Use of the deep learning approach to measure alveolar bone level

Abstract

Aim: The goal was to use a deep convolutional neural network to measure the radiographic alveolar bone level to aid periodontal diagnosis. Materials and Methods: A deep learning (DL) model was developed by integrating three segmentation networks (bone area, tooth, cemento-enamel junction) and image analysis to measure the radiographic bone level and assign radiographic bone loss (RBL) stages. The percentage of RBL was calculated to determine the stage of RBL for each tooth. A provisional periodontal diagnosis was assigned using the 2018 periodontitis classification. RBL percentage, staging, and presumptive diagnosis were compared with the measurements and diagnoses made by the independent examiners. Results:The average Dice Similarity Coefficient (DSC) for segmentation was over 0.91. There was no significant difference in the RBL percentage measurements determined by DL and examiners (p ¼ :65). The area under the receiver operating characteristics curve of RBL stage assignment for stages I, II, and III was 0.89, 0.90, and 0.90, respectively. The accuracy of the case diagnosis was 0.85. Conclusions:The proposed DL model provides reliable RBL measurements and image-based periodontal diagnosis using periapical radiographic images. However, this model has to be further optimized and validated by a larger number of images to facilitate its application.

Artificial Intelligence in Dentistry—Narrative Review

“…In dental surgery, neural networks may be helpful in orthognathic surgery planning, prediction of post-extraction complications, bone lesion detection, and differentiation and implantological treatment planning [ 41 , 42 , 43 , 44 , 45 , 46 , 47 ]. Furthermore, artificial intelligence is spreading into periodontology and in the above-mentioned studies, it was used to evaluate the periodontal bone loss, peri-implant bone loss, and to predict the development of periodontitis due to the psychological features [ 49 , 50 , 51 , 52 , 53 , 54 ]. This review shows that artificial intelligence has developed very fast in recent years and it may become an ordinary tool in modern dentistry in the near future.…”

“…The difficulties in osteointegration might occur due to the presence of a soft tissue layer (non-mineralized bone tissue) around the bone–implant interface, which can be exposed upon ultrasound examination [ 48 ]. Finally, artificial intelligence has been used in studies to measure the peri-implant bone loss [ 49 ].…”

“…The accuracy for the neural network was 85%. Thus, neural networks may be useful tools to assess radiographic bone loss and to obtain image-based periodontal diagnosis [ 49 ]. Other authors have also used neural networks to evaluate radiographic bone loss, and in this way, developed an automatic method for staging periodontitis according to the new criteria proposed at the 2017 World Workshop on the Classification of Periodontal and Peri-implant Diseases and Conditions.…”

Evaluation of the Progression of Periodontitis with the Use of Neural Networks

“…Thanathornwong et al used a convolutional neural network (CNN) in their work to identify periodontally compromised teeth on digital panoramic radiographs. The average precision of these networks was 81% [ 46 ]. Lee et al measured the radiographic alveolar bone level and assessed alveolar bone loss by using CNN.…”

Performance comparison of three deep learning models for impacted mesiodens detection on periapical radiographs

“…In dentistry, artificial intelligence (AI) models have been introduced that use panoramic radiographs 8 , 9 or cone-beam computed tomography (CBCT) 10 , 11 to make automatic diagnoses. A few studies have investigated tooth identification 12 , 13 and dental caries 14 using periapical radiographs, but no studies have focused on detecting mesiodens.…”