{"success":true,"data":{"author":"Yasin Shokrollahi (VitaminP); ORION data from Jia-Ren Lin, Peter Sorger et al.; packaged by SPORA (csmp-hub)","bucketAccessType":"public","bucketName":"csmp-hub/vitaminp-orion-crc","bucketPrefix":"","category":"Health & Biomedical","cloudProvider":"huggingface","createdAt":"2026-11-21T20:45:56.246Z","description":"VitaminP ORION-CRC is the colorectal cancer subset of the VitaminP release for cross-modal whole-cell segmentation in histopathology. It holds 3,337 patches of 512x512 pixels from 41 tissue samples of 40 cases, each pairing a color H&E image with two co-registered fluorescence channels from the ORION same-section imaging platform: Hoechst (nuclei) and a combined E-cadherin plus CD45 cell-boundary signal. Nucleus and whole-cell instance masks were produced by segmentation models, and pathologists selected regions where outlines were reliable; 158,851 nucleus-to-cell links are provided. The original train, validation and test splits are kept with each case in one split. Fluorescence is stored as uint16 TIFF and masks as uint32 TIFF bytes inside train.parquet, val.parquet and test.parquet, about 3.9 GB (measured 2026-11-21); read with pandas or polars plus tifffile and Pillow. Pixel spacing is not recorded. Use for whole-cell segmentation from H&E, H&E-to-immunofluorescence translation and nucleus segmentation. This is a different product from the 19-channel ORION-CRC crops. Repackaged by SPORA (csmp-hub) from Zenodo DOI 10.5281/zenodo.21224234, released under CC BY 4.0; cite Shokrollahi 2026 and Lin et al., Nature Cancer 2023 (doi 10.1038/s43018-023-00576-1).","displayTitle":"VitaminP ORION-CRC: paired H&E and Hoechst immunofluorescence patches of colorectal cancer with nucleus and whole-cell masks","doi":"10.5281/zenodo.21224234","huggingFaceUrl":"https://huggingface.co/datasets/csmp-hub/vitaminp-orion-crc","id":"1a9a0e2f-de3b-50db-ae2d-685d5322fae0","license":"CC-BY-4.0","licenseEvidenceUrl":"https://zenodo.org/records/21224234","objectCountEstimate":18,"probeStatus":"ok","referenceUrl":"https://doi.org/10.1038/s43018-023-00576-1","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"csmp-hub/vitaminp-orion-crc\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"csmp-hub/vitaminp-orion-crc\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=csmp-hub, license:cc-by-4.0, size_categories:1K<n<10K, downloads=603"],"format":"mixed","generatedAt":"2026-11-21T20:48:40.497Z","layout":{"example":"crc-source-splits.csv","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"final_split,sample_name,sub_sample,patch_idx,dataset_source,source_note\r\ntrain,CRC01,,0,crc,declared_train_non_xenium\r\ntrain,CRC01,,1,crc,declared_train_non_xenium\r\ntrain,CRC01,,2,crc,declared_train_non_xenium\r\ntrain,CRC01,,3,crc,declared_train_non_xenium\r\ntrain,CRC01,,4,crc,declared_train_non_xenium\r\ntrain,CRC01,,5,crc,declared_train_non_xenium\r\ntrain,CRC01,,6,crc,declared_train_non_xenium\r\ntrain,CRC01,,7,crc,declared_train_non_xenium\r\ntrain,CRC01,,8,crc,declared_train_non_xenium\r\ntrain,CRC01,,9,crc,declar","encoding":"utf8","key":"crc-source-splits.csv","truncatedAt":512}],"sampleObjects":["NOTICE.md","crc-source-splits.csv","examples.png","masks.png","original-orion-markers.csv"],"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- histology\n- fluorescence\n- colorectal-cancer\n- nuclei\n- whole-cell\n- paired-images\nsize_categories:\n- 1K<n<10K\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# VitaminP ORION-CRC\n\n## Overview\n\nVitaminP ORION-CRC contains **3,337 paired H&E and fluorescence image patches of human colorectal cancer**. Each patch is 512 × 512 pixels, with masks outlining individual nuclei and whole cells. The patches come from 41 tissue samples representing 40 source cases.\n\nEach pair includes a colour H&E image and two fluorescence signals: **Hoechst** highlights nuclei, while a combined **E-cadherin and CD45** signal helps show cell boundaries. These are the two signals retained in the VitaminP preparation of the ORION images.\n\n**License: [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/).