Research status and outlook of deep learning in oral and maxillofacial medical imaging / 中华口腔医学杂志

Artificial intelligence, represented by deep learning, has received increasing attention in the field of oral and maxillofacial medical imaging, which has been widely studied in image analysis and image quality improvement. This narrative review provides an insight into the following applications of deep learning in oral and maxillofacial imaging: detection, recognition and segmentation of teeth and other anatomical structures, detection and diagnosis of oral and maxillofacial diseases, and forensic personal identification. In addition, the limitations of the studies and the directions for future development are summarized.

Research status and outlook of deep learning in oral and maxillofacial medical imaging / 中华口腔医学杂志

Artificial intelligence, represented by deep learning, has received increasing attention in the field of oral and maxillofacial medical imaging, which has been widely studied in image analysis and image quality improvement. This narrative review provides an insight into the following applications of deep learning in oral and maxillofacial imaging: detection, recognition and segmentation of teeth and other anatomical structures, detection and diagnosis of oral and maxillofacial diseases, and forensic personal identification. In addition, the limitations of the studies and the directions for future development are summarized.

Preliminary study on the accuracy of infrabony root surface area of single-root teeth by periapical films / 北京大学学报(医学版)

OBJECTIVE@#To explore the accuracy of periapical film in evaluating the infrabone root surface area of single-root teeth with different types of resorption of the alveolar bone.@*METHODS@#In this study, the medical records and imaging data of patients in Department of Periodontology of Peking University Hospital of Stomatology from January 2014 to December 2016 were retrospectively analysed. The involved teeth were divided into two groups: horizontal bone loss group and vertical bone loss group. Digital three-dimensional models of teeth were segmented from cone beam computed tomography (CBCT) data, and the percentage of the infrabone root surface area was measured. While the percentage of the residual alveolar bone was measured on the periapical films. The correlation of the two percentages were analyzed, and the differences between the two percentages of each group were compared.@*RESULTS@#Thirty-two patients and 79 single-root teeth were involved totally, in whom there were 25 men and 7 women, aged 26-60 years, and of which there were 41 teeth in the horizontal bone loss group and 38 teeth in the vertical bone loss group. The remaining alveolar bone height percentages were 45.13%-90.39% on the periapical films, and at the same time, the infrabone root surface area percentages were 36.27%-93.03% on CBCT. The residual alveolar bone height percentage of the horizontal bone loss group was not subject to normal distribution, so the residual alveolar bone height percentage and the infrabone root surface area percentage were analyzed by Wilcoxon matched pair rank test (P=0.382), and the result showed no statistically significant difference. The mean value of difference value of the two measurements was 0, and the consistency limit of 95 % was -9.7%-9.8%, between the clinical consistency limit (-15%-15%).The residual alveolar bone height percentage and the infrabone root surface area percentage of the vertical bone loss group were analyzed by paired T test (P< 0.001), and the result showed statistically significant difference. The mean value of difference value of the two measurements was 7.2 %, and the consistency limit of 95% was -13.1%-27.5%, beyond clinical consistency limit (-15%-15%).@*CONCLUSION@#For the teeth with horizontal bone loss, the proximal residual alveo-lar bone height reflected on the periapical films could show the infrabone root surface area relatively accurately. For the teeth with vertical bone loss, the periapical films could not show the degree of periodontal tissue loss accurately, and it would underestimate the extent of bone destruction usually, so CBCT might be taken to show the bone destruction condition of this type of teeth if necessary.

CT imaging features of diffuse tenosynovial giant cell tumor of temporomandibular joint / 华西口腔医学杂志

<p><b>OBJECTIVE</b>This study aimed to investigate the characteristic features of CT images of diffuse tenosynovial giant cell tumor (D-TSGCT) from the temporomandibular joint (TMJ) to understand CT imaging diagnosis of the disease.</p><p><b>METHODS</b>CT images and clinical information from 10 patients with D-TSGCT of TMJ were recruited for retrospective analysis from March 2013 to March 2017 in Peking University Hospital of Stomatology. All cases were treated by surgery and confirmed by pathology.</p><p><b>RESULTS</b>CT scan demonstrated hyperdense soft-tissue masses and various kinds of calcification in all of the 10 subjects. Contrast-enhanced scan exhibited obvious enhancement in six patients. Bone destruction of the mandibular condyles and skull base was found in seven and six subjects, respectively.</p><p><b>CONCLUSIONS</b>CT appearance of D-TSGCT of the TMJ is characterized by hyperdense soft tissue with calcification, further enhancement in contrast-enhanced scan, and bone destruction in the mandibular condyles and skull base in some cases.</p>