Automatic Classification and Segmentation of Teeth on 3D Dental Model Using Hierarchical Deep Learning Networks

Abstract

To solve the problem of low efficiency, the complexity of the interactive operation, and the high degree of manual intervention in existing methods, we propose a novel approach based on the sparse voxel octree and 3D convolution neural networks (CNNs) for segmenting and classifying tooth types on the 3D dental models. First, the tooth classification method capitalized on the two-level hierarchical feature learning is proposed to solve the misclassification problem in highly similar tooth categories. Second, we exploit an improved three-level hierarchical segmentation method based on the deep convolution features to conduct segmentation of teeth-gingiva and inter-teeth, respectively, and the conditional random field model is used to refine the boundary of the gingival margin and the inter-teeth fusion region. The experimental results show that the classification accuracy in Level_1 network is 95.96%, the average classification accuracy in Level_2 network is 88.06%, and the accuracy of tooth segmentation is 89.81%. Compared with the existing state-of-the-art methods, the proposed method has higher accuracy and universality, and it has great application potential in the computer-assisted orthodontic treatment diagnosis.

A Symmetric Fully Convolutional Residual Network With DCRF for Accurate Tooth Segmentation

“…We modified the original U-Net method [31] to achieve CBCT tooth segmentation. Not only that, we also compared with several latest learning-based segmentation methods [35], [37], [40] for tooth segmentation. The accuracy comparison between our proposed method and other excellent algorithms is shown in Table 3.…”

“…Miki et al [40] used AlexNet [44] network architecture and image data enhancement strategy to realize the classification of teeth, which can be used for automatic filing of dental records for forensic identification. Tian et al [35] presented a two-level hierarchical feature learning 3D tooth segmentation and classification model by combining sparse voxel octree and 3D CNNs, which solved the problem of misclassification of highly similar tooth categories. Cui et al [41] obtained the edge map of CBCT images firstly, and then proposed a new learning similarity matrix based on Mask R-CNN [45] to realize the automatic segmentation and recognition of tooth from CBCT images.…”

Clinically Applicable System For 3D Teeth Segmentation in Intraoral Scans using Deep Learning

“…Baseline and Evaluation Metrics. We select Xu's work as our baseline since it outperforms other deep learning models and traditional geometry-based methods [11][12][13] . We use three metrics to evaluate the segmentation results, i.e., the mean intersection over union (mIoU), the per-face accuracy and the average-area accuracy.…”

SRF‐Net: Spatial Relationship Feature Network for Tooth Point Cloud Classification

“…Eun et al [EK16] used CNN on Dental X‐ray Images. Tian et al [TDZ∗19] proposed to use an octree structure to represent a 3D tooth model and use CNN to extract features for tooth classification. With the use of 3D data in stomatology, such as scanning equipment, depth cameras, CBCT, more research on 3D data are proposed.…”

“…Multimodal dental data generally include the CBCT images of teeth and the scanning data of teeth inversions, the 3D model of teeth can be reconstructed and classified based on these data [MMH∗17, TDZ∗19, XLZ18, WLC∗17, WZL∗15, LWW∗20]. However, the task of tooth classification is challenging because of two reasons.…”