{"success":true,"data":{"author":"Yaqi Wang et al. (STS challenge organizers); mirrored by MedOtter","bucketAccessType":"public","bucketName":"MedOtter/STS-3D-Tooth","bucketPrefix":"","category":"Health & Biomedical","cloudProvider":"huggingface","createdAt":"2026-11-17T19:29:33.534Z","description":"The 3D cone-beam CT (CBCT) part of the STS (Semi-supervised Teeth Segmentation) multimodal dental dataset used in the MICCAI 2023 and 2024 STS challenges. It contains 371 CBCT volumes of 512x512x400 in NIfTI: 32 labeled and 339 unlabeled. Two subsets differ in field of view and labels: Integrity/ (whole head and jaw scans, 10 labeled with a binary teeth mask and 231 unlabeled, intensities pre-normalized to 0 to 1) and ROI/ (crops centered on the dental arch, 22 labeled with per-tooth instance labels in FDI numbering and 108 unlabeled, raw HU-like int16 at about 0.15 mm isotropic spacing). The companion 2D panoramic X-ray subset is a separate repository. About 31 GB (measured 2026-11-17). Use for semi-supervised 3D tooth segmentation and per-tooth instance labeling in CBCT; load with nibabel. Mirrored on Hugging Face by MedOtter from the authors' release, with a card documenting provenance and any normalization. The authors' Zenodo record 10597292 is CC BY 4.0; cite Wang et al., Scientific Data 12, 117 (2025).","displayTitle":"STS-3D-Tooth: 371 dental cone-beam CT volumes for semi-supervised tooth segmentation (MICCAI STS challenge)","doi":"10.5281/zenodo.10597292","huggingFaceUrl":"https://huggingface.co/datasets/MedOtter/STS-3D-Tooth","id":"44669743-55b0-5ded-9f6f-2e4cbe7630f0","license":"CC-BY-4.0","licenseEvidenceUrl":"https://zenodo.org/records/10597292","objectCountEstimate":409,"probeStatus":"ok","referenceUrl":"https://doi.org/10.1038/s41597-024-04306-9","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"MedOtter/STS-3D-Tooth\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"MedOtter/STS-3D-Tooth\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=MedOtter, license:cc-by-4.0, size_categories:n<1K, format:parquet, downloads=1293"],"format":"mixed","generatedAt":"2026-11-17T19:40:32.908Z","layout":{"example":"data/integrity_labeled-00000-of-00001.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"data/integrity_labeled-00000-of-00001.parquet","truncatedAt":512}],"sampleObjects":["Integrity/Labeled/Image/Integrity_L_001.nii.gz","Integrity/Labeled/Image/Integrity_L_002.nii.gz","Integrity/Labeled/Image/Integrity_L_003.nii.gz","Integrity/Labeled/Image/Integrity_L_004.nii.gz","Integrity/Labeled/Image/Integrity_L_005.nii.gz"],"sidecars":[{"bytes":"---\nlicense: cc-by-4.0\ntask_categories:\n  - image-segmentation\nmodality:\n  - CBCT\nlanguage: []\ntags:\n  - medical-imaging\n  - dental\n  - tooth-segmentation\n  - cbct\n  - 3d-segmentation\n  - semi-supervised\npretty_name: STS-3D-Tooth\nsize_categories:\n  - n<1K\nconfigs:\n  - config_name: default\n    data_files:\n      - split: integrity_labeled\n        path: data/integrity_labeled-*\n      - split: integrity_unlabeled\n        path: data/integrity_unlabeled-*\n      - split: roi_labeled\n        path: data/roi_labeled-*\n      - split: roi_unlabeled\n        path: data/roi_unlabeled-*\n---\n\n# STS-3D-Tooth\n\nThe 3D Cone-Beam CT (CBCT) subset of the STS (Semi-supervised Teeth\nSegmentation) multi-modal dental dataset, as released in\n[Wang et al., *Scientific Data* **12**, 117 (2025)](https://doi.org/10.1038/s41597-024-04306-9)\nand used in the MICCAI 2023/2024 STS Challenges.\n\nThe companion 2D panoramic X-ray subset is hosted at\n[`Angelou0516/STS-2D-Tooth`](https://huggingface.co/datasets/Angelou0516/STS-2D-Tooth).