Oral Radiology最新文献

Objective: The purpose of this study is the development of a deep learning model based on nnU-Net v2 for the automated segmentation of sphenoid sinus and middle skull base anatomic structures in cone-beam computed tomography (CBCT) volumes, followed by an evaluation of the model's performance.

Material and methods: In this retrospective study, the sphenoid sinus and surrounding anatomical structures in 99 CBCT scans were annotated using web-based labeling software. Model training was conducted using the nnU-Net v2 deep learning model with a learning rate of 0.01 for 1000 epochs. The performance of the model in automatically segmenting these anatomical structures in CBCT scans was evaluated using a series of metrics, including accuracy, precision, recall, dice coefficient (DC), 95% Hausdorff distance (95% HD), intersection on union (IoU), and AUC.

Results: The developed deep learning model demonstrated a high level of success in segmenting sphenoid sinus, foramen rotundum, and Vidian canal. Upon evaluation of the DC values, it was observed that the model demonstrated the highest degree of ability to segment the sphenoid sinus, with a DC value of 0.96.

Conclusion: The nnU-Net v2-based deep learning model achieved high segmentation performance for the sphenoid sinus, foramen rotundum, and Vidian canal within the middle skull base, with the highest DC observed for the sphenoid sinus (DC: 0.96). However, the model demonstrated limited performance in segmenting other foramina of the middle skull base, indicating the need for further optimization for these structures.

Objectives: To validate the accuracy of the Cameriere's method and a previous Egyptian formula for determining the dental age (DA) of Egyptian children living along the Mediterranean region, while also assessing the accuracy in applying the new regression formula that is tailored to the Mediterranean region.

Methods: This study analyzed a training sample of 440 orthopantomograms (OPGs) and a testing sample of 212 OPGs of children and adolescents aged 515 years. Dental age (DA) was calculated and compared to chronological age (CA) using various formulas. Cameriere's formula was used to create a new Egyptian formula specific to the Mediterranean region, and a linear regression model was used to evaluate all morphological variables as well as gender. The new Egyptian formula specific to the Mediterranean region was later compared to Cameriere's formula and a previous Egyptian formula.

Results: The new Egyptian formula (Mediterranean region) for DA estimation (adjusted mean = 10.329, 95% CI 10.046, 10.611), highly matched the CA (adjusted mean = 10.331, 95%CI 10.033, 10.630, p = 1.00) showing no statistically significant difference. However, DA calculated by Cameriere's formula (adjusted mean = 9.469, 95% CI 9.236, 9.703, p < 0.001) as well as the previous Egyptian formula (adjusted mean = 10.000, 95%CI 9.729, 10.272, p < 0.001) significantly underestimated the age.

Conclusion: The new Egyptian formula is the most accurate for estimating dental age in Egyptian children living along the Mediterranean region and recommended as the first choice in forensic and clinical application for DA estimation in Egyptian children living along the Mediterranean.

Objectives: This study aimed to evaluate the ultrasonographic features of the parotid and submandibular salivary glands in individuals with long-term cigarette use and to investigate the effects of smoking on these glands.

Methods: A total of 120 participants were included in the study: 55 individuals who had been actively smoking for at least 5 years and 65 healthy controls who had never smoked. The echogenicity, parenchymal architecture, glandular margins, and vascular patterns of the parotid and submandibular glands were assessed using ultrasonography. In addition, the volume of each gland was measured, and tissue stiffness (elasticity) was quantitatively evaluated using shear wave elastography. For statistical analysis, the independent samples t-test was used for continuous variables, and the chi-square test was applied for categorical variables.

Results: There was no statistically significant difference between groups in terms of echogenicity. However, parenchymal heterogeneity and margin irregularity were more frequently observed in smokers. Vascular assessments revealed a significant increase in peripheral vascular patterns and poor vascularity, along with a decrease in central-peripheral distribution in the smoking group. Moreover, both parotid and submandibular gland volumes were significantly higher in smokers compared to controls. Elastography analysis showed a notable increase in tissue stiffness, particularly in the left-sided glands.

Conclusions: The findings suggest that long-term cigarette use may lead to structural alterations, increased tissue stiffness, and vascular changes in the parotid and submandibular salivary glands. These changes may reflect inflammatory, fibrotic, and microcirculatory disturbances induced by smoking in the glandular parenchyma. Non-invasive imaging modalities such as ultrasonography and shear wave elastography appear to be effective tools for detecting these alterations in a reliable and early manner.

