Sophisticated segmentation of the craniomaxillofacial bones (the mandible and maxilla) in computed tomography (CT) is essential for diagnosis and treatment planning for craniomaxillofacial surgeries. Conventional manual segmentation is time-consuming and challenging due to intrinsic properties of craniomaxillofacial bones and head CT such as the variance in the anatomical structures, low contrast of soft tissue, and artifacts caused by metal implants. However, data-driven segmentation methods, including deep learning, require a large consistent dataset, which creates a bottleneck in their clinical applications due to limited datasets. In this study, we propose a deep learning approach for the automatic segmentation of the mandible and maxilla in CT images and enhanced the compatibility for multi-center datasets. Four multi-center datasets acquired by various conditions were applied to create a scenario where the model was trained with one dataset and evaluated with the other datasets. For the neural network, we designed a hierarchical, parallel and multi-scale residual block to the U-Net (HPMR-U-Net). To evaluate the performance, segmentation with in-house dataset and with external datasets from multi-center were conducted in comparison to three other neural networks: U-Net, Res-U-Net and mU-Net. The results suggest that the segmentation performance of HPMR-U-Net is comparable to that of other models, with superior data compatibility.

Patients who received orthodontic treatment are likely to have apical root shortening. It appears that external apical root resorption results from a combination of patient-related risk factors such as genetic influences, systemic factors, and orthodontic treatment-related factors. Regarding the fact that the anterior segmental osteotomy (ASO) has been known for its possibility of complementing external apical root resorption and of buffering periodontal problems, it has been the preferred treatment. However, the studies on the efficacy of ASO in preserving the root are not sufficient. In this study, we compared the amount of root resorption between the patients who only received orthodontic treatment and the patients who received orthodontic treatment with ASO. This study included 28 patients (the number of incisor = 198) who received orthodontic treatment with or without ASO. We categorize them into groups A and B by the type of orthodontic treatment (group A: conventional orthodontic treatment; group B: orthodontic treatment with ASO). Cone-beam computed tomographic and cephalometric evaluations were retrospectively performed on the radiographs taken for the diagnosis of the treatment before treatment and at the end of active treatment. In group B, root resorption itself and its rate both turned out to have significantly lower than those in group A. Also, the change of incisal angle is significantly smaller in group B than in group A. On the other hand, in group A, the change of incisal angle was positively correlated with the change of AP (anteroposterior) position. In group B, the change of incisal angle was negatively correlated with the duration of the orthodontic treatment. In group B, amount of root resorption (mm) was positively correlated with the duration of the orthodontic treatment. The results show lesser root resorption and shorter treatment duration with ASO than with conventional orthodontic treatment. Therefore, if the indications are accurately determined, ASO can be an effective treatment option when the amount of root resorption is expected to be high, especially in late adults.

This study aimed to present and evaluate alternative lesion-specific mandibulectomy methods for preserving the mandibular anatomical structures as compared with the conventional virtual surgical plan. Fifteen patients who received segmental mandibulectomy were included in this study, and the following parameters were evaluated: (1) the disease-free bone margin, (2) the volume and surface between the tailor-made resection simulation and conventional resection simulation, and (3) the preserved mandibular anatomical structures. In all 15 patients, disease-free bone margins were confirmed by histopathology. Volumes of conventional resection simulation and tail-made resection simulation were 49,468.66 ± 14,007.96 mm³ and 52,610.01 ± 13,755.33 mm³ and the surfaces were 20,927.38 ± 4471.70 mm² and 22,356.49 ± 4185.73 mm², respectively; these were statistically significant (both, p < 0.001). Mandibular dentition was partially preserved in six patients. Twelve of the 15 patients had changes in defect classification with preservation of the mandibular inferior border. In conclusion, alternative lesion-specific mandibulectomy was a less invasive method for effectively removing mandibular lesions while preserving the important anatomical structures of the mandible.

Hyun-Soo Jung, In-Seo Chang, Jae-yeon Kim, In-Seok Song, Bo-Yeon Hwang, Sang-Ho Jun. JKAOMS 2025;51:117-22. https://doi.org/10.5125/jkaoms.2025.51.2.117

Bo-Yeon Hwang, Chandong Jeen, Myoung-Kyun Oh, Jaeyeon Kim, and Jung-Woo Lee. Oral Biol Res 2025;49:. https://doi.org/10.21851/obr.25.031