{"success":true,"data":{"author":"Johannes Raufeisen, Kunpeng Xie, Fabian Hörst et al.; curated by SPORA","bucketAccessType":"public","bucketName":"csmp-hub/cytonuke","bucketPrefix":"","category":"Health & Biomedical","cloudProvider":"huggingface","createdAt":"2026-11-17T19:29:32.969Z","description":"CytoNuke addresses whole-cell (not just nucleus) segmentation in bright-field histology. It has 83 H&E images of head and neck squamous cell carcinoma (HNSCC) from 13 patients, each 256x256 pixels at a reported 0.5 micrometers per pixel, with hand-drawn outlines of nuclei and of the full cell boundary where it could be identified, so whole-cell masks cover only part of the labeled nuclei. The images come from the TCIA CPTAC-HNSCC collection with stain colors adjusted by the dataset creators. Patient-level splits: train.parquet (43 images, 9 patients, 2,344 nuclei, 1,547 whole cells), val.parquet (11 images, 2 patients) and test.parquet (29 images, 2 patients); these differ from the paper's evaluation groups. The original Zenodo archive and its README are included. About 30 MB (measured 2026-11-17). Images and masks are stored as PNG or TIFF bytes inside Parquet; read them with polars or pandas plus Pillow, or with Hugging Face datasets. Curated and packaged on Hugging Face by the SPORA project (csmp-hub), which keeps the original license and citation files. The creators' Zenodo record 15310323 is CC BY 4.0, and the CPTAC/TCIA source images are CC BY 3.0; keep both notices. Cite Raufeisen, Xie, Hörst et al., Computer Methods and Programs in Biomedicine 2024 (doi 10.1016/j.cmpb.2024.108215).","displayTitle":"CytoNuke: whole-cell and nucleus outlines in H&E images of head and neck squamous cell carcinoma","doi":"10.5281/zenodo.15310323","huggingFaceUrl":"https://huggingface.co/datasets/csmp-hub/cytonuke","id":"d7b46e42-9216-5c30-a58b-50308a15b5b8","license":"CC-BY-4.0","licenseEvidenceUrl":"https://zenodo.org/records/15310323","objectCountEstimate":13,"probeStatus":"ok","referenceUrl":"https://doi.org/10.1016/j.cmpb.2024.108215","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"csmp-hub/cytonuke\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"csmp-hub/cytonuke\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=csmp-hub, license:cc-by-4.0, size_categories:n<1K, format:parquet, downloads=194"],"format":"mixed","generatedAt":"2026-11-17T19:40:09.739Z","layout":{"example":"test.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"504152311504153415464c1502150012000028b52ffd201ad100001600000043334c2d30303939352d32312e7376737837387936321500151215242c15021510150615061c3600281643334c2d30303939352d32312e737673783738793632181643334c2d30303939352d32312e737673783738793632111100000028b52ffd2009490000020000000201010200150415bec60e15d6c60e4c1502150012000028b52ffda09fd10100f98c0e9bd1010089504e470d0a1a0a0000000d4948445200000100000001000802000000d3103f310001000049444154789cecfdd9922d59921d88e9b407333b830fd7ef8db81119595995852a340082e0031ff861fc137e0b3f800f2da4480bd90da0811a32ab322323e20e3e9dc1863da82a1f3c5af8116c137f3d72c4e56cb5adba740df87fffbffedf3acbd3da4beb4b693fbecc97a7c7617ff8f857dfefa7dcaef36137ec6ff7b576ed9a19a7fd08293e3e9e972f4f3fdc8effe6f71fe6383c7e7e4daef966dfb6fa3ffe971febe5fac3c77b1ac75eeabc96e7d7ab68fff7bffbf0dbbffde8fbdd1ffff0e5e5f1747d3d27a671ca0040c2004808eeb0396c5bfd707ff8e1af3f34c0f3d379b2deccf7bb4c4cd7cbbace5b08cc312a7344f8fc78aeafe7efeea6fb1f3e14a0cbcb459b22228077b552bb766dddb6a6d7d273600a61e9b6d64e44bd54dbb6cc686adfbcbf391cc6bf","encoding":"hex","key":"test.parquet","truncatedAt":512}],"sampleObjects":["CytoNuke Dataset.zip","NOTICE.md","examples.png","masks.png","source-README.txt"],"sidecars":[{"bytes":"Attribution 4.0 International\n\n=======================================================================\n\nCreative Commons Corporation (\"Creative Commons\") is not a law firm and\ndoes not provide legal services or legal advice. Distribution of\nCreative Commons public licenses does not create a lawyer-client or\nother relationship. Creative Commons makes its licenses and related\ninformation available on an \"as-is\" basis. Creative Commons gives no\nwarranties regarding its licenses, any material licensed under their\nterms and conditions, or any related information. Creative Commons\ndisclaims all liability for damages resulting from their use to the\nfullest extent possible.