Real-time segmentation and detection of ponticulus posticus in lateral cephalometric radiographs using YOLOv8: a step towards enhanced clinical evaluation

Abstract

Objectives: Ponticulus posticus (PP) is a bony structure in the cervical spine, often difficult to identify in radiographic images, and its detection is important for both orthodontic diagnosis and clinical decision-making related to craniovertebral pathologies. The purpose of this study is to develop a deep learning-based approach for detecting the PP in lateral cephalometric radiographs using the YOLOv8-seg model. Methods: This retrospective study analyzed a dataset of 1000 anonymized lateral cephalometric radiographs, focusing on the segmentation and detection of the PP. Images were resized to 640 × 640 pixels and labeled by two experienced dentomaxillofacial radiologists. The YOLOv8-seg model, designed for segmentation tasks, was trained over 100 epochs with a batch size of sixteen, using pre-trained weights from the COCO dataset. Model performance was evaluated using precision, recall, mean average precision (mAP), and F1 score metrics. Results: The YOLOv8s-seg model demonstrated high accuracy in detecting the PP, with a precision of 62.81%, recall of 88.7%, mAP50 of 75.27%, mAP95 of 62.28%, and an F1 score of 73.54%. Even in cases where the boundaries of the C1 cervical vertebra were not clearly distinguishable, the model performed effectively, suggesting its reliability in clinical applications. Conclusions: The proposed YOLOv8-seg model shows promising potential for improving the accuracy and efficiency of PP detection in lateral cephalometric radiographs. By integrating AI into the diagnostic process, orthodontic practices can benefit from more precise and reliable identification of small but clinically significant anatomical structures, ultimately enhancing patient care and diagnostic accuracy.