Datasets:
image_id stringlengths 36 36 | image imagewidth (px) 332 1.92k | organ stringclasses 2
values | category stringclasses 4
values | finding stringclasses 23
values | label int32 0 22 |
|---|---|---|---|---|---|
0033588e-ae3b-4cc0-aa8e-2ac34857c9cb | upper-gi-tract | pathological-findings | barretts | 0 | |
0c9e0051-684f-4cad-9bd2-ed081d876249 | upper-gi-tract | pathological-findings | barretts | 0 | |
0ca3fdfe-1ab5-401c-a7f1-fb13f2586340 | upper-gi-tract | pathological-findings | barretts | 0 | |
0ceb7736-72bf-4480-9f98-9060fdddba9f | upper-gi-tract | pathological-findings | barretts | 0 | |
12b11a85-a5ee-4770-b643-254d631c1585 | upper-gi-tract | pathological-findings | barretts | 0 | |
1778d904-b13e-4461-9cd4-6a5bf1d3b37c | upper-gi-tract | pathological-findings | barretts | 0 | |
1aab090c-fd5d-45ae-b7a5-32c5eca64840 | upper-gi-tract | pathological-findings | barretts | 0 | |
1ba881b6-36de-4e9f-8867-ac9bded80858 | upper-gi-tract | pathological-findings | barretts | 0 | |
1e5ac457-5800-43de-91cf-813a5a18e148 | upper-gi-tract | pathological-findings | barretts | 0 | |
1f904b6e-a184-4d2f-8dae-8876b293568a | upper-gi-tract | pathological-findings | barretts | 0 | |
217721be-f88d-4630-a85f-c4b22a752ea8 | upper-gi-tract | pathological-findings | barretts | 0 | |
226a262c-3a76-41fb-91ec-7122790ab452 | upper-gi-tract | pathological-findings | barretts | 0 | |
22fbd274-1353-4bb1-8133-9f0e8433b2e8 | upper-gi-tract | pathological-findings | barretts | 0 | |
270c7eb3-1c2e-4ecf-9b61-f38ebf146673 | upper-gi-tract | pathological-findings | barretts | 0 | |
2b2a52de-29c0-44ac-b21c-0b47092e566d | upper-gi-tract | pathological-findings | barretts | 0 | |
318bf32b-1d5f-4833-a8a9-b1755e6121a4 | upper-gi-tract | pathological-findings | barretts | 0 | |
3463a99e-3b67-4576-946f-e437ac21cb50 | upper-gi-tract | pathological-findings | barretts | 0 | |
4b7e9f4f-bf72-43d4-9cfa-d70d58ec5891 | upper-gi-tract | pathological-findings | barretts | 0 | |
52ad7d62-e67c-4fdd-ac89-a8f8a7d722a2 | upper-gi-tract | pathological-findings | barretts | 0 | |
5460818f-5c1d-4e61-9f90-4f974d150689 | upper-gi-tract | pathological-findings | barretts | 0 | |
5a8ecda7-8963-4536-bc2a-414564b2447b | upper-gi-tract | pathological-findings | barretts | 0 | |
5fdcef4e-fb00-489e-b877-88a614662866 | upper-gi-tract | pathological-findings | barretts | 0 | |
6548c39a-3f73-491e-99b9-4c364adf23ff | upper-gi-tract | pathological-findings | barretts | 0 | |
68183bba-a974-4dd2-8f5a-15ab429d83c4 | upper-gi-tract | pathological-findings | barretts | 0 | |
7c243b75-2082-46ff-b6b2-4b3a734c0dbd | upper-gi-tract | pathological-findings | barretts | 0 | |
7ce23f01-78dd-4133-8b1b-68636217f878 | upper-gi-tract | pathological-findings | barretts | 0 | |
8595f95a-5216-44c7-bdf4-dc325f230ecd | upper-gi-tract | pathological-findings | barretts | 0 | |
8b730bab-4e4e-4f92-95f3-91dd345d78a0 | upper-gi-tract | pathological-findings | barretts | 0 | |
9393fa83-caa6-4863-a5de-7fa1028da946 | upper-gi-tract | pathological-findings | barretts | 0 | |
a6c44fe9-bbf5-4bb5-98f9-082006d6dd05 | upper-gi-tract | pathological-findings | barretts | 0 | |
aee0204e-fa72-44f4-82ab-4e89bbaaf67f | upper-gi-tract | pathological-findings | barretts | 0 | |
b1be233d-931b-4859-bb87-fe0263384553 | upper-gi-tract | pathological-findings | barretts | 0 | |
b470a84e-6560-4d7c-bb4e-2dbcbd96d17d | upper-gi-tract | pathological-findings | barretts | 0 | |
b994ff9f-38c9-426b-8197-c6fcedd3ea5b | upper-gi-tract | pathological-findings | barretts | 0 | |
