{"success":true,"data":{"author":"Gorkem Polat, Haluk Tarik Kani, Ilkay Ergenc et al. (Marmara University); repackaged by Chehab Lab","bucketAccessType":"public","bucketName":"chehablab/LIMUC","bucketPrefix":"","category":"Health & Biomedical","cloudProvider":"huggingface","createdAt":"2026-11-21T22:16:41.670Z","description":"LIMUC (Labeled Images for Ulcerative Colitis) is a colonoscopy image dataset for grading the severity of ulcerative colitis, a chronic inflammatory bowel disease. It holds 11,276 endoscopic images from 564 patients examined at Marmara University School of Medicine, Istanbul, each graded by experienced gastroenterologists with the Mayo endoscopic subscore: Mayo 0 normal or inactive (6,105), Mayo 1 mild (3,052), Mayo 2 moderate (1,254), and Mayo 3 severe (865). This copy is built from the authors' patient-based release, so every row carries a patient ID for patient-level evaluation; images were converted losslessly from BMP to PNG at native resolution and an integer label added. Three Parquet files under data/, about 1.3 GB (measured 2026-11-21); read with Hugging Face datasets. Use for ulcerative colitis severity classification and inflammatory bowel disease endoscopy models. Original data on Zenodo (DOI 10.5281/zenodo.5827695), released under CC BY 4.0. Cite Polat et al., Inflammatory Bowel Diseases 29(9):1431-1439 (2023), doi 10.1093/ibd/izac226.","displayTitle":"LIMUC: 11,276 colonoscopy images of ulcerative colitis labeled with Mayo endoscopic score, 564 patients (Marmara University)","doi":"10.5281/zenodo.5827695","huggingFaceUrl":"https://huggingface.co/datasets/chehablab/LIMUC","id":"4a19bfdf-b295-5002-868e-3fa2888a7b69","license":"CC-BY-4.0","licenseEvidenceUrl":"https://zenodo.org/records/5827695","objectCountEstimate":5,"probeStatus":"ok","referenceUrl":"https://doi.org/10.1093/ibd/izac226","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"chehablab/LIMUC\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"chehablab/LIMUC\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=chehablab, license:cc-by-4.0, size_categories:10K<n<100K, format:parquet, downloads=48"],"format":"mixed","generatedAt":"2026-11-21T22:32:17.251Z","layout":{"example":"data/train-00000-of-00003.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"50415231150415d4c80215c6484c15ae0d1500120000aaa4015011000000315f55435f70617469656e745f315f31364a15000432384e1500043912013f00303240000c305f321505163a17000c305f31355e190000375e1900043814564b0000345a180000375a18000038096136ea001030315f31305e190000315e19000dc5464b0000365e32005e7d0000325e64000432335a4b0000325e7d0000322d4242960000325e7d0000335efa000433332944424b000d63003236ee0100323544461900003409c74619000dc74619002d4446190000395a4b0004323209ae42320000365a18000d49003336c600003315c64619000dad4619002d114619000039097c4219000dac4218002d724218004dce003436ac0004345f4d384618006219002d584632004d1f421900515142190004333009dd4219006dab003536ad00003511ad4618004d814619000dad4619002dbc42190000325e4b0071494632005e640011f84632002d2a4619002dee4219004d3800373611010037556a4619002dbf4619008d104619006d62461900627d000dc74632000dc7421900adce4618002d754619006232005ec7000032adcf424b000035496c4218000dc50038365b0104385f31d74619006dac42190000336d944219004d210039366300003935be4619000df74619000d7c42190031424619005e4b0031424632004d094619004d0946","encoding":"hex","key":"data/train-00000-of-00003.parquet","truncatedAt":512}],"sampleObjects":["data/train-00000-of-00003.parquet","data/train-00001-of-00003.parquet","data/train-00002-of-00003.parquet"],"sidecars":[{"bytes":"---\ndataset_info:\n  features:\n  - name: image_id\n    dtype: string\n  - name: image\n    dtype: image\n  - name: patient_id\n    dtype: string\n  - name: mayo\n    dtype: string\n  - name: label\n    dtype: int32\n  splits:\n  - name: train\n    num_bytes: 1318668664\n    num_examples: 11276\n  download_size: 1319315720\n  dataset_size: 1318668664\nconfigs:\n- config_name: default\n  data_files:\n  - split: train\n    path: data/train-*\nlicense: cc-by-4.0\ntask_categories:\n- image-classification\nlanguage:\n- en\ntags:\n- endoscopy\n- colonoscopy\n- gastroenterology\n- ulcerative-colitis\n- inflammatory-bowel-disease\n- mayo-endoscopic-score\n- medical\n- classification\npretty_name: LIMUC (Labeled Images for Ulcerative Colitis)\nsize_categories:\n- 10K<n<100K\n---\n# LIMUC — Labeled Images for Ulcerative Colitis\n\n> [!NOTE]\n> This dataset description was generated with the help of generative AI. If you find anything wrong or missing, please report it in the [Community discussions](https://huggingface.co/datasets/chehablab/LIMUC/discussions).