Combined Deep Learning Techniques for Mandibular Fracture Diagnosis Assistance

Abstract

Mandibular fractures are the most common fractures in dentistry. Since diagnosing a mandibular fracture is difficult when only panoramic radiographic images are used, most doctors use cone beam computed tomography (CBCT) to identify the patient’s fracture location. In this study, considering the diagnosis of mandibular fractures using the combined deep learning technique, YOLO and U-Net were used as auxiliary diagnostic methods to detect the location of mandibular fractures based on panoramic images without CBCT. In a previous study, mandibular fracture diagnosis was performed using YOLO learning; in the detection performance result of the YOLOv4-based mandibular fracture diagnosis module, the precision score was approximately 97%, indicating that there was almost no misdiagnosis. In particular, fractures in the symphysis, body, angle, and ramus tend to be distributed in the middle of the mandible. Owing to the irregular fracture types and overlapping location information, the recall score was approximately 79%, which increased the detection of undiagnosed fractures. In many cases, fractures that are clearly visible to the human eye cannot be grasped. To overcome these shortcomings, the number of undiagnosed fractures can be reduced using a combination of the U-Net and YOLOv4 learning modules. U-Net is advantageous for the segmentation of fractures spread over a wide area because it performs semantic segmentation. Consequently, the undiagnosed case in the middle of the mandible, where YOLO was weak, was somewhat supplemented by the U-Net module. The precision score of the combined module was 95%, similar to that of the previous method, and the recall score improved to 87%, as the number of undiagnosed cases was reduced. Through this study, the performance of a deep learning method that can be used for the diagnosis of the mandibular bone has been improved, and it is anticipated that as an auxiliary diagnostic inspection device, it will assist dentists in making diagnoses.

Semantic segmentation for tooth cracks using improved DeepLabv3+ model

“…Research has demonstrated that AI-based deep learning methods, such as wavelet-guided deep models, can outperform human observers in terms of diagnostic accuracy [ 21 , 22 ]. While certain deep learning methods have been shown to provide a relatively high diagnosis rate (exceeding 90% using binary classification neural networks [ 20 , 24 ] as well as combined U-Net and YOLOv4 network [ 37 ]), issues such as extended detection time and limited crack information suggest that semantic segmentation methods may offer more detailed and specific information. This can facilitate a more comprehensive understanding of the crack structure and aid in the formulation of subsequent treatment plans.…”

“…In traditional small object detection, DeepLabv3 demonstrates significant advantages compared to FCN and U-Net model [ 58 ]. With improvements to the ASPP module and the fully connected conditional random field module, DeepLabv3+ has achieved even better results in terms of mandibular fractures detection [ 37 ], particularly in detecting smaller targets such as cracks in terms of over-segmentation and precision. Based on our experiments DeepLabv3+ performs best in tooth crack segmentation compared to other commonly used networks.…”

“…With the development of computer vision technology, semantic segmentation methods have been widely used in medical image processing. In recent years, some scholars have proposed fracture detection models tailored to specific tasks by considering the unique features of fractures and background characteristics in different application scenarios, such as wrist cracks and bone fracture [ [34] , [35] , [36] , [37] ]. As illustrated in Fig.…”

A review on artificial intelligence for the diagnosis of fractures in facial trauma imaging

“…Because the diagnosis of mandibular fractures on panoramic radiographic (PR) images is quite difficult, CBCT imaging is preferred by radiologists and surgeons. To alleviate the unavailability of CBCT, Son et al (2022) combined two deep networks U-Net and YOLOv4 for detecting the location of mandibular fractures based on PR images without the use of CBCT. Another motivation of for the combination of the two deep networks was to minimize the number of likely undiagnosed fractures (symphysis, body, angle, ramus, condyle, and coronoid regions).…”

YOLOv9 for Fracture Detection in Pediatric Wrist Trauma X-ray Images

“…Compared to twostage object detectors, the models of YOLO series offer a more balanced combination of accuracy and inference speed, making them suitable for deployment on mobile computing platforms for medical image recognition. Son et al [28] utilized YOLOv4 [9] and U-Net [29] as auxiliary diagnostic tools to assist dentists in identifying mandibular fractures without resorting to cone beam computed tomography (CBCT). Jeon et al [30] employed YOLOv4 [9] to aid surgeons in diagnosing trauma by detecting the fracture and mapping it onto a 3D reconstructed bone image, providing a clear display of the fracture region through the red mask overlaid on the 3D bone image.…”