c701e6b5-4fbf-45a5-ae6c-4a82b9592e94 | upper-gi-tract | pathological-findings | barretts | 0 | |
d7c33881-2184-405b-b115-a86f2118aa18 | upper-gi-tract | pathological-findings | barretts | 0 | |
fbc1e03f-2416-40e7-b6fe-16fcf1ebee11 | upper-gi-tract | pathological-findings | barretts | 0 | |
fbd2dfd7-e110-47a9-b69f-f404aa0652f8 | upper-gi-tract | pathological-findings | barretts | 0 | |
fee3b50b-abad-4bc5-a909-cf1b291f7ef7 | upper-gi-tract | pathological-findings | barretts | 0 | |
ff05511a-b483-4589-bd65-89d8077f2c50 | upper-gi-tract | pathological-findings | barretts | 0 | |
ff217c66-272b-4144-85b2-e77e8d14f6c4 | upper-gi-tract | pathological-findings | barretts | 0 | |
03b9c09d-9018-4104-a249-2ec9c471dbf0 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
0411e66e-a3f4-4833-8133-971cbbebefc2 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
08b531a8-4ae8-44d7-b92c-69566e84a519 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
0bc45252-73c2-4eca-9ed8-6a0e77427fe0 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
0efc2d9e-f46d-40d9-afdc-8189c29336e9 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
101b7a09-3116-4f28-afc4-e4705999a038 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
14f12b72-ec9d-4040-a995-2420072dbd20 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
18650e78-2cae-4609-8384-a1b9911f27e7 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
237b6b27-a102-46b7-9aa5-ba99c7bcb6c7 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
38e800ec-67eb-4218-9e6f-98444df41eaa | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
3d73b85f-b2aa-4836-b413-ed50f6637b30 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
3dcbbdf8-4c04-4f47-8aa8-e6859258ddab | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
44c452dd-f004-4e6f-8a9b-dc779e4d48d2 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
463c84ec-2e3e-46c4-9de7-f5f834b28010 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
4722aed9-b974-466c-a350-5870ec0383e6 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
50219546-53ac-4927-95ea-f81245053521 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
5e554e89-b4d6-4926-afe8-d2819693617c | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
5f586e2d-2269-41aa-b801-d5ae87129b75 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
617ae14a-7c51-4e19-a192-75c7a03bf658 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
6361016f-717f-4c1e-a94f-e9750f0944e4 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
70921113-09bd-4795-9050-4d633dea80ce | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
70d62048-39ec-411f-8e51-81f54a0170cb | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
785307ca-667a-4e09-ad4c-e81db22b7d3b | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
7a9d0c13-00d2-4220-a47e-ff24d23905e9 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
84ceedf7-5ada-4546-8f3f-14fc0c658edb | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
8a044c30-eb04-4bdb-be4d-734e1780a4c2 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
95a96a24-c097-4d79-b62b-cb86c3840537 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
9af95b54-502a-4459-b3b7-faf5fa3ed5a1 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
9b852df7-5851-4620-a3ec-1a43a0593171 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
9c1c6c31-dfa3-4929-97c8-c4bc0068f9d9 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