Objectives: Temporomandibular joint disorders (TMDs) require precise imaging of both osseous and soft-tissue components for accurate diagnosis. Current diagnostic protocols often necessitate separate cone-beam computed tomography (CBCT) and magnetic resonance imaging (MRI) scans, increasing cost, complexity, and patient burden. This study aimed to develop and validate the Sivan 3D CBCT Volumetric Imaging Protocol, a novel post-processing technique designed to enable simultaneous high-resolution visualization of the temporomandibular joint (TMJ) disc, retrodiscal tissue, and adjacent osseous structures using a single CBCT data set.

Methods: This retrospective study analyzed anonymized CBCT volumes from 42 patients diagnosed with TMD. The data sets were processed using NNT software, incorporating dual-mode volumetric rendering, voxel intensity modulation, and anisotropic filtering. Fifteen expert raters assessed conventional CBCT, enhanced CBCT, and MR images across eight diagnostic parameters using a 5-point Likert scale. Inter- and intra-rater reliability was assessed using intraclass correlation coefficients (ICC), and statistical comparisons were performed using one-way ANOVA, Tukey's post-hoc test, and Cohen's d effect size analysis.

Results: Enhanced CBCT images showed excellent inter- and intra-rater reliability (ICC > 0.95) and significantly outperformed conventional CBCT and MRI in visualizing both soft and hard TMJ structures (p < 0.001), with large effect sizes (Cohen's d > 2.3), especially for disc and retrodiscal tissue delineation.

Conclusions: The Sivan 3D CBCT Volumetric Imaging Protocol provides a robust, single-modality alternative to MRI for comprehensive TMJ evaluation, improving diagnostic accuracy, efficiency, and accessibility in clinical settings with limited MRI availability.

Objectives: The primary objective of this study is to evaluate the effectiveness of artificial intelligence-assisted segmentation methods in detecting carotid artery calcification (CAC) in panoramic radiographs and to compare the performance of different YOLO models: YOLOv5x-seg, YOLOv8x-seg, and YOLOv11x-seg. Additionally, the study aims to investigate the association between patient gender and the presence of CAC, as part of a broader epidemiological analysis.

Methods: In this study, 30,883 panoramic radiographs were scanned. Annotations were made on 652 radiographs exhibiting features consistent with CAC, totaling 1,086 annotations. Deep learning-based analysis was conducted using three distinct YOLO segmentation models. The performance of these models was assessed using metrics such as precision, accuracy, and F1 score.

Results: The YOLOv5x-seg model exhibited a balanced performance with a precision, sensitivity, and F1 score of 84.62% each. The YOLOv8x-seg model demonstrated higher sensitivity at 88.46%, albeit with a slightly higher false positive rate, evidenced by a precision of 78.63%. The YOLOv11x-seg model achieved the highest precision at 93.41%, an F1 score of 87.18%, and a sensitivity of 81.73%.

Conclusions: AI-based segmentation models utilizing YOLO algorithms can be considered reliable tools for detecting CAC in panoramic radiographs. These models display promising performance appropriate for clinical applications; however, further research with larger and more diverse datasets is required to verify their generalizability and efficacy.

Objective: The present study examined the efficacy of the ¹⁸F-FDG PET/CT (18F-FDG PET)-based Lugano Classification for Oral Cancer (Lugano-OC) to stage treatment-resistant oral cancer and its utility for predicting treatment responses.

Patients and methods: We included 29 oral squamous cell carcinoma (OSCC) patients with disease progression post the primary treatment, who had ¹⁸F-FDG PET scans prior to secondary treatment initiation, and were evaluable via Lugano-OC between April 2013 and September 2022. To adapt Lugano-OC to the head and neck region, we designated the clavicle level, a regional lymph node in oral cancer, as the reference standard and visually evaluated 18F-FDG PET accumulation patterns. The primary endpoint was the best overall response, which was assessed according to the Response Evaluation Criteria in Solid Tumors.

Results: Patients were categorized into a responder group (n = 7) and non-responder group (n = 22). A univariate analysis revealed significant differences in the CT maximum diameter and metabolic tumor volume (p = 0.050, and 0.037, respectively). A multivariate Cox regression analysis revealed a correlation (hazard ratio 3.59, p = 0.046, 95% confidence interval 1.02-12.6) between Lugano-OC and the duration of treatment responses.

Conclusion: Among OSCC patients with disease progression post the primary treatment, those with Lugano-OC-localized disease who received radiation chemotherapy had a significantly longer treatment response than patients with advanced-stage disease. These results suggest the predictive value of Lugano-OC for assessing treatment responses in this patient population.

Objectives: This study aimed to evaluate the applicability of a paralleling technique in dental patients in Hong Kong via cone-beam computed tomography (CBCT) images.

