---
title: "DENTEX Dataset"
license: cc-by-nc-sa-4.0
---
<p align="center">
<img src="https://huggingface.co/datasets/ibrahimhamamci/DENTEX/resolve/main/figures/dentex.jpg?download=true" width="100%">
</p>
Welcome to the official page of the DENTEX dataset, which has been released as part of the [Dental Enumeration and Diagnosis on Panoramic X-rays Challenge (DENTEX)](https://dentex.grand-challenge.org/), organized in conjunction with the International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) in 2023. The primary objective of this challenge is to develop algorithms that can accurately detect abnormal teeth with dental enumeration and associated diagnosis. This not only aids in accurate treatment planning but also helps practitioners carry out procedures with a low margin of error.
The challenge provides three types of hierarchically annotated data and additional unlabeled X-rays for optional pre-training. The annotation of the data is structured using the Fédération Dentaire Internationale (FDI) system. The first set of data is partially labeled because it only includes quadrant info. The second set of data is also partially labeled but contains additional enumeration information along with the quadrant. The third set is fully labeled because it includes all quadrant-enumeration-diagnosis information for each abnormal tooth, and all participant algorithms have been benchmarked on this third set, with an example output shown below.
<p align="center">
<img src="https://huggingface.co/datasets/ibrahimhamamci/DENTEX/resolve/main/figures/output.png?download=true" width="100%">
</p>
## DENTEX Dataset
The DENTEX dataset comprises panoramic dental X-rays obtained from three different institutions using standard clinical conditions but varying equipment and imaging protocols, resulting in diverse image quality reflecting heterogeneous clinical practice. The dataset includes X-rays from patients aged 12 and above, randomly selected from the hospital's database to ensure patient privacy and confidentiality.
To enable effective use of the FDI system, the dataset is hierarchically organized into three types of data:
- (a) 693 X-rays labeled for quadrant detection and quadrant classes only,
- (b) 634 X-rays labeled for tooth detection with quadrant and tooth enumeration classes,
- (c) 1005 X-rays fully labeled for abnormal tooth detection with quadrant, tooth enumeration, and diagnosis classes.
The diagnosis class includes four specific categories: caries, deep caries, periapical lesions, and impacted teeth. An additional 1571 unlabeled X-rays are provided for pre-training.
<p align="center">
<img src="https://huggingface.co/datasets/ibrahimhamamci/DENTEX/resolve/main/figures/data.png?download=true" width="100%">
</p>
## Annotation Protocol
The DENTEX dataset provides three hierarchically annotated datasets to support various dental detection tasks: (1) quadrant-only for quadrant detection, (2) quadrant-enumeration for tooth detection, and (3) quadrant-enumeration-diagnosis for abnormal tooth detection. While offering a quadrant detection dataset might appear redundant, it's essential for effectively using the FDI Numbering System. This globally recognized system assigns numbers from 1 through 4 to each mouth quadrant: top right (1), top left (2), bottom left (3), and bottom right (4). Additionally, it numbers each of the eight teeth and each molar from 1 to 8, starting from the front middle tooth and increasing towards the back. For instance, the back tooth on the lower left side is designated as 48 in FDI notation, indicating quadrant 4, tooth 8. Thus, the quadrant segmentation dataset greatly simplifies the dental enumeration task, though evaluations are conducted only on the fully annotated third dataset.
## Data Split for Evaluation and Training
The DENTEX 2023 dataset comprises three types of data: (a) partially annotated quadrant data, (b) partially annotated quadrant-enumeration data, and (c) fully annotated quadrant-enumeration-diagnosis data. The first two types of data are intended for training and development purposes, while the third type is used for training and evaluations.
To comply with standard machine learning practices, the fully annotated third dataset, consisting of 1005 panoramic X-rays, is partitioned into training, validation, and testing subsets, comprising 705, 50, and 250 images, respectively. Ground truth labels are provided only for the training data, while the validation data is provided without associated ground truth. All the ground truth data is now available for researchers.
