Dental panoramic radiographs (DPRs) provide information required to potentially evaluate bone density changes through a textural and morphological feature analysis on a mandible. This study aims to evaluate the discriminating performance of deep convolutional neural networks (CNNs), employed with various transfer learning strategies, on the classification of specific features of osteoporosis in DPRs. For objective labeling, we collected a dataset containing 680 images from different patients who underwent both skeletal bone mineral density and digital panoramic radiographic examinations at the Korea University Ansan Hospital between 2009 and 2018. Four study groups were used to evaluate the impact of various transfer learning strategies on deep CNN models as follows: a basic CNN model with three convolutional layers (CNN3), visual geometry group deep CNN model (VGG-16), transfer learning model from VGG-16 (VGG-16_TF), and fine-tuning with the transfer learning model (VGG-16_TF_FT). The best performing model achieved an overall area under the receiver operating characteristic of 0.858. In this study, transfer learning and fine-tuning improved the performance of a deep CNN for screening osteoporosis in DPR images. In addition, using the gradient-weighted class activation mapping technique, a visual interpretation of the best performing deep CNN model indicated that the model relied on image features in the lower left and right border of the mandibular. This result suggests that deep learning-based assessment of DPR images could be useful and reliable in the automated screening of osteoporosis patients.

Diagnosis and treatment planning are the most important steps in the orthognathic surgery for the successful treatment. The purpose of this study was to develop a new artificial intelligent model for surgery/non-surgery decision and extraction determination, and to evaluate the performance of this model. The sample used in this study consisted of 316 patients in total. Of the total sample, 160 were planned with surgical treatment and 156 were planned with non-surgical treatment. The input values of artificial neural network were obtained from 12 measurement values of the lateral cephalogram and 6 additional indexes. The artificial intelligent model of machine learning consisted of 2-layer neural network with one hidden layer. The learning was carried out in 3 stages, and 4 best performing models were adopted. Using these models, decision-making success rates of surgery/non-surgery, surgery type, and extraction/non-extraction were calculated. The final diagnosis success rate was calculated by comparing the actual diagnosis with the diagnosis obtained by the artificial intelligent model. The success rate of the model showed 96% for the diagnosis of surgery/non-surgery decision, and showed 91% for the detailed diagnosis of surgery type and extraction decision. This study suggests the artificial intelligent model using neural network machine learning could be applied for the diagnosis of orthognathic surgery cases.

Determining the severity of dental crowding and the necessity of tooth extraction for orthodontic treatment planning are time-consuming processes and there are no firm criteria. Thus, automated assistance would be useful to clinicians. This study aimed to construct and evaluate artificial intelligence (AI) systems to assist with such treatment planning. A total of 3,136 orthodontic occlusal photographs with annotations by two orthodontists were obtained. Four convolutional neural network (CNN) models, namely ResNet50, ResNet101, VGG16, and VGG19, were adopted for the AI process. Using the intraoral photographs as input, the crowding group and the necessity of tooth extraction were obtained. Arch length discrepancy analysis with AI-detected landmarks was used for crowding categorization. Various statistical and visual analyses were conducted to evaluate the performance. The maxillary and mandibular VGG19 models showed minimum mean errors of 0.84 mm and 1.06 mm for teeth landmark detection, respectively. Analysis of Cohen’s weighted kappa coefficient indicated that crowding categorization performance was best in VGG19 (0.73), decreasing in the order of VGG16, ResNet101, and ResNet50. For tooth extraction, the maxillary VGG19 model showed the highest accuracy (0.922) and AUC (0.961). By utilizing deep learning with orthodontic photographs, dental crowding categorization and diagnosis of orthodontic extraction were successfully determined. This suggests that AI can assist clinicians in the diagnosis and decision making of treatment plans.

The aim of this study was to investigate the relationship between image patterns in cephalometric radiographs and the diagnosis of orthognathic surgery and propose a method to improve the accuracy of predictive models according to the depth of the neural networks. The study included 640 and 320 patients requiring non-surgical and surgical orthodontic treatments, respectively. The data of 150 patients were exclusively classified as a test set. The data of the remaining 810 patients were split into five groups and a five-fold cross-validation was performed. The convolutional neural network models used were ResNet-18, 34, 50, and 101. The number in the model name represents the difference in the depth of the blocks that constitute the model. The accuracy, sensitivity, and specificity of each model were estimated and compared. The average success rate in the test set for the ResNet-18, 34, 50, and 101 was 93.80%, 93.60%, 91.13%, and 91.33%, respectively. In screening, ResNet-18 had the best performance with an area under the curve of 0.979, followed by ResNets-34, 50, and 101 at 0.974, 0.945, and 0.944, respectively. This study suggests the required characteristics of the structure of an artificial intelligence model for decision-making based on medical images.

The aim of this study was to evaluate the deep convolutional neural networks (DCNNs) based on analysis of cephalometric radiographs for the differential diagnosis of the indications of orthognathic surgery. Among the DCNNs, Modified-Alexnet, MobileNet, and Resnet50 were used, and the accuracy of the models was evaluated by performing 4-fold cross validation. Additionally, gradient-weighted class activation mapping (Grad-CAM) was used to perform visualized interpretation to determine which region affected the DCNNs’ class classification. The prediction accuracy of the models was 96.4% for Modified-Alexnet, 95.4% for MobileNet, and 95.6% for Resnet50. According to the Grad-CAM analysis, the most influential regions for the DCNNs’ class classification were the maxillary and mandibular teeth, mandible, and mandibular symphysis. This study suggests that DCNNs-based analysis of cephalometric radiograph images can be successfully applied for differential diagnosis of the indications of orthognathic surgery.