RESUMEN
Automatic mandible segmentation of CT images is an essential step to achieve an accurate preoperative prediction of an intended target in three-dimensional (3D) virtual surgical planning. Segmentation of the mandible is a challenging task due to the complexity of the mandible structure, imaging artifacts, and metal implants or dental filling materials. In recent years, utilizing convolutional neural networks (CNNs) have made significant improvements in mandible segmentation. However, aggregating data at pooling layers in addition to collecting and labeling a large volume of data for training CNNs are significant issues in medical practice. We have optimized data-efficient 3D-UCaps to achieve the advantages of both the capsule network and the CNN, for accurate mandible segmentation on volumetric CT images. A novel hybrid loss function based on a weighted combination of the focal and margin loss functions is also proposed to handle the problem of voxel class imbalance. To evaluate the performance of our proposed method, a similar experiment was conducted with the 3D-UNet. All experiments are performed on the public domain database for computational anatomy (PDDCA). The proposed method and 3D-UNet achieved an average dice coefficient of 90% and 88% on the PDDCA, respectively. The results indicate that the proposed method leads to accurate mandible segmentation and outperforms the popular 3D-UNet model. It is concluded that the proposed approach is very effective as it requires more than 50% fewer parameters than the 3D-UNet.
Asunto(s)
Tomografía Computarizada de Haz Cónico , Mandíbula , Artefactos , Bases de Datos Factuales , Humanos , Mandíbula/diagnóstico por imagen , Mandíbula/cirugía , Márgenes de EscisiónRESUMEN
The aim of the present study was to compare the application of alloderm and mucosal graft for vestibuloplasty. This randomized controlled trial with split mouth design was carried out on 20 edentulous patients. Patients underwent vestibuloplasty surgery with the Clark technique. Half of the prepared bed in each patient was covered with alloderm and the other half with mucosal graft. Vestibule depth (width of fixed tissue) and relapse in the two sides immediately after surgery, and 1, 3 and 6 months after surgery were measured and compared. Statistical analysis was carried out using the Kolmogorov-Smirnov, Student's paired t and Friedman tests. The width of the fixed tissue in the alloderm graft at 1, 3 and 6 month intervals was significantly lower than that in the autograft (P<0.05). The difference in relapse between the two grafts was not statistically significant at any time. The results of the study suggest that alloderm is as effective as mucosal grafts in vestibuloplasty.
Asunto(s)
Colágeno , Encía/trasplante , Mucosa Bucal/trasplante , Piel Artificial , Vestibuloplastia/métodos , Anciano , Anciano de 80 o más Años , Femenino , Humanos , Masculino , Persona de Mediana Edad , Boca Edéntula/rehabilitación , Recurrencia , Método Simple Ciego , Estadísticas no ParamétricasRESUMEN
Surgically assisted rapid palatal expansion (SARPE) is associated with postoperative cephalometric changes. In this study we analyse these changes in the sagittal plane in orthognathic patients undergoing SARPE followed by orthodontic treatment and Le Fort I, bilateral sagittal split osteotomy (BSSO), or bimaxillary surgery. This is a retrospective review of 50 patients (20 males, 30 females) undergoing orthognathic treatment with SARPE to correct transversal deficiency of the maxilla as part of a comprehensive treatment plan. PP-SN, SNA, and ANB angles were increased and U1-SN and U1-PP angles were decreased. All changes were statistically significant. Changes of SNB, PP-Mand plane angle, and SN-Mand. plane angle were not statistically significant. Surgically assisted rapid palatal expansion using a bone-borne appliance as a preparative step for later orthognathic surgery results in clockwise rotation of the maxilla.