Unsupervised Pre-training Improves Tooth Segmentation in 3-Dimensional Intraoral Mesh Scans

Xiaoxuan He, Hualiang Wang, Haoji Hu, Jianfei Yang, Yang Feng, Gaoang Wang, Liu Zuozhu
Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, PMLR 172:493-507, 2022.

Abstract

Accurate tooth segmentation in 3-Dimensional (3D) intraoral scanned (IOS) mesh data is an essential step for many practical dental applications. Recent research highlights the success of deep learning based methods for end-to-end 3D tooth segmentation, yet most of them are only trained or validated with a small dataset as annotating 3D IOS dental surfaces requires complex pipelines and intensive human efforts. In this paper, we propose a novel method to boost the performance of 3D tooth segmentation leveraging large-scale unlabeled IOS data. Our tooth segmentation network is first pre-trained with an unsupervised learning framework and point-wise contrastive learning loss on the large-scale unlabeled dataset and subsequently fine-tuned on a small labeled dataset. With the same amount of annotated samples, our method can achieve a mIoU of 89.38%, significantly outperforming the supervised counterpart. Moreover, our method can achieve better performance with only 40% of the annotated samples as compared to the fully supervised baselines. To the best of our knowledge, we present the first attempt of unsupervised pretraining for 3D tooth segmentation, demonstrating its strong potential in reducing human efforts for annotation and verification.

Cite this Paper

BibTeX
@InProceedings{pmlr-v172-he22b, title = {Unsupervised Pre-training Improves Tooth Segmentation in 3-Dimensional Intraoral Mesh Scans}, author = {He, Xiaoxuan and Wang, Hualiang and Hu, Haoji and Yang, Jianfei and Feng, Yang and Wang, Gaoang and Zuozhu, Liu}, booktitle = {Proceedings of The 5th International Conference on Medical Imaging with Deep Learning}, pages = {493--507}, year = {2022}, editor = {Konukoglu, Ender and Menze, Bjoern and Venkataraman, Archana and Baumgartner, Christian and Dou, Qi and Albarqouni, Shadi}, volume = {172}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v172/he22b/he22b.pdf}, url = {https://proceedings.mlr.press/v172/he22b.html}, abstract = {Accurate tooth segmentation in 3-Dimensional (3D) intraoral scanned (IOS) mesh data is an essential step for many practical dental applications. Recent research highlights the success of deep learning based methods for end-to-end 3D tooth segmentation, yet most of them are only trained or validated with a small dataset as annotating 3D IOS dental surfaces requires complex pipelines and intensive human efforts. In this paper, we propose a novel method to boost the performance of 3D tooth segmentation leveraging large-scale unlabeled IOS data. Our tooth segmentation network is first pre-trained with an unsupervised learning framework and point-wise contrastive learning loss on the large-scale unlabeled dataset and subsequently fine-tuned on a small labeled dataset. With the same amount of annotated samples, our method can achieve a mIoU of 89.38%, significantly outperforming the supervised counterpart. Moreover, our method can achieve better performance with only 40% of the annotated samples as compared to the fully supervised baselines. To the best of our knowledge, we present the first attempt of unsupervised pretraining for 3D tooth segmentation, demonstrating its strong potential in reducing human efforts for annotation and verification.} }
Endnote
%0 Conference Paper %T Unsupervised Pre-training Improves Tooth Segmentation in 3-Dimensional Intraoral Mesh Scans %A Xiaoxuan He %A Hualiang Wang %A Haoji Hu %A Jianfei Yang %A Yang Feng %A Gaoang Wang %A Liu Zuozhu %B Proceedings of The 5th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2022 %E Ender Konukoglu %E Bjoern Menze %E Archana Venkataraman %E Christian Baumgartner %E Qi Dou %E Shadi Albarqouni %F pmlr-v172-he22b %I PMLR %P 493--507 %U https://proceedings.mlr.press/v172/he22b.html %V 172 %X Accurate tooth segmentation in 3-Dimensional (3D) intraoral scanned (IOS) mesh data is an essential step for many practical dental applications. Recent research highlights the success of deep learning based methods for end-to-end 3D tooth segmentation, yet most of them are only trained or validated with a small dataset as annotating 3D IOS dental surfaces requires complex pipelines and intensive human efforts. In this paper, we propose a novel method to boost the performance of 3D tooth segmentation leveraging large-scale unlabeled IOS data. Our tooth segmentation network is first pre-trained with an unsupervised learning framework and point-wise contrastive learning loss on the large-scale unlabeled dataset and subsequently fine-tuned on a small labeled dataset. With the same amount of annotated samples, our method can achieve a mIoU of 89.38%, significantly outperforming the supervised counterpart. Moreover, our method can achieve better performance with only 40% of the annotated samples as compared to the fully supervised baselines. To the best of our knowledge, we present the first attempt of unsupervised pretraining for 3D tooth segmentation, demonstrating its strong potential in reducing human efforts for annotation and verification.
APA
He, X., Wang, H., Hu, H., Yang, J., Feng, Y., Wang, G. & Zuozhu, L.. (2022). Unsupervised Pre-training Improves Tooth Segmentation in 3-Dimensional Intraoral Mesh Scans. Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 172:493-507 Available from https://proceedings.mlr.press/v172/he22b.html.

Related Material