9cfda4f2-9693-41ed-b972-ee7e3d35ca5b | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
a0e5972f-a1af-4b52-a468-9aa028cfbe16 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
a30cd737-df9c-46aa-800d-8a2c5986f50b | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
a55101b4-5628-4c8e-acbb-78c0ca83e2f1 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
a9829b0d-5bfa-423b-b58e-599cb24f4453 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
b29a057b-efd5-4a4d-907d-177a8f081767 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
c2428751-0786-41e8-a451-0120e1624c78 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
c4b05b19-a861-4d69-8b01-57004367cf22 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
cae271f6-943d-4632-b860-60a0b928e5e7 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
cf06a21b-d3ea-4c51-afa9-329f18be8518 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
cf5189c0-a0f6-4c4c-9e96-d773ded6447e | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
cf6bfd1e-a2ea-4763-8024-da57df87ceb6 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
d0313453-b4ad-4112-ac4e-830966d7c946 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
d28adfe1-b3c6-43d1-9ea2-c8f3de9630e2 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
d6b620c7-4466-4398-ad48-c020e48d7b29 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
dc31d822-e062-4f78-813e-5f9b9ad2bcf7 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
deaa2cad-a016-4955-acd8-eb7ef305ba3e | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
e64aef67-1a24-4d3f-b6a5-67381555e7df | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
f06503c7-0f4d-4d2b-a30e-652b69b502db | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
f3298956-5b2d-4d1e-946e-a32c47b582b6 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
f3b86789-4b3c-4978-a26f-c4d95fa9124d | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
f6c64961-b41c-4569-9a44-1deee095c095 | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
fd091d93-f3bf-494d-bbee-5da2cc70989b | upper-gi-tract | pathological-findings | barretts-short-segment | 1 | |
00053206-b78e-4acd-b2ef-078130b71b26 | lower-gi-tract | quality-of-mucosal-views | bbps-0-1 | 2 | |
0013240d-bb50-4886-a616-e1995eb1a163 | lower-gi-tract | quality-of-mucosal-views | bbps-0-1 | 2 | |
004df018-504d-43c3-8071-a853945910c5 | lower-gi-tract | quality-of-mucosal-views | bbps-0-1 | 2 | |
007dc2fb-8d49-45d2-8268-b7f1be7ea2d7 | lower-gi-tract | quality-of-mucosal-views | bbps-0-1 | 2 | |
01772699-a9fd-4009-b17d-7f03844e285e | lower-gi-tract | quality-of-mucosal-views | bbps-0-1 | 2 | |
01eb45e2-0d2e-4fa2-bbd3-68d61163596b | lower-gi-tract | quality-of-mucosal-views | bbps-0-1 | 2 |
HyperKvasir β Labeled Gastrointestinal Endoscopy Images
This dataset is derived from HyperKvasir (Borgli et al., 2020), a multi-class image and video dataset of the gastrointestinal (GI) tract, originally hosted by Simula and mirrored on OSF. The images were collected during routine gastroscopies and colonoscopies in Norway and labelled by experienced gastrointestinal endoscopists into 23 classes covering anatomical landmarks, pathological findings, mucosal-view quality and therapeutic interventions. Here, we provide only the 10,662 labeled images. We kept the original JPG images unchanged and added the organ, category, finding and an integer label, all taken from the original folder structure. The dataset is designed for GI endoscopy image classification.