Methods: We analyzed CBCT images of 28 patients who underwent CBCT examination at Prince Philip Dental Hospital between March 2016 and July 2021. Overall, 136 teeth were studied. We created an 8 mm-thick reconstructed maximum intensity projection image to show each subject tooth, traced them with the surrounding structures, such as the hard palate and the buccal skin surface of the maxilla, and measured various parameters for determining the tooth length and palatal height. We also considered the applicability of the parallel technique.

Results: We successfully measured the parameters with excellent intra- and inter-observer reliability indices. The results revealed that the applicability of this technique was not feasible. These results were similar to those of a previous study. Only two periapical radiographs showed distorted shapes and inconsistent root lengths.

Conclusions: Using CBCT image analysis, this study clarified that the paralleling technique is not applicable even for patients in Hong Kong. Clinicians should note that the paralleling technique is not strictly a paralleling method and may lead to image distortion.

Objectives: The aim of this study is to comparatively evaluate morphologic changes in the mandibular angle and condylar region, assess the mandibular cortical index (MCI), and analyze the relationships between these parameters in bruxers and non-bruxers using panoramic radiographs.

Methods: A total of 182 patients (364 mandibular condyle and angle), 91 bruxers and 91 non-bruxers (controls), aged between 18 and 35, were participated in this study. Three hundred sixty four mandibular angles were classified as G0, G1, G2, G3 in terms of bone apposition and direction change. In addition, osseous changes in the mandibular condyle were determined and the endoestal edge of the basal cortex was evaluated according to the MCI classification. Parameters were analyzed using the Pearson's Chi-Square test and Fisher's Exact test. A p value < 0.05 was considered significant.

Result: MCI class C1, G0 class, and normal condyle type were found to be significantly more common in the non-bruxist group than in the bruxist group (p < 0.001, p = 0.025, p = 0.006, respectively). It was determined that deformity and MCI-C2 class were more common in the bruxist group (p = 0.006, p < 0.001). Morphologic changes in the endosteal margin of the cortex and in the condylar region were observed more frequently in bruxist individuals.

Conclusion: For a general overview of the probable presence of bruxism, osseous changes in the mandibular condyle, and MCI can be used as auxiliary diagnostic markers.

Introduction: This study aims to assess the structural features of adenoids and nasopharyngeal airways in boys with nonsyndromic unilateral complete cleft of lip and palate (UCLP group) compared to non-cleft lip and palate patients of the same age (control group), and to identify differences in the growth and development of adenoids and nasopharyngeal airways between two groups.

Methods: Adenoid and nasopharyngeal airway length, area, and volume were retrospectively analyzed using CBCT images (n = 120). The UCLP group comprised boys aged 6-11 with UCLP (n = 60), while the control group consisted of boys without cleft lip and palate (n = 60).

Results: Statistically significant differences (p < 0.05) were observed between the UCLP group and the control group for adenoids, including the upper and lower parts of the adenoids, adenoid thickness, and adenoidal-nasopharyngeal ratio (ANR) values. In addition, significant differences (p < 0.05) were found in the lower part of the nasopharyngeal airway, area of palatal plane, area of lower nasopharyngeal airway, and volume of lower nasopharyngeal airway.

Conclusions: UCLP boys aged 6-11 exhibit a broader nasopharyngeal airway, particularly in the palate and lower nasopharyngeal airway region, adenoid degeneration becomes more pronounced with age. When UCLP boys aged 6-11 experience adenoid hypertrophy, special attention should be given to the upper palatal plane and nasal ventilation. These factors should be the primary considerations when deciding whether to perform adenoidectomy. Similar to cephalometric X-ray films, utilizing CBCT image measurement to calculate ANR values can provide a more accurate method for evaluating adenoid size and nasopharyngeal airway obstruction in UCLP children.

The study aimed to review the applicability and performance of various Convolutional Neural Network (CNN) models for the identification of periodontal bone loss (PBL) in digital periapical radiographs achieved through classification, detection, and segmentation approaches. We searched the PubMed, IEEE Xplore, and SCOPUS databases for articles published up to June 2024. After the selection process, a total of 11 studies were included in this review. The reviewed studies demonstrated that CNNs have a significant potential application for automatic identification of PBL on periapical radiographs through classification and segmentation approaches. CNN architectures can be utilized to classify the presence or absence of PBL, the severity or degree of PBL, and PBL area segmentation. CNN showed a promising performance for PBL identification on periapical radiographs. Future research should focus on dataset preparation, proper selection of CNN architecture, and robust performance evaluation to improve the model. Utilizing an optimized CNN architecture is expected to assist dentists by providing accurate and efficient identification of PBL.