A perspective on the diagnosis of cracked tooth: imaging modalities evolve to AI-based analysis

Semantic segmentation for tooth cracks using improved DeepLabv3+ model

Objective

Accurate and prompt detection of cracked teeth plays a critical role for human oral health. The aim of this paper is to evaluate the performance of a tooth crack segmentation model (namely, FDB-DeepLabv3+) on optical microscopic images.

Method

The FDB-DeepLabv3+ model proposed here improves feature learning by replacing the backbone with ResNet50. Feature pyramid network (FPN) is introduced to fuse muti-level features. Densely linked atrous spatial pyramid pooling (Dense ASPP) is applied to achieve denser pixel sampling and wider receptive field. Bottleneck attention module (BAM) is embedded to enhance local feature extraction.

Results

Through testing on a self-made hidden cracked tooth dataset, the proposed method outperforms four classical networks (FCN, U-Net, SegNet, DeepLabv3+) on segmentation results in terms of mean pixel accuracy (MPA) and mean intersection over union (MIoU). The network achieves an increase of 11.41% in MPA and 12.14% in MIoU compared to DeepLabv3+. Ablation experiments shows that all the modifications are beneficial.

Conclusion

An improved network is designed for segmenting tooth surface cracks with good overall performance and robustness, which may hold significant potential in computer-aided diagnosis of cracked teeth.

Influence of restorative materials on the mechanical properties of maxillary first molars with different degrees of cryptic fractures and defects: A finite element analysis

This study aimed to apply finite element analysis to evaluate the effects of pile materials with different elastic moduli and cement materials on the stress distribution between the remaining tooth tissue and cryptic fracture defects. A three-dimensional finite element model was established for 20 maxillary first molars with hidden fissures and mesial tongue-tip defects. Two levels of hidden cracks and three types of pile and adhesive materials were used in the design. The stress distribution and maximum stress peak in the remaining tooth tissue and crack defects were determined by simulating the normal bite, maximum bite, and lateral movement forces. When titanium posts, zinc phosphate binders, and porcelain crowns were used to repair the two types of deep cracked teeth, the maximum principal stress at the crack and dentin was the smallest. As the crack depth increased, the maximum principal stress of the residual dentin and crack defects increased.

An evaluation of quantitative percussion diagnostics for determining the probability of a microgap defect in restored and unrestored teeth: A prospective clinical study

STATEMENT OF PROBLEM

Current dental diagnostics are image based and cannot detect a structural microgap defect such as a crack in a tooth. Whether percussion diagnostics can effectively diagnose a microgap defect is unclear.

PURPOSE

The purpose of the present study was to determine from a large multicenter prospective clinical study whether quantitative percussion diagnostics (QPD) could detect structural damage in teeth and whether a probability of its presence could be provided.

MATERIAL AND METHODS

A nonrandomized prospective and multicenter clinical validation study with 224 participants was performed in 5 centers with 6 independent investigators. The study used QPD and the normal fit error to determine whether a microgap defect was present in a natural tooth. Teams 1 and 2 were blinded. Team 1 tested teeth scheduled for restoration with QPD, and Team 2 disassembled the teeth aided by a clinical microscope, transillumination, and a penetrant dye. Microgap defects were documented in written and video formats. Controls were participants without damaged teeth. The percussion response from each tooth was stored on a computer and analyzed. A total of 243 teeth were tested to provide approximately 95% power to test the performance goal of 70%, based on an assumed population overall agreement of 80%.

RESULTS

Regardless of the collection method, tooth geometry, restoration material used, or restoration type, the data on detecting a microgap defect in a tooth were accurate. The data also reflected good sensitivity and specificity consistent with previously published clinical studies. The combined study data showed an overall agreement of 87.5% with a 95% confidence interval (84.2 to 90.3), beyond the 70% predetermined performance goal. The combined study data determined whether it was possible to predict the probability of a microgap defect.

CONCLUSIONS

The results showed that the data on detecting microgap defects in a tooth site were consistently accurate and confirmed that QPD provided information to aid the clinician in treatment planning and early preventative treatment. QPD can also alert the clinician of probable diagnosed and undiagnosed structural problems via the use of a probability curve.

Automatic Detection of Cracks in Cracked Tooth Based on Binary Classification Convolutional Neural Networks

Cracked tooth syndrome is a commonly encountered disease in dentistry, which is often accompanied by dramatic painful responses from occlusion and temperature stimulation. Current clinical diagnostic trials include traditional methods (such as occlusion test, probing, cold stimulation, etc.) and X-rays based medical imaging (periapical radiography (PR), cone-beam computed tomography (CBCT), etc.). However, these methods are strongly dependent on the experience of the clinicians, and some inconspicuous cracks are also extremely easy to be overlooked by visual observation, which will definitely affect the subsequent treatments. Inspired by the achievements of applying deep convolutional neural networks (CNNs) in crack detection in engineering, this article proposes an image-based crack detection method using a deep CNN classifier in combination with a sliding window algorithm. A CNN model is designed by modifying the size of the input layer and adding a fully connected layer with 2 units based on the ResNet50, and then, the proposed CNN is trained and validated with a self-prepared cracked tooth dataset including 20,000 images. By comparing validation accuracy under seven different learning rates, 10⁻⁵ is chosen as the best learning rate for the following testing process. The trained CNN is tested on 100 images with 1920 × 1080-pixel resolutions, which achieves an average accuracy of 90.39%. The results show that the proposed method can effectively detect cracks in images under various conditions (stained, overexplosion, images affected by other diseases). The proposed method in this article provides doctors with a more intelligent diagnostic solution, and it is not only suitable for optical photographs but also for automated diagnosis of other medical imaging images.