Yasin Yasa

Objectives: Panoramic radiography is a widely used imaging modality in dentistry. Areas of increased mineral density within anatomical structures appear as radiopaque regions on radiographs. Radiopacities detected on panoramic radiographs may be associated with both physiological and pathological processes. The aim of this study was to retrospectively evaluate common radiopacities observed in both hard and soft tissues on panoramic radiographs.Materials and Methods: This retrospective study analyzed 10000 panoramic radiographs to evaluate radiopacities in hard and soft tissues. Radiopacities detected on the radiographs were classified according to their radiographic characteristics. The distribution of radiopacities was analyzed according to age, gender, and anatomical localization. Statistical analysis was performed using the Chi-square test, with a significance level set at p < 0.05.Results: A total of 1060 radiopacities were identified in 963 patients, including 284 (3.25%) in soft tissues and 776 (8.89%) in hard tissues. Tonsilloliths were the most common soft tissue calcification, whereas idiopathic osteosclerosis was the most frequent hard tissue radiopacity. Conclusion: Although most hard and soft tissue radiopacities are asymptomatic and do not require treatment, accurate diagnosis of certain lesions is clinically important. Therefore, systematic evaluation of panoramic radiographs is essential for the detection and appropriate management of these findings.

Background: Artificial intelligence (AI) has made significant contributions to numerous technological advancements, particularly in radiological diagnostics, and continues to demonstrate progressive development within the domain of dentistry. This study aims to assess the efficacy of deep learning algorithms in the detection and numbering of primary dentition through panoramic radiographic analysis. Methods: This study incorporated a total of 701 panoramic radiographs obtained from pediatric individuals aged 5-9 years. The methodology employed a convolutional neural network (CNN)- based framework, specifically utilizing the mask region based (R)-CNN architecture, for automated dental detection and enumeration. The model’s performance metrics were systematically evaluated through confusion matrix analysis. Results: The model demonstrated superior performance metrics in the detection and numbering of deciduous dentition within panoramic radiographic analyses. The precision, recall, and F1-score values calculated using the confusion matrix were 0.973, 0.947, and 0.960 for the age group 5, 0.944, 0.904, and 0.923 for the age group 6, 0.959, 0.932, and 0.945 for the age group 7, 0.984, 0.913, and 0.947 for the age group 8, 0.920, 0.904, and 0.912 for the age group 9, respectively. Conclusion: Deep learning–based AI models are a promising approach for automatic detection and numbering of deciduous teeth in panoramic radiographs from children.

This study aimed to evaluate and compare the performance of state-of-the-art deep learning models for detecting and segmenting both radiolucent and radiopaque jaw lesions on panoramic radiographs. A total of 2371 anonymized panoramic radiographs containing jaw lesions were retrospectively collected and categorized into radiolucent and radiopaque datasets. Expert annotation was performed to delineate lesion boundaries and assign anatomical localization (anterior/posterior maxilla and mandible). Four deep learning architectures—YOLOv8, YOLOv11, Mask R-CNN, and RT-DETR—were trained and evaluated under three experimental scenarios: (I) training without spatial labels, (II) data augmentation with unlabeled background images, and (III) inclusion of spatial localization annotations. Performance metrics included precision, recall, F1-score, and mean average precision (mAP@0.5 and mAP@0.5–0.95), with paired t-tests used for statistical comparison. In Scenario I, YOLOv11x-seg and YOLOv8x-seg achieved the highest segmentation performance for radiolucent and radiopaque lesions, respectively. For detection, YOLOv8x performed best on radiolucent lesions, while RT-DETR-L outperformed others on radiopaque lesions. In Scenario II, while YOLOv8x-seg achieved the best segmentation results across both lesion types, RT-DETR-L demonstrated superior detection performance, particularly for radiolucent lesions. In Scenario III, RT-DETR-L consistently outperformed all models across both lesion types. This study demonstrates the potential of state-of-the-art deep learning models for effective detection of lesions in panoramic radiographs. The developed models may offer valuable support to clinicians in lesion evaluation; however, it is recommended that they be employed primarily as decision support tools within clinical workflows, rather than as standalone diagnostic systems.

