Dental Panoramic X-Ray Detection Dataset (YOLO Format)

Combined dataset for dental pathology detection on panoramic radiographs, in YOLO format. Built for training liodon-ai/dental-panoramic-detector.

Classes

Dataset Sources

DENTEX provides: caries, periapical lesion, impacted tooth annotations on panoramic X-rays from 3 institutions.

OralXrays-9 provides: decay (→ caries), apical periodontitis (→ periapical_lesion), wisdom tooth (→ impacted_tooth) annotations across a hospital-scale panoramic X-ray collection.

Splits

Validation uses DENTEX val set only for consistent benchmarking.

Class Distribution (train)

Format

Standard YOLO format:

dental-panoramic-xray-yolo/
├── images/
│   ├── train/   # 9,928 images (.png / .jpg)
│   └── val/     # 46 images (.png)
├── labels/
│   ├── train/   # YOLO .txt per image
│   └── val/
└── dental.yaml

Each label file: one line per box — class cx cy w h (normalised 0–1).

Usage

from huggingface_hub import snapshot_download
path = snapshot_download(repo_id="liodon-ai/dental-panoramic-xray-yolo", repo_type="dataset")

Then point your YOLO config data: to {path}/dental.yaml.

Citation

If you use this dataset, please cite:

@dataset{liodonai2026dental,
  title={Dental Panoramic X-Ray Detection Dataset (YOLO Format)},
  author={Liodon AI},
  year={2026},
  publisher={Hugging Face},
  url={https://huggingface.co/datasets/liodon-ai/dental-panoramic-xray-yolo}
}

This dataset combines and remaps annotations from the following sources — please also cite them:

@inproceedings{hamamci2023dentex,
  title={DENTEX: An Abnormal Tooth Detection with Dental Enumeration and Diagnosis Benchmark for Panoramic X-Rays},
  author={Hamamci, Ibrahim Ethem and others},
  booktitle={MICCAI},
  year={2023}
}

@inproceedings{chen2025oralxrays9,
  title={OralXrays-9: Towards Hospital-Scale Panoramic X-ray Anomaly Detection via Personalized Multi-Object Query-Aware Mining},
  author={Chen, Bingzhi and others},
  booktitle={CVPR},
  year={2025}
}