Alireza Fahim and Hojjatollah Rahimi and Hamidreza Bolhasani

@article{mendeley_yfp5sfj47m,
title= {CRC-HGD-v1 (v5): A Histopathological Image Dataset for Grading Colorectal Cancer},
journal= {},
author= {Elham Amjadi and Amin Bahreini and Sayyed Mohammadreza Hakimian and Mohammad Hasan Emami and Alireza Fahim and Hojjatollah Rahimi and Hamidreza Bolhasani },
year= {2026},
url= {https://data.mendeley.com/datasets/yfp5sfj47m/5},
doi= {10.17632/yfp5sfj47m.5},
version= {v5},
abstract= {**CRC-HGD-v1 (Mendeley Data version 5)** is a histopathology image dataset for **grading colorectal adenocarcinoma**. It contains **1,914 H&E stained microscopy images** from **214 patients**, labeled with WHO tumor grade.

### At a glance

| | |
|---|---|
| Images | 1,914 (JPEG, RGB, 800 x 800 px, 96 DPI) |
| Patients | 214 |
| Magnifications | 4x, 10x, 20x, 40x |
| Stain | Hematoxylin and Eosin (H&E) |
| Collected | 2014 to 2019 |
| Grading standard | WHO criteria (gland formation) |
| Origin | Poursina Hakim Digestive Disease Research Center (PDDRC), Isfahan University of Medical Sciences, Iran |

### Classes

| Class | Patients | Images |
|---|---|---|
| Grade I, well differentiated (more than 95% gland formation) | 106 | 860 |
| Grade II, moderately differentiated (50 to 95% gland formation) | 75 | 712 |
| Grade III, poorly differentiated (less than 50% gland formation) | 33 | 327 |
| Normal colorectal tissue | n/a | 8 (2 per magnification) |
| Mixed normal and tumoral tissue | n/a | 7 |

Each specimen has one 4x image. The number of 10x, 20x and 40x images varies per specimen, because the pathologist captured extra views where they were informative.

### File naming

`[SampleNumber]_CRC_[GradeLabel]_[ArchiveNumber]_[Magnification]_[ImageIndex]`, for example `01_CRC_G1_16130_4x_1`.

### Notes

The samples were gathered by a multidisciplinary team in pathology, surgery, gastroenterology, AI and data science. License: CC BY 4.0. This is version 5 and replaces the earlier version 2 release. Latest updates: https://databiox.com

### References

1. Amjadi E, Bahreini A, Hakimian SM, Emami MH, Fahim A, Rahimi H, Bolhasani H. *CRC-HGD: A Histopathological Image Dataset for Grading Colorectal Cancer.* arXiv:2607.12750, 2026. https://arxiv.org/abs/2607.12750
2. Amjadi E, et al. *CRC-HGD-v1: A Histopathological Image Dataset for Grading Colorectal Cancer.* Mendeley Data, v5, doi:10.17632/yfp5sfj47m.5. https://data.mendeley.com/datasets/yfp5sfj47m/5
3. Kather JN, et al. *NCT-CRC-HE-100K and CRC-VAL-HE-7K: colorectal cancer histology image sets (9 tissue classes).* Zenodo, 2018. https://zenodo.org/record/1214455
4. Pataki BE, et al. *HunCRC: annotated pathological slides to enhance deep learning applications in colorectal cancer screening.* Scientific Data, 2022. https://pubmed.ncbi.nlm.nih.gov/35764660/
5. *CNR-CRC: a new colorectal carcinoma grading dataset.* ISASI-CNR. https://dataset.isasi.cnr.it/?p=173
6. *Extended colorectal cancer grading dataset* (300 H&E images). TIB. https://service.tib.eu/ldmservice/dataset/extended-colorectal-cancer-grading-dataset
7. *STARC-9: a large-scale dataset for multi-class tissue classification for CRC histopathology.* NeurIPS 2025. https://neurips.cc/virtual/2025/loc/san-diego/poster/121437
},
keywords= {Image classification, histopathology, colorectal cancer, tumor grading, H&E, digital pathology},
license= {https://creativecommons.org/licenses/by/4.0/},
terms= {},
superseded= {}
}