π¦ Dataset Structure
The dataset has a single train split, with each row corresponding to a single endoscopy image:
| Field | Description |
|---|---|
image_id |
Original image filename without extension (a UUID) |
image |
RGB endoscopy image (variable resolution, JPG) |
organ |
lower-gi-tract (colonoscopy) or upper-gi-tract (gastroscopy) |
category |
anatomical-landmarks, pathological-findings, quality-of-mucosal-views or therapeutic-interventions |
finding |
One of 23 classes (e.g., polyps, cecum, esophagitis-a) |
label |
Integer class 0β22: the index of finding in the alphabetically sorted list of classes |
βοΈ Preprocessing
hyper-kvasir-labeled-images.zipwas downloaded from https://datasets.simula.no/hyper-kvasir/ (the full 58.6 GB release, including videos and unlabeled images, was not used)- Images were kept as the original JPG files (no re-encoding, resizing, cropping or color normalization)
organ,categoryandfindingwere taken from each image's folder path:labeled-images/<organ>/<category>/<finding>/<image>.jpglabelis the index offindingin the alphabetically sorted list of the 23 classes- All labeled images were preserved
π Usage
from datasets import load_dataset
import matplotlib.pyplot as plt
ds = load_dataset("chehablab/HyperKvasir", split="train")
# One example image per class
findings = ds["finding"]
classes = sorted(set(findings))
fig, axes = plt.subplots(4, 6, figsize=(16, 11))
for ax in axes.flat:
ax.axis("off")
for ax, c in zip(axes.flat, classes):
sample = ds[findings.index(c)]
ax.imshow(sample["image"])
ax.set_title(f"{c}\n{sample['organ']}", fontsize=8)
plt.tight_layout()
plt.show()
# Filter by organ or category
polyps = ds.filter(lambda x: x["finding"] == "polyps")
upper_gi = ds.filter(lambda x: x["organ"] == "upper-gi-tract")
print(f"Polyps: {len(polyps)}, upper GI: {len(upper_gi)}")
# No split in this release: make a stratified train/test split
split = ds.class_encode_column("label").train_test_split(test_size=0.2, stratify_by_column="label", seed=0)
π·οΈ Feature Descriptions
Classes
23 classes, grouped by organ and category:
Lower GI tract (colonoscopy)
- Anatomical landmarks
- Cecum β Most proximal part of the colon; reaching it confirms a complete colonoscopy
- Ileum β Terminal ileum, reached through the ileocecal valve
- Retroflex rectum β Retroflexed view of the distal rectum
- Pathological findings
- Hemorrhoids β Dilated vascular structures in the anal canal
- Polyps β Mucosal growths, potential precursors of colorectal cancer
- Ulcerative colitis, grade 0β1 / 1 / 1β2 / 2 / 2β3 / 3 β Inflammation graded by severity on the Mayo endoscopic score
- Quality of mucosal views
- BBPS 0β1 β Poor bowel preparation (Boston Bowel Preparation Scale 0β1)
- BBPS 2β3 β Good bowel preparation (BBPS 2β3)
- Impacted stool β Stool that blocks the view of the mucosa
- Therapeutic interventions
- Dyed lifted polyps β Polyps lifted by injection of dyed saline before removal
- Dyed resection margins β Resection site after polyp removal
Upper GI tract (gastroscopy)
- Anatomical landmarks
- Pylorus β Opening from the stomach into the duodenum
- Retroflex stomach β Retroflexed view of the upper stomach and cardia
- Z-line β Junction between the esophagus and the stomach
- Pathological findings
- Barrett's β Barrett's esophagus (metaplasia of the distal esophagus)
- Barrett's, short segment β Short-segment Barrett's esophagus
- Esophagitis A β Mild esophagitis (Los Angeles grade A)
- Esophagitis BβD β More severe esophagitis (Los Angeles grades BβD)
Image Specifications
- Modality: White-light GI endoscopy (gastroscopy and colonoscopy)
- Image size: Variable resolution, RGB, JPG
- Content: Single endoscopy frames
Acquisition Protocol
- Procedures: Routine clinical gastroscopies and colonoscopies, Norway
- Annotation: Labelled by experienced gastrointestinal endoscopists
π§ͺ Use Cases
- Multi-class classification of GI endoscopy images
- Polyp detection and classification
- Grading of ulcerative colitis and esophagitis
- Bowel-preparation quality assessment (BBPS)
- Anatomical landmark recognition for automatic procedure reporting
- Representation learning, feature extraction and self-supervised pretraining for endoscopy
β οΈ Important Notes
- No split in this release: The dataset has a single
trainsplit. Make your own stratified train/validation/test split (see Usage). - No patient IDs: Images are named by UUID and the release does not say which images come from the same patient or procedure, so a random split can place frames from one examination in both training and evaluation.