** You may use, modify and share the data, including commercially, with credit to the creators, a license link and a description of changes.\n\n## Files and splits\n\n| Split | File | Patches | Nuclei | Whole cells |\n| --- | --- | ---: | ---: | ---: |\n| `train` | `train.parquet` | 2,946 | 550,303 | 537,862 |\n| `val` | `val.parquet` | 272 | 41,738 | 42,118 |\n| `test` | `test.parquet` | 119 | 19,223 | 19,063 |\n\nThe original train, validation and test assignments are kept. Each source case stays within one split. The two samples named `CRC33_01` and `CRC33_02` are kept together as case `CRC33`. Some patches within a split overlap, so object counts can include the same cells more than once.\n\n## Fields\n\nEach row holds one aligned H&E/fluorescence pair and its masks.\n\n| Field | Contents |\n| --- | --- |\n| `id` | Unique patch name based on the source sample and patch number |\n| `image` | H&E PNG bytes; RGB uint8, shape `(512, 512, 3)` |\n| `image_if` | Fluorescence TIFF bytes; uint16, shape `(2, 512, 512)`, axes channel, height, width |\n| `channel_names` | `Hoechst`, `E-cadherin + CD45`, in that order |\n| `inst`, `inst_cell` | Nucleus and whole-cell masks as uint32 TIFF bytes, shape `(512, 512)` |\n| `type` | Nucleus foreground as uint8 PNG bytes: 0 background, 1 nucleus |\n| `pairs` | Inferred links, each with `nucleus_id` and `cell_id` |\n| `mask_format`, `mask_dtype` | `tiff`, `uint32` |\n| `image_if_format`, `image_if_axes`, `image_if_dtype` | `tiff`, `CYX`, `uint16` |\n| `case`, `sample` | Source case and tissue-sample names |\n| `source_index`, `source_split`, `source_path` | Original patch number, split and archive folder |\n| `roi_index` | Original selected-region number |\n| `x_start`, `y_start`, `x_end`, `y_end` | Original requested patch coordinates; x is column, y is row |\n\nOriginal image values and object numbers are kept. Zero is background in both instance masks. Large source numbers are stored as uint32. Numbers can repeat across patches showing the same part of a sample.\n\n## Labels\n\nSegmentation models produced the masks. Pathologists then selected regions where the nucleus and cell outlines looked reliable.\n\nNucleus and whole-cell numbers are independent. Use `pairs` to look up the **158,851 inferred links** where every nucleus pixel lies inside one cell. These conservative links cover 26.0% of the nuclei. Some cells have several linked nuclei. All unmatched nuclei, cells and partial overlaps remain in the separate masks. A missing link does not mean the objects are unrelated: many nucleus outlines cross cell outlines, sometimes by only a few pixels.\n\nThe original H&E and fluorescence nucleus masks match exactly, as do their whole-cell masks. Each shared mask is stored once. `type` marks nucleus foreground; it does not assign biological cell types.\n\n## Download and read a sample\n\n```python\nfrom io import BytesIO\n\nimport numpy as np\nimport polars as pl\nimport tifffile\nfrom huggingface_hub import hf_hub_download\nfrom PIL import Image\n\npath = hf_hub_download(\n    repo_id=\"csmp-hub/vitaminp-orion-crc\", filename=\"train.parquet\", repo_type=\"dataset\"\n)\nrow = pl.scan_par","encoding":"utf8","key":"README.md","name":"README.md","size":10178,"truncated":true},{"bytes":"{\n  \"dataset\": \"VitaminP ORION-CRC\",\n  \"source_record\": \"https://zenodo.org/records/21224234\",\n  \"source_doi\": \"10.5281/zenodo.21224234\",\n  \"license\": \"CC-BY-4.0\",\n  \"source_archive\": {\n    \"url\": \"https://zenodo.org/records/21224234/files/VitaminP_crc_zarr_v1.tar.zst?download=1\",\n    \"bytes\": 4411506650,\n    \"md5\": \"29086407f36fb5549907590e6cc6852f\",\n    \"sha256\": \"6633316c15424030d293c6215d2cc88e01bced13fc54588753c35d3f31c9697b\",\n    \"authentication\": false\n  },\n  \"conversion\": \"CRC subset only. H&E uint8 RGB saved losslessly as PNG. Both original independent uint32 instance masks saved as lossless YX TIFF bytes with source IDs unchanged. The identical H&E/IF masks are stored once after checking both originals. Two retained uint16 IF channels saved losslessly as CYX TIFF bytes, Hoechst first and E-cadherin/CD45 composite second. No further normalization, clipping or resizing. type is derived generic nucleus foreground, values 0/1. Full-containment links inferred; all objects retained. Source coordinates, case/sample IDs, patch/ROI indices and original split retained.