Objective: This study investigated knowledge, attitudes, and perceptions regarding the future of artificial intelligence (AI) for radiological diagnosis among a group of Turkish dental students. Methods: An online survey was conducted consisting of 11 questions using Google Forms and circulated among 4th and 5th grade students at Marmara University, Faculty of Dentistry. The survey consisted of questions regarding participants’ recognition of and attitudes toward AI, their opinions on directions of AI development, and their perceptions about the future of AI in oral radiology. IBM SPSS Statistics 25.0 (IBM SPSS, Turkey) program is used for statistical analysis. Results: The study group consists of 75 4th and 65 5th grades and a total of 140 students. Of the 140 participating dental students, 60.0 % were already familiar with the concept of AI, 92.9% agreed stated that they would like to use a software/program that can be helpful in radiological diagnosis and 37.9 % reported that AI would have a future in Turkey. Among two grades, there was no statistically significant difference of answers to questions regarding the future and role of artificial intelligence in oral radiology (p>0.05). Conclusion: According to the findings of the study, most dental students were aware of AI, AI systems could be used to improve diagnostic accuracy when reading radiographs, and AI has a promising role in radiological diagnosis.

Objectives: The purpose of the present study was to examine ultrasonographic appearances of Masseter Muscle (MM) in dentate and edentulous patients without Temporomandibular Disorder (TMD). Materials and Methods: The thickness of the MM in 25 dentate (mean age: 30,68 ± 10,49) and 24 edentulous (mean age: 61,46 ± 9,71) patients, who visited routine dental examination, was measured at rest and at maximum contraction bilaterally. Examinations were performed using an Aloka Prosound α6 (Hitachi Aloka Medical Systems, Tokyo, Japan) equipped with an 8 MHz-wide bandwidth linear active matrix transducer (ranging from 1 to 15 MHz). The visibility and width of the internal echogenic bands of the MM were also assessed and the muscle appearance was classified as I of III types. Type I, characterized by the clear visibility of the fine bands; Type II, thickening echogenicity of the bands; Type III, disappearance or reduction in a number of the bands. Results: MM thickness at rest and contraction in the dentate group were significantly higher than the edentulous group (p <0.05). Type I was the most common echogenic type in both dentate (right:16 (64%), left; 15 (60%)) and edentulous patients (right; 22 (91.7%), left; 18 (75%)). In a dentate group, type II was significantly higher than the edentulous group in both the right and left sides (p <0.05; p <0.01, respectively). Age and gender seemed to have no significant effect on the echogenic type (p ˃0.05). Conclusion: There were significant differences in the thickness at rest and contraction between the dentate and edentulous groups. It was clarified that ultrasonographic features of the MM in dentate and edentulous patients were different.

Aim: Deep learning algorithms have lately been used for medical image processing, and they have showed promise in a range of applications. The purpose of this study was to develop and test computer-based diagnostic tools for evaluating masseter muscle segmentation on ultrasonography images. Materials and methods: A total of 388 anonymous adult masseter muscle retrospective ultrasonographic images were evaluated. The masseter muscle was labeled on ultrasonography images using the polygonal type labeling method with the CranioCatch labeling program (CranioCatch, Eskişehir, Turkey). All images were re-checked and verified by Oral and Maxillofacial Radiology experts. This data set was divided into training (n = 312), verification (n = 38) and test (n = 38) sets. In the study, an artificial intelligence model was developed using PyTorch U-Net architecture, which is a deep learning approach. Results: In our study, the artificial intelligence deep learning model known as U-net provided the detection and segmentation of all test images, and when the success rate in the estimation of the images was evaluated, the F1, sensitivity and precision results of the model were 1.0, 1.0 and 1.0, respectively. Conclusion: Artificial intelligence shows promise in automatic segmentation of masseter muscle on ultrasonography images. This strategy can aid surgeons, radiologists, and other medical practitioners in reducing diagnostic time.