\n\nUsing Creative Commons Public Licenses\n\nCreative Commons public licenses provide a standard set of terms and\nconditions that creators and other rights holders may use to share\noriginal works of authorship and other material subject to copyright\nand certain other rights specified in the public license below. The\nfollowing considerations are for informational purposes only, are not\nexhaustive, and do not form part of our licenses.\n\n     Considerations for licensors: Our public licenses are\n     intended for use by those authorized to give the public\n     permission to use material in ways otherwise restricted by\n     copyright and certain other rights. Our licenses are\n     irrevocable. Licensors should read and understand the terms\n     and conditions of the license they choose before applying it.\n     Licensors should also secure all rights necessary before\n     applying our licenses so that the public can reuse the\n     material as expected. Licensors should clearly mark any\n     material not subject to the license. This includes other CC-\n     licensed material, or material used under an exception or\n     limitation to copyright. More considerations for licensors:\n    wiki.creativecommons.org/Considerations_for_licensors\n\n     Considerations for the public: By using one of our public\n     licenses, a licensor grants the public permission to use the\n     licensed material under specified terms and conditions. If\n     the licensor's permission is not necessary for any reason--for\n     example, because of any applicable exception or limitation to\n     copyright--then that use is not regulated by the license. Our\n     licenses grant only permissions under copyright and certain\n     other rights that a licensor has authority to grant. Use of\n     the licensed material may still be restricted for other\n     reasons, including because others have copyright or other\n     rights in the material. A licensor may make special requests,\n     such as asking that all changes be marked or described.\n     Although not required by our licenses, you are encouraged to\n     respect those requests where reasonable. More considerations\n     for the public:\n    wiki.creativecommons.org/Considerations_for_licensees\n\n=======================================================================\n\nCreative Commons Attribution 4.0 International Public License\n\nBy exercising the Licensed Rights (defined below), You accept and agree\nto be bound by the terms and conditions of this Creative Commons\nAttribution 4.0 International Public License (\"Public License\"). To the\nextent this Public License may be interpreted as a contract, You are\ngranted the Licensed Rights in consideration of Your acceptance of\nthese terms and conditions, and the Licensor grants You such rights in\nconsideration of benefits the Licensor receives from making the\nLicensed Material available under these terms and conditions.\n\n\nSection 1 -- Definitions.\n\n  a. Adapted Material means material subject to Copyright and Similar\n     Rights that is derived from or based upon the Licensed Material\n     and in which the Licensed Material is translated, altered,\n     arranged, transformed, or otherwise modified in a manner requiring\n     permission under the Copyright and Similar Rights held by the\n     Licensor. For purposes of this Public License, where the Licensed\n     Material is a musical work, performance, or sound recording,\n ","encoding":"utf8","key":"LICENSE","name":"LICENSE","size":18656,"truncated":true},{"bytes":"---\nlicense: cc-by-4.0\ntask_categories:\n- image-segmentation\ntags:\n- histopathology\n- nuclei\n- whole-cell\n- H&E\nsize_categories:\n- n<1K\nconfigs:\n- config_name: default\n  data_files:\n  - split: train\n    path: train.parquet\n  - split: val\n    path: val.parquet\n  - split: test\n    path: test.parquet\n---\n\n# CytoNuke\n\n## Overview\n\nCytoNuke contains **83 H&E images of head and neck squamous cell carcinoma from 13 patients**, with hand-drawn nucleus and whole-cell outlines. Each image is 256 × 256 pixels at a reported 0.5 µm per pixel.\n\nSome cell boundaries could not be identified, so whole-cell masks are available for only part of the labelled nuclei. The images come from CPTAC-HNSCC; the dataset creators adjusted their stain colours.