\",\n  \"source_code_revision\": \"de637b2102f9be6fa064879d5a29d4491061ec8e\",\n  \"image_contract\": {\n    \"image\": \"HWC RGB uint8 PNG\",\n    \"image_if\": \"CYX uint16 TIFF, 2x512x512\",\n    \"if_channel_names\": [\n      \"Hoechst\",\n      \"E-cadherin + CD45\"\n    ],\n    \"masks\": \"YX uint32 TIFF, original instance IDs\",\n    \"type\": \"YX uint8 PNG, 0 background/1 nucleus\"\n  },\n  \"source_files\": {\n    \"LICENSE\": \"9e5f1b3c610b9c2da5c313bf81d577a7d1acec686bdb0384edefa6df0f90cd94\",\n    \"source-record.json\": \"4c557ca7ca67791b7a7bc1470ab3e76ffcc5bd21c9268200e196afe5c651d1e3\",\n    \"source-license.json\": \"7a65d67a90f421ad6ad6b01da7b7af94a1736abd51a0b47219f7de458212ce60\",\n    \"source-sha256sums.txt\": \"5fac3d31b25e06b1aa2a158ad8faa74f9975d2e0f7df5c05fbf4740422284287\",\n    \"crc-source-splits.csv\": \"921498bce5a6b6c11b1956b5ed1612aa6e4a5502b0317384dc07ec217b06bdbc\",\n    \"source-audit.json\": \"2adb5dd791d116d61c329b4115e75313a77d6cee6842965c6fa90d60e88c474e\",\n    \"split-audit.json\": \"8d4babdf9540b12b517522a3f67392f4f00d5480a68815cbf6f8bea76cba9de8\",\n    \"original-orion-markers.csv\": \"e7587ca37d48370fcc1d7c0246bef4d8bd9d119df7dede9a6e18a7bf29fcffd3\",\n    \"validation.json\": \"01fa20729aee678fbe7f5c77451fb898fb1363fae874acd4223d7ab30d1dd892\"\n  },\n  \"splits\": {\n    \"train\": {\n      \"samples\": 2946,\n      \"nuclei\": 550303,\n      \"cells\": 537862,\n      \"pairs\": 139757,\n      \"unmatched_nuclei\": 7172,\n      \"partial_single_cell_nuclei\": 89145,\n      \"multiple_cell_nuclei\": 314229,\n      \"cells_without_nucleus_overlap\": 14916,\n      \"cells_with_multiple_nucleus_overlaps\": 307468,\n      \"cells_with_multiple_pairs\": 2460,\n      \"same_number_pairs\": 1,\n      \"sha256\": \"3669450e269caca7a8f1404b441aaa5361c6db96b2d9bdab4a2606e468f25c63\"\n    },\n    \"val\": {\n      \"samples\": 272,\n      \"nuclei\": 41738,\n      \"cells\": 42118,\n      \"pairs\": 14347,\n      \"unmatched_nuclei\": 679,\n      \"partial_single_cell_nuclei\": 7583,\n      \"multiple_cell_nuclei\": 19129,\n      \"cells_without_nucleus_overlap\": 1763,\n      \"cells_with_multiple_nucleus_overlaps\": 19048,\n      \"cells_with_multiple_pairs\": 328,\n      \"same_number_pairs\": 0,\n      \"sha256\": \"c1f9626a5029f7ff39cd408aedd247b8483de2bd522ea26a58862707a2a9d1d9\"\n    },\n    \"test\": {\n      \"samples\": 119,\n      \"nuclei\": 19223,\n      \"cells\": 19063,\n      \"pairs\": 4747,\n      \"unmatched_nuclei\": 247,\n      \"partial_single_cell_nuclei\": 4369,\n      \"multiple_cell_nuclei\": 9860,\n      \"cells_without_nucleus_overlap\": 631,\n      \"cells_with_multiple_nucleus_overlaps\": 9703,\n      \"cells_with_multiple_pairs\": 76,\n      \"same_number_pairs\": 1,\n      \"sha256\": \"4dc9d00e5a061c8c6fec44b64c134cc16dc0e60a5c89cce03f7b578736d0f27d\"\n    }\n  },\n  \"totals\": {\n    \"nuclei\": 611264,\n    \"cells\": 599043,\n    \"pairs\": 158851,\n    \"unmatched_nuclei\": 8098,\n    \"partial_single_cell_nuclei\": 101097,\n    \"multiple_cell_nuclei\": 343218,\n    \"cells_without_nucleus_overlap\": 17310,\n    \"cells_with_multiple_nucleus_overlaps\": 336219,\n    \"cells_with_multiple_pairs\": 2864,\n    \"same_number_pairs\": ","encoding":"utf8","key":"manifest.json","name":"manifest.json","size":7416630,"truncated":true}],"sourceObjectKey":"crc-source-splits.csv","starterCell":"import pandas as pd\ndf = pd.read_csv(\"hf://datasets/csmp-hub/vitaminp-orion-crc/crc-source-splits.csv\")\ndf.head()","starterSamplePath":"crc-source-splits.csv","totalObjects":18},"summary":"VitaminP ORION-CRC is the colorectal cancer subset of the VitaminP release for cross-modal whole-cell segmentation in histopathology.","title":"VitaminP ORION-CRC: paired H&E and Hoechst immunofluorescence patches of colorectal cancer with nucleus and whole-cell masks","updatedAt":"2026-11-22T01:09:47.976Z","visibility":"public","websiteUrl":"https://zenodo.org/records/21224234"}}