\n\n## Dataset Summary\n\n| Field | Details |\n|---|---|\n| Modality | Cone-Beam CT (CBCT), NIfTI (`.nii.gz`) |\n| Body Part | Teeth (32 permanent teeth, FDI numbering) |\n| Volumes | 371 total: 32 labeled, 339 unlabeled |\n| Volume shape | 512 x 512 x 400 (consistent across all volumes) |\n| License | CC-BY-4.0 |\n| Source | https://zenodo.org/records/10597292 |\n\n## Subsets\n\nThe release ships two distinct CBCT subsets that differ in field-of-view, intensity\nrepresentation, and label semantics. They are **not interchangeable**.\n\n### Integrity (whole-FOV scan)\n\nFull head/jaw CBCT acquisitions captured in their original field of view.\n\n- 10 labeled volumes (image + binary tooth-vs-background mask)\n- 231 unlabeled volumes (image only, no GT)\n- Image dtype: `float32`, intensity range `[0.0, 1.0]` (pre-normalized to unit interval)\n- Affine: identity (`1.0 x 1.0 x 1.0` voxel spacing) — true acquisition spacing is **not** preserved\n- Mask labels: `{0, 1}` — binary foreground = teeth\n\n### ROI (tooth-region crop)\n\nCropped sub-volumes centered on the dental arch.\n\n- 22 labeled volumes (image + multi-class instance mask)\n- 108 unlabeled volumes (image only, no GT)\n- Image dtype: `int16`, intensity range approximately `[-1000, 3095]` (raw HU-like)\n- Affine: real isotropic spacing, approximately `0.156 x 0.156 x 0.150` mm\n- Mask labels: integer class indices > 0 — per-tooth instance labels following the FDI\n  numbering convention (not all 32 teeth appear in every volume)\n\n| Subset    | Total | Labeled | Unlabeled |\n|-----------|------:|--------:|----------:|\n| Integrity |   241 |      10 |       231 |\n| ROI       |   130 |      22 |       108 |\n| **Total** |   371 |      32 |       339 |\n\n## Recommended Ground Truth\n\nAnnotation pipeline (per the source paper):\n1. 10 junior dentists each annotated CBCT scans layer-by-layer in ITK-SNAP.\n2. 3 senior dentists (>10 years experience) reviewed and corrected the layer-wise\n   annotations.\n3. Remaining inter-reviewer discrepancies were resolved by consensus.\n\nThe shipped masks are post-consensus refined and reflect senior-expert agreement.\nThere is no alternative mask source.\n\n## Data Structure\n\n```\nSTS-3D-Tooth/\n|-- README.md\n|-- Integrity/\n|   |-- Labeled/\n|   |   |-- Image/Integrity_L_NNN.nii.gz   # 10 CBCT volumes (float32, [0,1])\n|   |   `-- Mask/Integrity_L_NNN.nii.gz    # 10 binary masks\n|   `-- Unlabeled/\n|       `-- Image/Integrity_U_NNN.nii.gz   # 231 unlabeled CBCT volumes\n`-- ROI/\n    |-- Labeled/\n    |   |-- Image/ROI_L_NNN.nii.gz         # 22 CBCT crops (int16, raw HU-like)\n    |   `-- Mask/ROI_L_NNN.nii.gz          # 22 multi-class FDI instance masks\n    `-- Unlabeled/\n        `-- Image/ROI_U_NNN.nii.gz         # 108 unlabeled CBCT crops\n```\n\nImage and mask filenames match exactly within each `Labeled/` directory.\n\n## Splits\n\nThe released dataset has **no official train/val/test split** — define your own\ndownstream. The labeled / unlabeled distinction is intrinsic to the\nsemi-supervised challenge format; the unlabeled volumes have no ground truth and\nare typically used only for self-trainin","encoding":"utf8","key":"README.md","name":"README.md","size":4783,"truncated":true}],"sourceObjectKey":"data/integrity_labeled-00000-of-00001.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"MedOtter/STS-3D-Tooth\")\nprint(ds)","starterSamplePath":"Integrity/Labeled/Image/Integrity_L_001.nii.gz","totalObjects":409},"summary":"The 3D cone-beam CT (CBCT) part of the STS (Semi-supervised Teeth Segmentation) multimodal dental dataset used in the MICCAI 2023 and 2024 STS challenges.","title":"STS-3D-Tooth: 371 dental cone-beam CT volumes for semi-supervised tooth segmentation (MICCAI STS challenge)","updatedAt":"2026-11-17T21:40:19.671Z","visibility":"public","websiteUrl":"https://huggingface.co/datasets/MedOtter/STS-3D-Tooth"}}