\n\n**License:** [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/), with credit to the dataset creators. The original CPTAC/TCIA images also carry [CC BY 3.0](https://creativecommons.org/licenses/by/3.0/); keep both notices.\n\n## Files and splits\n\n| Split | File | Images | Patients | Source nuclei | Source whole cells |\n| --- | --- | --- | --- | --- | --- |\n| `train` | `train.parquet` | 43 | 9 | 2,344 | 1,547 |\n| `val` | `val.parquet` | 11 | 2 | 360 | 259 |\n| `test` | `test.parquet` | 29 | 2 | 1,287 | 801 |\n\nAll images from a patient stay in one split: nine patients in training, two in validation and two in testing. These groups were made for this version using seed 42; they differ from the paper's evaluation groups. Classes were not balanced between groups, and the small patient count limits comparisons. Exact patient lists are in `manifest.json`.\n\n`CytoNuke Dataset.zip` keeps all original images and outlines. Source descriptions, licenses and attribution notices are included alongside the Parquet files.\n\n## Fields\n\nEach row contains one image and its masks, stored as PNG bytes. Images are 256 × 256 pixels with three colour channels; masks are 256 × 256 arrays.\n\n| Field | Contents |\n| --- | --- |\n| `id` | Original image filename without `.png` |\n| `image` | RGB colour image; uint8 values from 0 to 255 |\n| `inst` | Numbered nucleus mask; uint16 |\n| `inst_cell` | Numbered whole-cell mask, including the nucleus; uint16 |\n| `type` | Simple nucleus mask: 1 for a marked nucleus, 0 elsewhere; uint8 |\n| `pairs` | Suggested nucleus-to-cell matches, listed as `nucleus_id` and `cell_id` |\n| `patient` | Patient ID taken from the original CPTAC filename |\n| `slide_name` | Original slide name, including `.svs` |\n| `file_name`, `source_image_id` | Original image filename and annotation-file image ID |\n| `mpp_x`, `mpp_y` | Reported pixel size: 0.5 µm along both axes |\n\nPositive mask numbers are the original annotation IDs; zero means no marked object. A nucleus and its cell have different numbers. Keep the two masks separate: a whole cell includes its nucleus, so they overlap.\n\nA zero in the cell mask can mean that a cell was not annotated. It is not proof that no cell is present. There is no tissue-region mask.\n\n## Labels\n\nThe source labels identify nuclei and cells, not different cell types:\n\n| Original class number | Name | Stored in |\n| --- | --- | --- |\n| 0 | `NUCLEUS` | `inst`; `type` uses 1 for a marked nucleus |\n| 1 | `CELL` | `inst_cell` |\n\n`pairs` lists **2,312 suggested nucleus-to-cell matches**. A match is included only when a nucleus lies completely inside one cell and neither object overlaps another object in the other mask. The matches were calculated from the masks, not checked by a person. Partial, missing or ambiguous matches are left out; the objects themselves are kept.\n\nUse these matches rather than assuming equal numbers identify a nucleus and its cell. The original archive keeps all 3,991 nucleus outlines and 2,607 cell outlines. There are no separate cytoplasm-only masks.\n\n## Download and read a sample\n\n```python\nfrom io import BytesIO\n\nimport numpy as np\nimport polars as pl\nfrom huggingface_hub import hf_hub_download\nfrom PIL import Image\n\npath = hf_hub_download(\n    repo_id=\"csmp-hub/cytonuke\", filename=\"train.parquet\", repo_type=\"dataset\"\n)\nrow = pl.scan_parquet(path).head(1","encoding":"utf8","key":"README.md","name":"README.md","size":10756,"truncated":true},{"bytes":"{\n  \"dataset\": \"CytoNuke\",\n  \"source\": {\n    \"record\": \"https://zenodo.org/records/15310323\",\n    \"archive_sha256\": \"834836444fe749cc48daa7557458ba578bc00306a59286896e3d29666e98777a\",\n    \"coco_sha256\": \"75343e200456e0a39753ed06338aa0915df1e503eecc853f9d6ee66388bc9d3f\",\n    \"categories\": [\n      {\n        \"supercategory\": \"Cell\",\n        \"id\": 0,\n        \"name\": \"NUCLEUS\"\n      },\n      {\n        \"supercategory\": \"Cell\",\n        \"id\": 1,\n        \"name\": \"CELL\"\n      }\n    ]\n  },\n  \"software\": {\n    \"numpy\": \"2.5.4\",\n    \"pillow\": \"12.3.0\",\n    \"pyarrow\": \"23.0.1\",\n    \"pycocotools\": \"2.0.10\"\n  },\n  \"split_method\": \"New partitions; sorted CPTAC case prefixes, random.Random(42).shuffle, first 9 train, next 2 val, last 2 test. Original paper assignments unavailable in release and author repository 4433eb275d478375fd7066e8f7f9152416af5e10.