In this study, a deep learning-based method for developing an automated diagnostic support system that detects periodontal bone loss in the panoramic dental radiographs is proposed. The presented method called DeNTNet not only detects lesions but also provides the corresponding teeth numbers of the lesion according to dental federation notation. DeNTNet applies deep convolutional neural networks(CNNs) using transfer learning and clinical prior knowledge to overcome the morphological variation of the lesions and imbalanced training dataset. With 12,179 panoramic dental radiographs annotated by experienced dental clinicians, DeNTNet was trained, validated, and tested using 11,189, 190, and 800 panoramic dental radiographs, respectively. Each experimental model was subjected to comparative study to demonstrate the validity of each phase of the proposed method. When compared to the dental clinicians, DeNTNet achieved the F1 score of 0.75 on the test set, whereas the average performance of dental clinicians was 0.69.

The objective of this retrospective study is to verify the effectiveness and safety of patient-specific titanium implants on maxillofacial bones, with a long-term follow-up. Total 16 patients with various maxillofacial defects underwent reconstruction using patient-specific titanium implants. Titanium implants, manufactured by electron beam melting, selective laser sintering, or milling, were inserted into the maxilla, mandible, or zygoma. Long-term follow‐up (36.7 ± 20.1 months) was conducted after the surgery. Bone fusion of the titanium implant body, postoperative infection, implant malunion, functional results, patient satisfaction, subsidence, osteolysis around the implants, and complications were recorded and analyzed at the last follow-up. Of the 28 implants, only one failed to unite with the bone; therefore, revision surgery was performed. No osteolysis or subsidence around the titanium implants nor adverse events were observed; the mean VAS score for satisfaction was 9. All patients enrolled in this trial were esthetically and functionally satisfied with their surgical results, and fixation failure and esthetic dissatisfaction complications were well resolved. Patient-specific titanium showed satisfactory outcomes when used to treat various oral and maxillofacial defects. A 3D printed titanium implant can be effectively used in the reconstruction of the zygoma and mandible instead of autogenous bone without donor site morbidity.

The aim of this study was to segment the maxillary sinus into the maxillary bone, air, and lesion, and to evaluate its accuracy by comparing and analyzing the results performed by the experts. We randomly selected 83 cases of deep active learning. Our active learning framework consists of three steps. This framework adds new volumes per step to improve the performance of the model with limited training datasets, while inferring automatically using the model trained in the previous step. We determined the effect of active learning on cone-beam computed tomography (CBCT) volumes of dental with our customized 3D nnU-Net in all three steps. The dice similarity coefficients (DSCs) at each stage of air were 0.920 ± 0.17, 0.925 ± 0.16, and 0.930 ± 0.16, respectively. The DSCs at each stage of the lesion were 0.770 ± 0.18, 0.750 ± 0.19, and 0.760 ± 0.18, respectively. The time consumed by the convolutional neural network (CNN) assisted and manually modified segmentation decreased by approximately 493.2 s for 30 scans in the second step, and by approximately 362.7 s for 76 scans in the last step. In conclusion, this study demonstrates that a deep active learning framework can alleviate annotation efforts and costs by efficiently training on limited CBCT datasets.

Article published by the Endocrine Society Article published by the Endocrine Society