Note: The datasets are fully identical to the data used for our baseline method, named HierarchicalDet. For more information, please visit the [MICCAI paper](https://conferences.miccai.org/2023/papers/205-Paper2550.html) and the [GitHub repository](https://github.com/ibrahimethemhamamci/DENTEX) of HierarchicalDet (Diffusion-Based Hierarchical Multi-Label Object Detection to Analyze Panoramic Dental X-rays).
## Citing Us
If you use DENTEX, we would appreciate references to the following papers:
```
1. @article{hamamci2023dentex,
title={DENTEX: An Abnormal Tooth Detection with Dental Enumeration and Diagnosis Benchmark for Panoramic X-rays},
author={Hamamci, Ibrahim Ethem and Er, Sezgin and Simsar, Enis and Yuksel, Atif Emre and Gultekin, Sadullah and Ozdemir, Serife Damla and Yang, Kaiyuan and Li, Hongwei Bran and Pati, Sarthak and Stadlinger, Bernd and others},
journal={arXiv preprint arXiv:2305.19112},
year={2023}
}
2. @inproceedings{hamamci2023diffusion,
title={Diffusion-based hierarchical multi-label object detection to analyze panoramic dental x-rays},
author={Hamamci, Ibrahim Ethem and Er, Sezgin and Simsar, Enis and Sekuboyina, Anjany and Gundogar, Mustafa and Stadlinger, Bernd and Mehl, Albert and Menze, Bjoern},
booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
pages={389--399},
year={2023},
organization={Springer}
}
```
## License
We are committed to fostering innovation and collaboration in the research community. To this end, all elements of the DENTEX dataset are released under a [Creative Commons Attribution (CC-BY-NC-SA) license](https://creativecommons.org/licenses/by-nc-sa/4.0/). This licensing framework ensures that our contributions can be freely used for non-commercial research purposes, while also encouraging contributions and modifications, provided that the original work is properly cited and any derivative works are shared under similar terms.
---
title: "DENTEX数据集"
license: cc-by-nc-sa-4.0
---
<p align="center">
<img src="https://huggingface.co/datasets/ibrahimhamamci/DENTEX/resolve/main/figures/dentex.jpg?download=true" width="100%">
</p>
欢迎来到DENTEX数据集的官方页面,本数据集作为2023年与国际医学图像计算与计算机辅助干预会议(MICCAI)联合举办的**全景牙科X射线牙齿计数与诊断挑战赛(Dental Enumeration and Diagnosis on Panoramic X-rays Challenge, DENTEX)**的配套资源正式发布。本次挑战赛的核心目标是开发可精准检测异常牙齿、完成牙齿计数并给出对应诊断的算法,该技术不仅有助于制定精准的治疗方案,还能帮助临床从业者降低操作误差。
本次挑战赛提供三类层级标注数据集,以及额外的未标注X射线影像供选手进行可选预训练。数据集的标注采用国际牙科联盟(Fédération Dentaire Internationale, FDI)标准。第一类数据为部分标注数据集,仅包含牙象限信息;第二类数据同样为部分标注,但额外包含牙象限与牙齿计数信息;第三类为全标注数据集,涵盖每颗异常牙齿的象限、计数与诊断信息,所有参赛算法均在第三套数据集上完成基准测试,示例输出如下。
