STS-3D-Tooth: 371 dental cone-beam CT volumes for semi-supervised tooth segmentation (MICCAI STS challenge)
- Provider
- Yaqi Wang et al. (STS challenge organizers); mirrored by MedOtter
- License
- CC-BY-4.0 (source)
- Size
- 409 files
- Format
- NIfTI, Parquet (in sampled files)
- Stored at
- Hugging Face (https://huggingface.co/datasets/MedOtter/STS-3D-Tooth)
- Access
- Free and public. No account with the data provider is needed.
- DOI
- 10.5281/zenodo.10597292
- Category
- Health & Biomedical
- Source
- https://doi.org/10.1038/s41597-024-04306-9
About this dataset
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).
How to load this dataset in Python
from datasets import load_dataset
ds = load_dataset("MedOtter/STS-3D-Tooth")
print(ds)Data structure
File types in sampled files: NIfTI, Parquet
Files
Example file: data/integrity_labeled-00000-of-00001.parquet
Analyze this dataset without downloading it
On Scigantic, STS-3D-Tooth: 371 dental cone-beam CT volumes for semi-supervised tooth segmentation (MICCAI STS challenge) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.
This page is also available as Markdown and as JSON.
Related Health & Biomedical datasets
- DRAC22: Diabetic Retinopathy Analysis Challenge, ultra-wide optical coherence tomography angiography (UW-OCTA) images
- IDC CMB-MEL: Melanoma CT, whole-slide microscopy, PET, ultrasound (54 patients)
- IDC CCDI-MCI: Various whole-slide microscopy (4,407 patients)
- IDC TCGA-THYM: Thymoma whole-slide microscopy (124 patients)
- C-NMC 2019: 12,528 segmented B-lymphoblast cell images, leukemic (B-ALL) versus normal, from 101 subjects (ISBI 2019, TCIA)
- IDC COVID-19-AR: COVID-19 (non-cancer) radiographs, CT (105 patients)
- IDC TCGA-OV: Ovarian Serous Cystadenocarcinoma whole-slide microscopy, CT, MRI, other (591 patients)
- IDC TCGA-PCPG: Paraganglioma, Pheochromocytoma whole-slide microscopy (179 patients)