\n\nThis dataset is derived from the **LIMUC dataset** (Polat et al., 2022), a collection of colonoscopy images from patients with ulcerative colitis, originally hosted on **Zenodo**. The images were collected at Marmara University School of Medicine, Istanbul, between December 2011 and July 2019, and graded with the **Mayo endoscopic score** (MES) by experienced gastroenterologists.\nHere, we used the patient-based release, read the patient ID and Mayo score from its folder structure, converted the images from BMP to PNG without loss, and added an integer label.\nThe dataset is designed for ulcerative colitis severity classification with **patient-level** evaluation.\n\n\n## 📦 Dataset Structure\n\nThe dataset has a single `train` split, with each row corresponding to a single colonoscopy image:\n\n| Field | Description |\n|---|---|\n| `image_id` | Original image filename without extension (e.g., `UC_patient_190_16`) |\n| `image` | RGB colonoscopy image (PNG, converted losslessly from the original BMP, native resolution) |\n| `patient_id` | Patient ID from the original folder (e.g., `190`) |\n| `mayo` | Mayo endoscopic score: `Mayo 0`, `Mayo 1`, `Mayo 2`, `Mayo 3` |\n| `label` | Integer Mayo score: `0`–`3` (see Feature Descriptions) |\n\n**Class Distribution:**\n- Mayo 0: 6,105 images (54.1%)\n- Mayo 1: 3,052 images (27.1%)\n- Mayo 2: 1,254 images (11.1%)\n- Mayo 3: 865 images (7.7%)\n- Total: 11,276 images from 564 patients\n\n\n## ⚙️ Preprocessing\n\n- `patient_based_classified_images.zip` was downloaded from Zenodo (record `5827695`); the authors' image-level `train_and_validation_sets.zip` and `test_set.zip` hold the same images and were not used\n- The patient ID was taken from each image's patient folder, and the Mayo score from its `Mayo k` subfolder (`<patient>/Mayo k/UC_patient_<patient>_<n>.bmp`)\n- Images were converted from BMP to PNG (lossless; no resizing, cropping or color change)\n- All images were preserved\n\n\n## 🚀 Usage\n\n```python\nfrom datasets import load_dataset\nimport matplotlib.pyplot as plt\n\nds = load_dataset(\"chehablab/LIMUC\", split=\"train\")\n\n# One example image per Mayo score\nclasses = [\"Mayo 0\", \"Mayo 1\", \"Mayo 2\", \"Mayo 3\"]\nlabels = ds[\"label\"]\nfig, axes = plt.subplots(1, len(classes), figsize=(12, 3))\nfor ax, label in zip(axes, range(len(classes))):\n    sample = ds[labels.index(label)]\n    ax.imshow(sample[\"image\"])\n    ax.set_title(f\"{classes[label]} · patient {sample['patient_id']}\", fontsize=9)\n    ax.axis(\"off\")\nplt.tight_layout()\nplt.show()\n\n# Split by patient (never by image)\nfrom sklearn.model_selection import GroupShuffleSplit\ndf = ds.select_columns([\"patient_id\", \"label\"]).to_pandas()\ngss = GroupShuffleSplit(n_splits=1, test_size=0.2, random_state=0)\ntrain_idx, test_idx = next(gss.split(df, groups=df[\"patient_id\"]))\ntrain, test = ds.select(train_idx), ds.select(test_idx)\n```\n\n---\n\n## 🏷️ Feature Descriptions\n\n### Mayo Endoscopic Score\nFour grades of ulcerative colitis activity:\n\n- **Mayo 0** (`label=0`): 6,105 images — Normal or inactive disease\n- **Mayo 1** (`label=1`): 3,052 images — Mild di","encoding":"utf8","key":"README.md","name":"README.md","size":8385,"truncated":true}],"sourceObjectKey":"data/train-00000-of-00003.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"chehablab/LIMUC\")\nprint(ds)","starterSamplePath":"data/train-00000-of-00003.parquet","totalObjects":5},"summary":"LIMUC (Labeled Images for Ulcerative Colitis) is a colonoscopy image dataset for grading the severity of ulcerative colitis, a chronic inflammatory bowel disease.","title":"LIMUC: 11,276 colonoscopy images of ulcerative colitis labeled with Mayo endoscopic score, 564 patients (Marmara University)","updatedAt":"2026-11-22T01:04:15.163Z","visibility":"public","websiteUrl":"https://zenodo.org/records/5827695"}}