\",\n  \"pairing_method\": \"Inferred on COCO raster masks: nucleus fully contained in cell, with exactly one positive-overlap partner in each direction. Source has no explicit links. IDs remain independent source annotation IDs.\",\n  \"rasterization\": \"pycocotools polygon union at source size; no repair, resizing, renumbering or artificial pixels. Same-compartment overlap rejected.\",\n  \"type_mapping\": {\n    \"0\": \"not an annotated nucleus\",\n    \"1\": \"NUCLEUS (source category_id 0)\"\n  },\n  \"splits\": {\n    \"train\": {\n      \"samples\": 43,\n      \"patients\": [\n        \"C3L-00997\",\n        \"C3L-00999\",\n        \"C3L-02651\",\n        \"C3L-03378\",\n        \"C3L-04844\",\n        \"C3N-04275\",\n        \"C3N-04277\",\n        \"C3N-04279\",\n        \"C3N-04611\"\n      ],\n      \"source_nuclei\": 2344,\n      \"source_cells\": 1547,\n      \"pairs\": 1351,\n      \"sha256\": \"680f371224fbddb941020c997d6e55ae16f7ba33c9b929797add74131b523c0a\"\n    },\n    \"val\": {\n      \"samples\": 11,\n      \"patients\": [\n        \"C3L-00977\",\n        \"C3L-02621\"\n      ],\n      \"source_nuclei\": 360,\n      \"source_cells\": 259,\n      \"pairs\": 218,\n      \"sha256\": \"14b43aa25afc34b2124986f4a1a3555cb4eb3e762bf0b8632be48d97d8a4927b\"\n    },\n    \"test\": {\n      \"samples\": 29,\n      \"patients\": [\n        \"C3L-00995\",\n        \"C3N-04278\"\n      ],\n      \"source_nuclei\": 1287,\n      \"source_cells\": 801,\n      \"pairs\": 743,\n      \"sha256\": \"c4fd8a00f6cd47348d2f3e27956b5aa446c59dcafa4a1604a5ab59bad3cfa361\"\n    }\n  },\n  \"rows\": [\n    {\n      \"id\": \"C3L-00997-21.svsx35y38\",\n      \"patient\": \"C3L-00997\",\n      \"slide_name\": \"C3L-00997-21.svs\",\n      \"source_image_id\": 31,\n      \"pairs\": [\n        {\n          \"nucleus_id\": 1658,\n          \"cell_id\": 1690\n        },\n        {\n          \"nucleus_id\": 1659,\n          \"cell_id\": 1691\n        },\n        {\n          \"nucleus_id\": 1660,\n          \"cell_id\": 1692\n        },\n        {\n          \"nucleus_id\": 1661,\n          \"cell_id\": 1693\n        },\n        {\n          \"nucleus_id\": 1664,\n          \"cell_id\": 1694\n        },\n        {\n          \"nucleus_id\": 1665,\n          \"cell_id\": 1696\n        },\n        {\n          \"nucleus_id\": 1666,\n          \"cell_id\": 1695\n        },\n        {\n          \"nucleus_id\": 1668,\n          \"cell_id\": 1699\n        },\n        {\n          \"nucleus_id\": 1669,\n          \"cell_id\": 1698\n        },\n        {\n          \"nucleus_id\": 1670,\n          \"cell_id\": 1700\n        },\n        {\n          \"nucleus_id\": 1671,\n          \"cell_id\": 1697\n        },\n        {\n          \"nucleus_id\": 1673,\n          \"cell_id\": 1701\n        },\n        {\n          \"nucleus_id\": 1674,\n          \"cell_id\": 1704\n        },\n        {\n          \"nucleus_id\": 1676,\n          \"cell_id\": 1702\n        },\n        {\n          \"nucleus_id\": 1678,\n          \"cell_id\": 1708\n        },\n        {\n          \"nucleus_id\": 1680,\n          \"cell_id\": 1706\n        },\n        {\n          \"nucleus_id\": 1681,\n          \"cell_id\": 1709\n        },\n        {\n          \"nucleus_id\": 1683,\n          \"cell_id\": 1710\n        },\n        {\n          \"nucleus_id\": 1685,\n          \"cell_id\": 1712\n        },\n        {\n          \"nucleus_id\": 1687,\n          \"cell_id\": 1713\n        },\n        {\n          \"nucleus_id\": 1688,\n          \"cell_id\": 1716\n       ","encoding":"utf8","key":"manifest.json","name":"manifest.json","size":270834,"truncated":true}],"sourceObjectKey":"test.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"csmp-hub/cytonuke\")\nprint(ds)","starterSamplePath":"examples.png","totalObjects":13},"summary":"CytoNuke addresses whole-cell (not just nucleus) segmentation in bright-field histology. It has 83 H&E images of head and neck squamous cell carcinoma (HNSCC) from 13 patients,…","title":"CytoNuke: whole-cell and nucleus outlines in H&E images of head and neck squamous cell carcinoma","updatedAt":"2026-11-17T23:04:17.265Z","visibility":"public","websiteUrl":"https://huggingface.co/datasets/csmp-hub/cytonuke"}}