A lightweight detection algorithm for tooth cracks in optical images

• Published in: Computers in biology and medicine Vol. 182; p. 109153
• Main Authors: Xie, Zewen, Hu, Xian, Guo, Lide, Lin, Weiren, Liu, Jiakun, Zhang, Chunliang, Ge, Guanghua, Tang, Yadong, Wang, Wenlong
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.11.2024; Elsevier Limited
• Subjects: Adaptation; Algorithms; Artificial intelligence; Attention task; Automation; Classification; Cracked tooth detection; Cracked Tooth Syndrome - diagnostic imaging; Cracks; Datasets; Deep learning; Dentistry; Dentists; Fractures; Humans; Image enhancement; Image Processing, Computer-Assisted - methods; Impact analysis; Internal Medicine; Lightweight algorithm; Medical diagnosis; Medical research; Methods; Microcracks; Modules; Neural networks; Object detection; Object recognition; Optical image; Optical Imaging - methods; Other; Parameter identification; Performance evaluation; Semantics; Teeth; Thermal expansion; Tooth - diagnostic imaging; Tooth Fractures - diagnostic imaging; YOLOv8; Lightweight algorithm; Optical image; Cracked tooth detection; YOLOv8; Object detection
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2024.109153

Cracked tooth syndrome (CTS) is one of the major causes of tooth loss, presents the problem of early microcrack symptoms that are difficult to distinguish. This paper aims to investigate the practicality and feasibility of an improved object detection algorithm for automatically detecting cracks in dental optical images. A total of 286 teeth were obtained from Sun Yat-sen University and Guangdong University of Technology, and simulated cracks were generated using thermal expansion and contraction. Over 3000 images of cracked teeth were collected, including 360 real clinical images. To make the model more lightweight and better suited for deployment on embedded devices, this paper improves the YOLOv8 model for detecting tooth cracks through model pruning and backbone replacement. Additionally, the impact of image enhancement modules and coordinate attention modules on optimizing our model was analyzed. Through experimental validation, we conclude that that model pruning reduction maintains performance better than replacing a lightweight backbone network on a tooth crack detection task. This approach achieved a reduction in parameters and GFLOPs by 16.8 % and 24.3 %, respectively, with minimal impact on performance. These results affirm the effectiveness of the proposed method in identifying and labeling tooth fractures. In addition, this paper demonstrated that the impact of image enhancement modules and coordinate attention mechanisms on YOLOv8's performance in the task of tooth crack detection was minimal. An improved object detection algorithm has been proposed to reduce model parameters. This lightweight model is easier to deploy and holds potential for assisting dentists in identifying cracks on tooth surfaces. •Explored the feasibility of applying object detection algorithms to tooth crack syndrome.•Created an optical image dataset for tooth crack syndrome, containing real clinical data, intended for object detection tasks.•Improved the YOLOv8 and explored the impact of network pruning and backbone replacement on model complexity.