Aim: Panoramic radiography is frequently used in dental practice as an imaging technique that provides an extensive view of facial anatomical structures. Achieving high-quality radiographic images requires precise positioning and technique. This study aims to evaluate the quality of panoramic radiographs, to identify common errors that compromise diagnostic adequacy, and to analyze interobserver agreement levels concerning these errors. Material and Method: A study analyzed 947 panoramic radiographs from the archive of Ordu University based on nine specific error criteria. Four research assistants evaluated the diagnostic quality of each image, classifying them as “excellent,” “diagnostically acceptable,” or “unacceptable.” Inter-observer agreement was measured with kappa statistics, and overall agreement was evaluated using the Fleiss κ test. Results: The rate of incorrect radiographs was 66.1% to 78.8%, with the most common error criterion differing for each observer. The least common error was the chin tipped too low for observers 1 and 2, and the patient positioned forward for observers 3 and 4. The highest inter-observer agreement was observed regarding the presence of foreign objects on the radiographs, while the lowest agreement occurred in cases where the patient was positioned too far back. Overall, the diagnostic quality of the panoramic radiographs was rated as “acceptable,” with scores ranging from 60.5% to 69.5%. The Fleiss Kappa analysis indicated fair agreement among the four observers in assessing radiographic quality (k=0.252). Conclusion: This study demonstrates that errors in panoramic radiography significantly impact image quality and diagnostic accuracy, highlighting the need for standardization, the use of various imaging models, and enhanced training in radiographic education.

Objective: Artificial intelligence’s (AI) potential to analyze medical data and improve patient outcomes is driving its rapid integration into healthcare systems. This study addresses the knowledge gap regarding Kazakhstani dental students’ attitudes toward AI in dentistry. Methods: A survey with 213 participants assessed dental students’ views on AI in dentistry at two Kazakhstani universities. A self-administered questionnaire (22 items) via Google Forms was used to gather data on demographics, baseline knowledge (information sources, understanding of AI principles, and familiarity with dental AI applications), and attitudes/perceptions. Frequency analyses and chi-square tests (p < 0.05) explored the responses. Results: The survey included 213 dental students (mean age: 21.79 ± 1.82 years). Two-thirds of participants reported basic knowledge of AI principles; however, the rate of dental AI application awareness was lower (40.9%). Friends and family (65.7%) were the primary sources of information, followed by lectures (14.1%) and social media (13.6%). The year of study and sex did not significantly impact AI knowledge or attitudes (p > 0.05). Conclusion: This study in Kazakhstan reveals dental students’ limited awareness of dental AI applications despite possessing basic AI knowledge. This study recommends taking AI-related dental courses and participating in AI workshops to equip students for the future of dentistry and empower them to contribute to and potentially lead Kazakh-specific dental AI advancements.

Objectives Introduction Fractal analysis (FA) is a computational method used to quantify the complex trabecular structure of bone. While FA has been widely applied in dentistry, there are challenges in standardizing the technique due to factors such as image resolution, region of interest (ROI) selection, and image processing. This study aimed to investigate the impact of the direction of ROI selection (DROIS) on fractal dimension (FD) calculations. Methods Panoramic radiographs of 226 individuals aged 20-35 years were analyzed. ROIs were selected on the mandibular condyle, angular region, and mental region, and oriented at 0°, 22.5°, 45°, and 67.5° angles. FD was calculated using the box-counting method in ImageJ. The Friedman test and Wilcoxon signed-rank test were used for statistical analysis. Results The FD values differed significantly between the angled ROI groups in all three regions (Friedman test, p < 0.0001). Pairwise comparisons showed significant differences in FD between most ROI orientations, except between 22.5° and 67.5° in the angular region. Conclusions DROIS is an important factor that should be considered in FA studies to ensure reliable and reproducible FD values. Appropriate methodological choices can help account for the influence of DROIS on FD calculations..

Objectives The aim of this study is to recommend an automatic caries detection and segmentation model based on the Convolutional Neural Network (CNN) algorithms in dental bitewing radiographs using VGG-16 and U-Net architecture and evaluate the clinical performance of the model comparing to human observer. Methods A total of 621 anonymized bitewing radiographs were used to progress the Artificial Intelligence (AI) system (CranioCatch, Eskisehir, Turkey) for the detection and segmentation of caries lesions. The radiographs were obtained from the Radiology Archive of the Department of Oral and Maxillofacial Radiology of the Faculty of Dentistry of Ordu University. VGG-16 and U-Net implemented with PyTorch models were used for the detection and segmentation of caries lesions, respectively. Results The sensitivity, precision, and F-measure rates for caries detection and caries segmentation were 0.84, 0.81; 0.84, 0.86; and 0.84, 0.84, respectively. Comparing to 5 different experienced observers and AI models on external radiographic dataset, AI models showed superiority to assistant specialists. Conclusion CNN-based AI algorithms can have the potential to detect and segmentation of dental caries accurately and effectively in bitewing radiographs. AI algorithms based on the deep-learning method have the potential to assist clinicians in routine clinical practice for quickly and reliably detecting the tooth caries. The use of these algorithms in clinical practice can provide to important benefit to physicians as a clinical decision support system in dentistry.