<p align="center">
<img src="https://huggingface.co/datasets/ibrahimhamamci/DENTEX/resolve/main/figures/output.png?download=true" width="100%">
</p>
## DENTEX数据集
DENTEX数据集包含来自三家不同机构的全景牙科X射线影像,采集时均采用标准临床流程,但设备与成像协议存在差异,因此图像质量呈现多样性,反映了真实临床实践的异质性。数据集选取12岁及以上患者的X射线影像,从医院数据库中随机抽取,以充分保障患者隐私与信息保密性。
为适配FDI标注系统,数据集按层级划分为三类:
- (a) 693张仅标注牙象限检测与象限类别的X射线影像;
- (b) 634张标注了牙齿检测、牙象限与牙齿计数类别的X射线影像;
- (c) 1005张针对异常牙齿检测完成全标注的X射线影像,包含牙象限、牙齿计数与诊断类别。
诊断类别包含四类具体场景:龋病、深龋、根尖周病变与阻生齿。此外还提供1571张未标注X射线影像用于预训练。
<p align="center">
<img src="https://huggingface.co/datasets/ibrahimhamamci/DENTEX/resolve/main/figures/data.png?download=true" width="100%">
</p>
## 标注规范
DENTEX数据集提供三类层级标注数据集,以支持不同的牙科检测任务:(1) 仅牙象限数据集用于牙象限检测;(2) 牙象限-计数数据集用于牙齿检测;(3) 牙象限-计数-诊断数据集用于异常牙齿检测。尽管仅提供牙象限检测数据集看似冗余,但对于有效使用FDI编号系统至关重要。这套全球通用的编号系统为口腔四个象限分配1至4的编号:右上(1)、左上(2)、左下(3)、右下(4)。同时,从中切牙开始向后方依次为1至8,标注每颗恒牙。例如,左下后方的磨牙在FDI标注体系中记为48,代表象限4、牙齿8。因此,牙象限分割数据集可极大简化牙齿计数任务,不过模型评估仅在全标注的第三套数据集上进行。
## 训练与评估数据拆分
DENTEX 2023数据集包含三类数据:(a) 部分标注的牙象限数据、(b) 部分标注的牙象限-计数数据、(c) 全标注的牙象限-计数-诊断数据。前两类数据仅用于模型训练与开发,第三类数据则用于模型训练与评估。
遵循标准机器学习实践,包含1005张全景X射线的全标注第三套数据集被划分为训练、验证与测试子集,分别包含705、50与250张影像。仅训练数据提供真值标签,验证数据不附带真值标签。目前所有真值标签数据均已对研究人员开放。
> 注:本数据集与我们的基线方法HierarchicalDet所使用的数据完全一致。如需了解更多信息,请访问HierarchicalDet的[MICCAI论文](https://conferences.miccai.org/2023/papers/205-Paper2550.html)与[GitHub仓库](https://github.com/ibrahimethemhamamci/DENTEX)(基于扩散的层级多标签目标检测算法用于全景牙科X射线分析)。
## 引用声明
若您使用DENTEX数据集,请引用以下论文:
1. @article{hamamci2023dentex,
title={DENTEX: 面向全景牙科X射线的异常牙齿检测、牙齿计数与诊断基准数据集},
author={Hamamci, Ibrahim Ethem and Er, Sezgin and Simsar, Enis and Yuksel, Atif Emre and Gultekin, Sadullah and Ozdemir, Serife Damla and Yang, Kaiyuan and Li, Hongwei Bran and Pati, Sarthak and Stadlinger, Bernd 等},
journal={arXiv预印本arXiv:2305.19112},
year={2023}
}
2. @inproceedings{hamamci2023diffusion,
title={基于扩散的层级多标签目标检测算法用于全景牙科X射线分析},
author={Hamamci, Ibrahim Ethem and Er, Sezgin and Simsar, Enis and Sekuboyina, Anjany and Gundogar, Mustafa and Stadlinger, Bernd and Mehl, Albert and Menze, Bjoern},
booktitle={国际医学图像计算与计算机辅助干预会议},
pages={389--399},
year={2023},
organization={Springer}
}
## 许可协议
我们致力于推动研究社区的创新与合作。为此,DENTEX数据集的所有内容均采用[知识共享署名-非商业性使用-相同方式共享(Creative Commons Attribution-NonCommercial-ShareAlike, CC-BY-NC-SA)许可协议](https://creativecommons.org/licenses/by-nc-sa/4.0/)发布。该许可框架确保我们的成果可免费用于非商业研究目的,同时鼓励对原工作进行引用与修改,且衍生作品需采用相同许可协议共享。