RESUMO
Transfer learning (TL) is an alternative approach to the full training of deep learning (DL) models from scratch and can transfer knowledge gained from large-scale data to solve different problems. ImageNet, which is a publicly available large-scale dataset, is a commonly used dataset for TL-based image analysis; many studies have applied pre-trained models from ImageNet to clinical prediction tasks and have reported promising results. However, some have questioned the effectiveness of using ImageNet, which consists solely of natural images, for medical image analysis. The aim of this study was to evaluate whether pre-trained models using RadImageNet, which is a large-scale medical image dataset, could achieve superior performance in classification tasks in dental imaging modalities compared with ImageNet pre-trained models. To evaluate the classification performance of RadImageNet and ImageNet pre-trained models for TL, two dental imaging datasets were used. The tasks were (1) classifying the presence or absence of supernumerary teeth from a dataset of panoramic radiographs and (2) classifying sex from a dataset of lateral cephalometric radiographs. Performance was evaluated by comparing the area under the curve (AUC). On the panoramic radiograph dataset, the RadImageNet models gave average AUCs of 0.68 ± 0.15 (p < 0.01), and the ImageNet models had values of 0.74 ± 0.19. In contrast, on the lateral cephalometric dataset, the RadImageNet models demonstrated average AUCs of 0.76 ± 0.09, and the ImageNet models achieved values of 0.75 ± 0.17. The difference in performance between RadImageNet and ImageNet models in TL depends on the dental image dataset used.
RESUMO
Sturge-Weber syndrome (SWS) is characterized by hemangiomas, glaucoma, and central nervous system disorders. Here, we report the case of a 15-year-old boy with SWS and upper-lip hypertrophy who underwent surgical orthodontic treatment for correction of a large overjet and deep overbite. In addition to the a large overjet and deep overbite, interdental spacing was observed in both the arches. The mandible was retrognathic and deviated to the right side. No maxillary occlusal canting or temporomandibular joint symptoms were observed. The patient was diagnosed with skeletal maxillary protrusion with spaced dentition and mandibular deviation to the right due to SWS. After presurgical orthodontic treatment using a multibracket appliance, we performed a sagittal split ramus osteotomy (SSRO) alone due to the presence of a hemangioma around the maxilla. No abnormal bleeding or cerebral hemorrhage due to increased blood pressure was observed during the SSRO. Postoperatively, the maxillary and mandibular arches were well-aligned, the deep overbite and excessive overjet improved, and bilateral angle class I molar and canine relationships were established. Furthermore, mandibular deviation improved, and the midlines of both arches approximately coincided with the facial midline. In conclusion, orthognathic surgery is feasible in patients with SWS after carefully evaluating the sites and sizes of the hemangiomas.
RESUMO
The purpose of this study was to construct deep learning models for more efficient and reliable sex estimation. Two deep learning models, VGG16 and DenseNet-121, were used in this retrospective study. In total, 600 lateral cephalograms were analyzed. A saliency map was generated by gradient-weighted class activation mapping for each output. The two deep learning models achieved high values in each performance metric according to accuracy, sensitivity (recall), precision, F1 score, and areas under the receiver operating characteristic curve. Both models showed substantial differences in the positions indicated in saliency maps for male and female images. The positions in saliency maps also differed between VGG16 and DenseNet-121, regardless of sex. This analysis of our proposed system suggested that sex estimation from lateral cephalograms can be achieved with high accuracy using deep learning.
Assuntos
Aprendizado Profundo , Humanos , Feminino , Masculino , Estudos Retrospectivos , Cefalometria/métodos , Adulto , Determinação do Sexo pelo Esqueleto/métodos , Curva ROCRESUMO
This cross-sectional study aimed to explore the correlation between maxillofacial morphology and caries risk, assessed using salivary tests, in orthodontic patients. Despite enhancing the oral health-related quality of life, orthodontic treatment may adversely affect oral hygiene and increase caries risk. This study included 1071 patients all of whom underwent orthodontic examinations and salivary tests before starting orthodontic treatment at a hospital. Salivary tests were performed to assess the secretion rate, pH, buffering capacity, and counts of cariogenic bacteria. The maxillofacial morphology was evaluated using cephalometric X-rays and dental models. Statistical analyses revealed significant correlations among salivary characteristics, bacterial scores, and maxillofacial morphology. Notably, the facial angle and Y-axis values were associated with salivary secretion (p < 0.001), pH (p < 0.001), buffering capacity (p < 0.05), and cariogenic bacterial scores (p < 0.01), respectably. In conclusion, assessing the maxillofacial morphology before orthodontic treatment may aid in predicting the risk of bacterial oral diseases, offering valuable insights into personalized preventive measures. These findings underscore the potential for comprehensive evaluations to enhance caries risk assessment in orthodontic patients.
