{"success":true,"data":{"author":"Broad Institute (Caicedo et al., BBBC038); COCO conversion by DataTorch","bucketAccessType":"public","bucketName":"DataTorchio/nuclei-segmentation-dsb2018","bucketPrefix":"","category":"Life Sciences","cloudProvider":"huggingface","createdAt":"2026-11-17T14:12:21.894Z","description":"The stage 1 training set of the 2018 Data Science Bowl, image set BBBC038v1 from the Broad Bioimage Benchmark Collection: 670 microscopy images with 29,430 individually segmented nuclei. Images span fluorescence, brightfield and histology stains at several magnifications, from human, mouse and fly samples, and were assembled to train generalist nucleus segmentation models. In this copy the official per-nucleus masks are converted to polygon annotations in one COCO-format annotations.json (12 of 29,442 masks were too small to outline and were dropped), with the images under images/; about 70 MB. Load with pycocotools, torchvision or any COCO-compatible detection and instance segmentation tool, for example to train Mask R-CNN, Cellpose or StarDist style models. The copy is maintained by DataTorch, an annotation platform. The Broad Institute releases BBBC038v1 under CC0 (contributors have waived copyright); cite Caicedo et al., \"Nucleus segmentation across imaging experiments: the 2018 Data Science Bowl\", Nature Methods 16, 1247-1253 (2019).","displayTitle":"2018 Data Science Bowl nuclei segmentation (BBBC038v1) training set as COCO polygons","huggingFaceUrl":"https://huggingface.co/datasets/DataTorchio/nuclei-segmentation-dsb2018","id":"73114391-7cab-52d9-b6e5-e0c2e52f936f","license":"CC0-1.0","licenseEvidenceUrl":"https://bbbc.broadinstitute.org/BBBC038","objectCountEstimate":673,"probeStatus":"ok","referenceUrl":"https://doi.org/10.1038/s41592-019-0612-7","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"DataTorchio/nuclei-segmentation-dsb2018\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"DataTorchio/nuclei-segmentation-dsb2018\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=DataTorchio, license:cc0-1.0, size_categories:n<1K, format:imagefolder, downloads=2627"],"format":"mixed","generatedAt":"2026-11-17T14:23:15.174Z","layout":{"example":"images/00071198d059ba7f5914a526d124d28e6d010c92466da21d4a04cd5413362552.png","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"images/00071198d059ba7f5914a526d124d28e6d010c92466da21d4a04cd5413362552.png","truncatedAt":512}],"sampleObjects":["annotations.json","images/00071198d059ba7f5914a526d124d28e6d010c92466da21d4a04cd5413362552.png","images/003cee89357d9fe13516167fd67b609a164651b21934585648c740d2c3d86dc1.png","images/00ae65c1c6631ae6f2be1a449902976e6eb8483bf6b0740d00530220832c6d3e.png","images/0121d6759c5adb290c8e828fc882f37dfaf3663ec885c663859948c154a443ed.png"],"sidecars":[{"bytes":"---\nlicense: cc0-1.0\ntask_categories:\n  - image-segmentation\ntags:\n  - microscopy\n  - nuclei\n  - instance-segmentation\n  - biology\n  - coco\npretty_name: \"Nuclei Segmentation (DSB2018, BBBC038)\"\nsize_categories:\n  - n<1K\n---\n\n# Nuclei Segmentation (DSB2018, BBBC038)\n\nThe training set of the 2018 Data Science Bowl: 670 microscopy images with 29,430 individually\nsegmented nuclei, as polygon annotations in COCO format. Images span fluorescence and\nbrightfield or histology stains, several magnifications, and samples from humans, mice and\nflies.\n\n**Browse it before you download it.** Every image, with every nucleus drawn on it, is viewable\nin the browser at https://datatorch.io/datatorch/nuclei-segmentation-dsb2018. The annotations here are the official per-nucleus masks converted to polygons (12 of the\n29,442 masks were too small to outline and were dropped).\n\n## Contents\n\n| | |\n|---|---|\n| Images | 670 (stage1_train) |\n| Nuclei | 29,430, one polygon each |\n| Format | images plus one COCO annotations JSON |\n| License | CC0 1.0 (public domain) |\n| Source | https://bbbc.broadinstitute.org/BBBC038 |\n\n## Citation\n\nWe used image set BBBC038v1, available from the Broad Bioimage Benchmark Collection.\n\nCaicedo, J.C. et al. Nucleus segmentation across imaging experiments: the 2018 Data Science\nBowl. Nature Methods 16, 1247-1253 (2019).\n\n## About this mirror\n\nMirrored by DataTorch, an annotation platform. A step-by-step tutorial that uses this dataset is\nat https://datatorch.io/resources/nuclei-segmentation-tutorial, and a description of the\ndataset is at https://datatorch.io/resources/nuclei-segmentation-dataset.\n","encoding":"utf8","key":"README.md","name":"README.md","size":1627,"truncated":false}],"sourceObjectKey":"images/00071198d059ba7f5914a526d124d28e6d010c92466da21d4a04cd5413362552.png","starterCell":"from huggingface_hub import hf_hub_download\npath = hf_hub_download(repo_id=\"DataTorchio/nuclei-segmentation-dsb2018\", filename=\"images/00071198d059ba7f5914a526d124d28e6d010c92466da21d4a04cd5413362552.png\", repo_type=\"dataset\")\nprint(path)","starterSamplePath":"images/00071198d059ba7f5914a526d124d28e6d010c92466da21d4a04cd5413362552.png","totalObjects":673},"summary":"The stage 1 training set of the 2018 Data Science Bowl, image set BBBC038v1 from the Broad Bioimage Benchmark Collection: 670 microscopy images with 29,430 individually segmented…","title":"2018 Data Science Bowl nuclei segmentation (BBBC038v1) training set as COCO polygons","updatedAt":"2026-11-17T20:11:31.830Z","visibility":"public","websiteUrl":"https://huggingface.co/datasets/DataTorchio/nuclei-segmentation-dsb2018"}}