Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

STS MICCAI 2023 Challenge: Grand challenge on 2D and 3D semi-supervised tooth segmentation

Yaqi Wang, Yifan Zhang, Xiaodiao Chen, Shuai Wang, Dahong Qian, Fan Ye, Feng Xu, Hongyuan Zhang, Qianni Zhang, Chengyu Wu, Yunxiang Li, Weiwei Cui, Shan Luo, Chengkai Wang, Tianhao Li, Yi Liu, Xiang Feng, Huiyu Zhou, Dongyun Liu, Qixuan Wang, Zhouhao Lin, Wei Song, Yuanlin Li, Bing Wang, Chunshi Wang, Qiupu Chen, Mingqian Li

TL;DR

The STS MICCAI 2023 Challenge addresses the lack of open, annotated datasets for tooth segmentation in both 2D PXI and 3D CBCT by providing a semi-supervised benchmarking platform. Participants used limited labeled data with abundant unlabeled data, employing pseudo-labeling, consistency regularization, and multi-stage training to achieve state-of-the-art segmentation on both modalities, with top methods relying on nnUNet backbones and innovative data augmentation. The results demonstrate high Dice and mIoU scores in 2D and strong performance in 3D, while HD reveals edge alignment challenges and the need for robust post-processing. The study highlights the feasibility and value of semi-supervised approaches for dental image analysis and outlines directions for improving dataset diversity, artifact handling, and clinical translation.

Abstract

Computer-aided design (CAD) tools are increasingly popular in modern dental practice, particularly for treatment planning or comprehensive prognosis evaluation. In particular, the 2D panoramic X-ray image efficiently detects invisible caries, impacted teeth and supernumerary teeth in children, while the 3D dental cone beam computed tomography (CBCT) is widely used in orthodontics and endodontics due to its low radiation dose. However, there is no open-access 2D public dataset for children's teeth and no open 3D dental CBCT dataset, which limits the development of automatic algorithms for segmenting teeth and analyzing diseases. The Semi-supervised Teeth Segmentation (STS) Challenge, a pioneering event in tooth segmentation, was held as a part of the MICCAI 2023 ToothFairy Workshop on the Alibaba Tianchi platform. This challenge aims to investigate effective semi-supervised tooth segmentation algorithms to advance the field of dentistry. In this challenge, we provide two modalities including the 2D panoramic X-ray images and the 3D CBCT tooth volumes. In Task 1, the goal was to segment tooth regions in panoramic X-ray images of both adult and pediatric teeth. Task 2 involved segmenting tooth sections using CBCT volumes. Limited labelled images with mostly unlabelled ones were provided in this challenge prompt using semi-supervised algorithms for training. In the preliminary round, the challenge received registration and result submission by 434 teams, with 64 advancing to the final round. This paper summarizes the diverse methods employed by the top-ranking teams in the STS MICCAI 2023 Challenge.

STS MICCAI 2023 Challenge: Grand challenge on 2D and 3D semi-supervised tooth segmentation

TL;DR

The STS MICCAI 2023 Challenge addresses the lack of open, annotated datasets for tooth segmentation in both 2D PXI and 3D CBCT by providing a semi-supervised benchmarking platform. Participants used limited labeled data with abundant unlabeled data, employing pseudo-labeling, consistency regularization, and multi-stage training to achieve state-of-the-art segmentation on both modalities, with top methods relying on nnUNet backbones and innovative data augmentation. The results demonstrate high Dice and mIoU scores in 2D and strong performance in 3D, while HD reveals edge alignment challenges and the need for robust post-processing. The study highlights the feasibility and value of semi-supervised approaches for dental image analysis and outlines directions for improving dataset diversity, artifact handling, and clinical translation.

Abstract

Computer-aided design (CAD) tools are increasingly popular in modern dental practice, particularly for treatment planning or comprehensive prognosis evaluation. In particular, the 2D panoramic X-ray image efficiently detects invisible caries, impacted teeth and supernumerary teeth in children, while the 3D dental cone beam computed tomography (CBCT) is widely used in orthodontics and endodontics due to its low radiation dose. However, there is no open-access 2D public dataset for children's teeth and no open 3D dental CBCT dataset, which limits the development of automatic algorithms for segmenting teeth and analyzing diseases. The Semi-supervised Teeth Segmentation (STS) Challenge, a pioneering event in tooth segmentation, was held as a part of the MICCAI 2023 ToothFairy Workshop on the Alibaba Tianchi platform. This challenge aims to investigate effective semi-supervised tooth segmentation algorithms to advance the field of dentistry. In this challenge, we provide two modalities including the 2D panoramic X-ray images and the 3D CBCT tooth volumes. In Task 1, the goal was to segment tooth regions in panoramic X-ray images of both adult and pediatric teeth. Task 2 involved segmenting tooth sections using CBCT volumes. Limited labelled images with mostly unlabelled ones were provided in this challenge prompt using semi-supervised algorithms for training. In the preliminary round, the challenge received registration and result submission by 434 teams, with 64 advancing to the final round. This paper summarizes the diverse methods employed by the top-ranking teams in the STS MICCAI 2023 Challenge.
Paper Structure (27 sections, 5 equations, 7 figures, 7 tables)

This paper contains 27 sections, 5 equations, 7 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Semi-supervised Medical Image Segmentation
  4. Tooth Segmentation
  5. Challenge Description
  6. Organization
  7. Dataset
  8. 2D-PXI Dataset
  9. 3D-CBCT Dataset
  10. Performance Evaluation
  11. Challenge Submissions
  12. Submitted methods in Task 1: STS on 2D-PXI dataset
  13. Methods of Task2: STS on 3D-CBCT dataset
  14. Evaluation Results and Ranking Analysis
  15. Dice metric analysis of Task1
  16. ...and 12 more sections

Figures (7)

  • Figure 1: The workflow of the Semi-supervised Teeth Segmentation Challenge (STS) challenge consists of three stages: Preparation, Progression, and Conclusion.
  • Figure 2: Boxplot visualization (the first row) and statistical significance maps (the second row) for 2D tooth segmentation track. (a) and (d), (b) and (e), and (c) and (f) are the results for the Dice, mIoU and HD, respectively.
  • Figure 3: Boxplot visualization (the first row) and statistical significance maps (the second row) for 3D tooth segmentation track. (a) and (d), (b) and (e), and (c) and (f) are the results for the Dice, mIoU and HD, respectively.
  • Figure 4: Ranking stability analysis for the top 10 teams with individual and weighted metrics. (a) 2D tooth segmentation track, (b) 3DCBCT tooth segmentation track. "All Metric" is the weighted ranking of three metrics (Dice, mIoU and HD).
  • Figure 5: The illustration of examples in the dataset.
  • ...and 2 more figures