STS-2D-Tooth: 4,000 panoramic dental X-rays (PXI) of adults and children, 900 with tooth masks (MICCAI STS challenge)

Provider
Wang et al. (STS challenge organizers); repackaged by MedOtter
License
CC-BY-4.0 (source)
Size
6 files
Format
Parquet (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/MedOtter/STS-2D-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

STS-2D-Tooth is the two-dimensional half of the STS (Semi-supervised Teeth Segmentation) dental dataset used in the MICCAI 2023 STS challenge: panoramic dental X-ray images (orthopantomograms, PXI) for segmenting teeth with few labels. It holds 4,000 radiographs at 640x320 pixels, 3,500 from adults (850 labeled) and 500 from children (50 labeled), so 900 have binary tooth-region masks and 3,100 are unlabeled for semi-supervised learning. Masks were seeded by 20 trained dental practitioners on 300 images, refined with an R2 U-Net assistant and checked by six dentists. Four Parquet splits under data/ (a_pxi_labeled, a_pxi_unlabeled, c_pxi_labeled, c_pxi_unlabeled), about 0.9 GB (measured 2026-11-21); read with Hugging Face datasets. Use for tooth segmentation on panoramic radiographs and pediatric versus adult dentition studies. The companion cone-beam CT set STS-3D-Tooth is listed separately. Released by the authors on Zenodo (DOI 10.5281/zenodo.10597292) under CC BY 4.0. Cite Wang et al., Scientific Data 12:117 (2025), doi 10.1038/s41597-024-04306-9.

How to load this dataset in Python

from datasets import load_dataset
ds = load_dataset("MedOtter/STS-2D-Tooth")
print(ds)

Data structure

File types in sampled files: Parquet

Files

Example file: data/a_pxi_labeled-00000-of-00001.parquet

Analyze this dataset without downloading it

On Scigantic, STS-2D-Tooth: 4,000 panoramic dental X-rays (PXI) of adults and children, 900 with tooth masks (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