RESUMO
OBJECTIVE: Root resorption may occur during orthodontic treatment. Herein, we investigated the effect of a culture supernatant of stem cells derived from human exfoliated deciduous teeth on root resorption. DESIGN: Twelve 8-week-old male Sprague-Dawley rats were used, and their maxillary first molars were pulled with excessive orthodontic force to induce root resorption. On days 1 and 7 after traction initiation, stem cells derived from human exfoliated deciduous teeth and alpha minimum essential medium (control group) were administered. After 14 days, the maxillary bone was evaluated for tooth movement. The expression of osteoprotegerin, receptor activator of nuclear factor κB ligand, tumor necrosis factor α, interleukin 1ß, interleukin 6, and interleukin 17 was evaluated on the compression side and tension side. RESULTS: No significant difference in tooth movement was observed between the two groups. Root resorption decreased in the group administered the culture supernatant compared with in the control. Immunohistochemical staining revealed increased osteoprotegerin expression and decreased receptor activators for nuclear factor κB ligand, tumor necrosis factor α, interleukin 1ß, interleukin 6, and interleukin 17 on the compression side and tension side. CONCLUSIONS: Administration of stem cells derived from human exfoliated deciduous teeth affected the expression of osteoprotegerin, receptor activator of nuclear factor κB ligand, tumor necrosis factor α, interleukin 1ß, interleukin 6 and interleukin 17; hence, these stem cells may inhibit root resorption by regulating their expression.
Assuntos
Reabsorção da Raiz , Ratos , Humanos , Masculino , Animais , Reabsorção da Raiz/metabolismo , Osteoprotegerina/metabolismo , Interleucina-17/metabolismo , Fator de Necrose Tumoral alfa/metabolismo , Osteoclastos , Interleucina-6/metabolismo , Ligante RANK/metabolismo , Interleucina-1beta/metabolismo , Ratos Sprague-Dawley , Células-Tronco/metabolismo , Dente Decíduo , Técnicas de Movimentação DentáriaRESUMO
Bacterial hyaluronate lyases (Hys) are enzymes that degrade hyaluronic acid in their host and are known to contribute to the pathogenesis of several illnesses. The first two identified Hys genes in Staphylococcus aureus were registered as hysA1 and hysA2. However, their annotations have been mistakenly reversed in some registered assembly data, and different abbreviations (hysA and hysB) in some reports complicates comparative analysis of Hys proteins. We investigated the hys loci of S. aureus genome sequences registered in public databases, analyzed the homology, and defined hysA as hys located in the core genome surrounded by a lactose metabolic operon and a ribosomal protein cluster present in almost all strains and hysB as that located on the genomic island νSaß of the accessory genome. Homology analysis of the amino acid sequences of HysA and HysB revealed that they are conserved among clonal complex (CC) groups with a few exceptions. Thus, we propose a new nomenclature for S. aureus Hys subtypes: HysACC*** for HysA and HysBCC*** for HysB, with the asterisks representing the clonal complex number of the S. aureus strain producing the Hys subtype. The application of this proposed nomenclature will facilitate the intuitive, straightforward, and unambiguous designation of Hys subtypes and contribute to enhancing comparative studies in this regard. IMPORTANCE Numerous whole-genome sequence data for Staphylococcus aureus harboring two hyaluronate lyase (Hys) genes have been registered. However, the assigned gene names for hysA1 and hysA2 are incorrect in some assembled data, and in some cases, the genes are annotated differently as hysA and hysB. This creates confusion with respect to the nomenclature of Hys subtypes and complicates analysis involving Hys. In this study, we compared the homology of Hys subtypes and observed that to some extent, their amino acid sequences are conserved in each clonal complex group. Hys has been implicated as an important virulence factor, but relative sequence heterogeneity among S. aureus clones raises the question of whether Hys activities are different among these clones. Our proposed Hys nomenclature will facilitate comparison of the virulence of Hys, as well as discussions of the subject.
Assuntos
Infecções Estafilocócicas , Staphylococcus aureus , Humanos , Infecções Estafilocócicas/microbiologia , Fatores de Virulência/genética , Polissacarídeo-Liases/genéticaRESUMO
We report the complete genome sequence of Staphylococcus aureus strain JP025, which was isolated from a furunculosis sample from a Japanese patient. The strain carried two hyaluronate lyase genes, JP025hysA and JP025hysB, on the chromosome and was classified as sequence type 30.