Akademik Veri Yönetim Sistemi
International Congress of DentoMaxilloFacial Radiology, London, İngiltere, 24 - 25 Haziran 2025, ss.36, (Özet Bildiri)
Aim: The temporomandibular joint (TMJ) capsule, a fibrous tissue that surrounds and stabilizes the joint, is crucial for assessing TMJ function, stability, and disease. This study aims to develop a deep learning model for the automatic detection and segmentation of TMJ capsular width on ultrasound (US) images.
Materials and Methods: A dataset of 490 TMJ ultrasound images was collected from the radiological archive of Eskisehir Osmangazi University Faculty of Dentistry. The images were labeled with capsular width measurements by a dentist using the CranioCatch annotation software (Eskisehir, Türkiye). The dataset was divided into training, validation, and test subsets. A YOLOv8x-based deep learning model was developed for the automatic detection and segmentation of TMJ capsular width on US images. Model performance was evaluated using precision, sensitivity, and F1 score metrics.
Results: The YOLOv8x-based AI model demonstrated strong performance in detecting and segmenting the TMJ capsular area in ultrasound images. The model achieved an F1 score of 0.98, with precision and sensitivity values of 1 and 0.96, respectively. These results indicate that the model effectively detects the capsular area across various images.
Conclusion: The YOLOv8x-based model successfully automates the detection and segmentation of TMJ capsular width in ultrasound images, enhancing diagnostic efficiency and consistency. This artificial intelligence tool has the potential to become a valuable asset in clinical practice for TMJ assessment and treatment planning.