- Class imbalance: Class sizes vary widely. Use weighted sampling, class balancing or class merging as needed.
- Fine-grained grades: Neighbouring ulcerative-colitis grades (e.g.,
grade-1vsgrade-1-2) and Barrett's subtypes are hard to tell apart and are often merged in practice.
π Citation
This dataset is derived from data originally published by Simula under a CC BY 4.0 license. If you use this dataset, please acknowledge Chehab Lab and cite the original HyperKvasir dataset:
@misc{Borgli2020HyperKvasirData,
author = {Borgli, Hanna and Thambawita, Vajira and Smedsrud, Pia H. and Hicks, Steven and Jha, Debesh and Eskeland, Sigrun L. and Randel, Kristin Ranheim and Pogorelov, Konstantin and Lux, Mathias and Nguyen, Duc Tien Dang and Johansen, Dag and Griwodz, Carsten and Stensland, H{\aa}kon K. and Garcia-Ceja, Enrique and Schmidt, Peter T. and Hammer, Hugo L. and Riegler, Michael A. and Halvorsen, P{\aa}l and de Lange, Thomas},
title = {HyperKvasir [dataset]},
year = {2020},
publisher = {OSF},
doi = {10.17605/OSF.IO/MH9SJ}
}
Additionally, cite the research paper that introduced this dataset:
@article{Borgli2020HyperKvasir,
author = {Borgli, Hanna and Thambawita, Vajira and Smedsrud, Pia H. and Hicks, Steven and Jha, Debesh and Eskeland, Sigrun L. and Randel, Kristin Ranheim and Pogorelov, Konstantin and Lux, Mathias and Nguyen, Duc Tien Dang and Johansen, Dag and Griwodz, Carsten and Stensland, H{\aa}kon K. and Garcia-Ceja, Enrique and Schmidt, Peter T. and Hammer, Hugo L. and Riegler, Michael A. and Halvorsen, P{\aa}l and de Lange, Thomas},
title = {HyperKvasir, a comprehensive multi-class image and video dataset for gastrointestinal endoscopy},
journal = {Scientific Data},
volume = {7},
pages = {283},
year = {2020},
doi = {10.1038/s41597-020-00622-y}
}
βοΈ Original Authors
Hanna Borgli, Vajira Thambawita, Pia H. Smedsrud, Steven Hicks, Debesh Jha, Sigrun L. Eskeland, Kristin Ranheim Randel, Konstantin Pogorelov, Mathias Lux, Duc Tien Dang Nguyen, Dag Johansen, Carsten Griwodz, HΓ₯kon K. Stensland, Enrique Garcia-Ceja, Peter T. Schmidt, Hugo L. Hammer, Michael A. Riegler, PΓ₯l Halvorsen, Thomas de Lange
Simula Research Laboratory, Norway, with collaborating hospitals and universities
Simula: https://datasets.simula.no/hyper-kvasir/ (original files)
OSF mirror: https://osf.io/mh9sj/
π License
This dataset is released under the Creative Commons Attribution 4.0 International (CC BY 4.0) license, consistent with the original HyperKvasir dataset.
You may copy, modify, distribute, and use the data for any purpose (commercial or non-commercial), provided that appropriate credit is given to Borgli et al., Simula, and the original paper.
Chehab Lab @ 2026
- Downloads last month
- -
