Automated segmentation of articular disc of the temporomandibular joint on magnetic resonance images using deep learning

Abstract

Temporomandibular disorders are typically accompanied by a number of clinical manifestations that involve pain and dysfunction of the masticatory muscles and temporomandibular joint. The most important subgroup of articular abnormalities in patients with temporomandibular disorders includes patients with different forms of articular disc displacement and deformation. Here, we propose a fully automated articular disc detection and segmentation system to support the diagnosis of temporomandibular disorder on magnetic resonance imaging. This system uses deep learning-based semantic segmentation approaches. The study included a total of 217 magnetic resonance images from 10 patients with anterior displacement of the articular disc and 10 healthy control subjects with normal articular discs. These images were used to evaluate three deep learning-based semantic segmentation approaches: our proposed convolutional neural network encoder-decoder named 3DiscNet (Detection for Displaced articular DISC using convolutional neural NETwork), U-Net, and SegNet-Basic. Of the three algorithms, 3DiscNet and SegNet-Basic showed comparably good metrics (Dice coefficient, sensitivity, and positive predictive value). This study provides a proof-of-concept for a fully automated deep learning-based segmentation methodology for articular discs on magnetic resonance images, and obtained promising initial results, indicating that the method could potentially be used in clinical practice for the assessment of temporomandibular disorders.

Automation and deep (machine) learning in temporomandibular joint disorder radiomics: A systematic review

“…Therefore, it is imperative to include the years of experience possessed by the clinical operators annotating the training datasets to exclude the likelihood of inaccurate classifications. For context, only 40% of the investigations reported on the clinical operators' years of experience prior to data annotation 20,23,28–32,34 . From the available information, the average years of experience held by the clinical operator was 13.7 years with at least 3 years of experience observed per operator.…”

“…The current review revealed several areas for improvement in the administrative handling of machine learning projects. It was seen that the majority of the funds and ethical approvals to use clinical TMJ data for deep learning were provided in South Korea (25%), 18,20,23,29,33 USA (25%) 10,22,24–26 and Japan (15%) 30,31,36 . and held by key research teams.…”

Analysis of the feasibility of using deep learning for multiclass classification of dental anomalies on panoramic radiographs

“…In the past 10 years, there has been increasing interest in the application of deep learning to dental treatment [3][4][5][6][7][22][23][24] . Medical images are regarded as the primary input for AI applications in healthcare.…”

“…From 2015 to March 2020, the United States Food and Drug Administration approved 222 AI-equipped medical devices, including three dental devices 1) . AI research related to dentistry is actively underway [3][4][5][6][7] .…”

“New Normal” Radiology

“…This preparation may be achieved by “deep mutational scanning” observing “surface antigens drifts and shifts of the pathogens in silico using artificial intelligence and machine learning.” 6 In the meanwhile, the dentists should review all existing technology, particularly radiographic, adapting and developing it further so that it not only addresses new infection control challenges, reducing the radiation dose to their patients, but also improves the quality of the display of the patient and the accompanying disease. This also means developing technologies that do not use ionising radiation, such as qualitative light-induced fluorescence for the diagnosis of early caries 37 and magnetic resonance imaging (MRI) optimally used for temporomandibular joint diseases 38 and lesions of the salivary glands and other soft-tissue of the face, jaws, and upper neck. Although MRI units are generally only sited in medical facilities, ultrasonography can be used in the dental office.…”