Three-Dimensional Assessment of Temporomandibular Joint Using MRI-CBCT Image Registration

Multi-class segmentation of temporomandibular joint using ensemble deep learning

Temporomandibular joint disorders are prevalent causes of orofacial discomfort. Diagnosis predominantly relies on assessing the configuration and positions of temporomandibular joint components in magnetic resonance images. The complex anatomy of the temporomandibular joint, coupled with the variability in magnetic resonance image quality, often hinders an accurate diagnosis. To surmount this challenge, we developed deep learning models tailored to the automatic segmentation of temporomandibular joint components, including the temporal bone, disc, and condyle. These models underwent rigorous training and validation utilizing a dataset of 3693 magnetic resonance images from 542 patients. Upon evaluation, our ensemble model, which combines five individual models, yielded average Dice similarity coefficients of 0.867, 0.733, 0.904, and 0.952 for the temporal bone, disc, condyle, and background class during internal testing. In the external validation, the average Dice similarity coefficients values for the temporal bone, disc, condyle, and background were 0.720, 0.604, 0.800, and 0.869, respectively. When applied in a clinical setting, these artificial intelligence-augmented tools enhanced the diagnostic accuracy of physicians, especially when discerning between temporomandibular joint anterior disc displacement and osteoarthritis. In essence, automated temporomandibular joint segmentation by our deep learning approach, stands as a promising aid in refining temporomandibular joint disorders diagnosis and treatment strategies.

A Three-branch Jointed Feature and Topology Decoder guided by game-theoretic interactions for temporomandibular joint segmentation

Segmentation of the temporomandibular joint (TMJ) disc and condyle from magnetic resonance imaging (MRI) is a crucial task in TMJ internal derangement research. The automatic segmentation of the disc structure presents challenges due to its intricate and variable shapes, low contrast, and unclear boundaries. Existing TMJ segmentation methods often overlook spatial and channel information in features and neglect overall topological considerations, with few studies exploring the interaction between segmentation and topology preservation. To address these challenges, we propose a Three-Branch Jointed Feature and Topology Decoder (TFTD) for the segmentation of TMJ disc and condyle in MRI. This structure effectively preserves the topological information of the disc structure and enhances features. We introduce a cross-dimensional spatial and channel attention mechanism (SCIA) to enhance features. This mechanism captures spatial, channel, and cross-dimensional information of the decoded features, leading to improved segmentation performance. Moreover, we explore the interaction between topology preservation and segmentation from the perspective of game theory. Based on this interaction, we design the Joint Loss Function (JLF) to fully leverage the features of segmentation, topology preservation, and joint interaction branches. Results on the TMJ MRI dataset demonstrate the superior performance of our TFTD compared to existing methods.

The utilization of decision trees on orthopantomographic and lateral panoramic graphs for the diagnosis of unilateral anterior disc displacement of the temporomandibular joint

BACKGROUND

Investigation is to utilize decision trees in conjunction with orthopantomography (OPT) and lateral panoramic graphy (LPG) to diagnose unilateral anterior disc displacement (ADD) of the temporomandibular joint.

METHODS

In this study, 161 patients with images obtained through all three imaging methods, MRI, OPT, and LPG, were selected from the archives. The participants were categorized into two groups: the study group, comprising 89 patients with unilateral anterior disc displacement, and the control group, consisting of 72 healthy individuals. Measurements, including 2 angles (antero-posterior angle and superior-inferior angle) and 3 distance parameters (anterior joint space distance, superior joint space distance, and posterior joint space distance), were conducted on each imaging modality dataset. To assess the obtained measurement data within each patient, the differences from each measurement were calculated. Statistical analysis of the measurement differences between the control and study groups was carried out with independent t test, and decision trees were generated using the SPSS 25 decision tree module 5.0.

RESULTS

In ADD patients, it was statistically significantly found that the APA increased while the SIA decreased for angle measurements. But for linear measurements, AS increased while the SS and PS decreased in MRI, OPT, and LPG.

CONCLUSION

ADD can be diagnosed in OPT and LPG. The identification of the specific type of ADD that occurs in the temporomandibular joint is not feasible.

Effect of TMJ disc position on condylar bone remodeling after arthroscopic disc repositioning surgery

PURPOSE

The objective of this study was to analyze the effect of TMJ disc position on condylar bone remodeling after arthroscopic disc repositioning surgery.

METHODS

Nine patients with anterior disc displacement without reduction (ADDWoR, 15 sides) who underwent arthroscopic disc repositioning surgery were included. Three-dimensional (3D) reconstruction of the articular disc and the condyle in the closed-mouth position was performed using cone-beam computed tomography (CBCT) and magnetic resonance imaging (MRI) data. Then, the CBCT and MRI images were fused and displayed together by multimodal image registration techniques. Morphological changes in the articular disc and condyle, as well as changes in their spatial relationship, were studied by comparing preoperative and 3-month postoperative CBCT-MRI fused images.

RESULTS

The volume and superficial area of the articular disc, as well as the area of the articular disc surface in the subarticular cavity, were significantly increased compared to that before the surgical treatment（P < 0.01）. There was also a significant increase in the volume of the condyle (P < 0.001). All condyles showed bone remodeling after surgery that could be categorized as one of two types depending on the position of the articular disc, suggesting that the location of the articular disc was related to the new bone formation.

CONCLUSIONS

The morphology of the articular disc and condyle were significantly changed after arthroscopic disc repositioning surgery. The 3D changes in the position of the articular disc after surgery tended to have an effect on condylar bone remodeling and the location of new bone formation.

Explainable deep learning-based clinical decision support engine for MRI-based automated diagnosis of temporomandibular joint anterior disk displacement

BACKGROUND AND OBJECTIVE

MRI is considered the gold standard for diagnosing anterior disc displacement (ADD), the most common temporomandibular joint (TMJ) disorder. However, even highly trained clinicians find it difficult to integrate the dynamic nature of MRI with the complicated anatomical features of the TMJ. As the first validated study for MRI-based automatic TMJ ADD diagnosis, we propose a clinical decision support engine that diagnoses TMJ ADD using MR images and provides heat maps as the visualized rationale of diagnostic predictions using explainable artificial intelligence.

METHODS

The engine builds on two deep learning models. The first deep learning model detects a region of interest (ROI) containing three TMJ components (i.e., temporal bone, disc, and condyle) in the entire sagittal MR image. The second deep learning model classifies TMJ ADD into three classes (i.e., normal, ADD without reduction, and ADD with reduction) within the detected ROI. In this retrospective study, the models were developed and tested on the dataset acquired between April 2005 to April 2020. The additional independent dataset acquired at a different hospital between January 2016 to February 2019 was used for the external test of the classification model. Detection performance was assessed by mean average precision (mAP). Classification performance was assessed by the area under the receiver operating characteristic (AUROC), sensitivity, specificity, and Youden's index. 95% confidence intervals were calculated via non-parametric bootstrap to assess the statistical significance of model performances.

RESULTS

The ROI detection model achieved mAP of 0.819 at 0.75 intersection over union (IoU) thresholds in the internal test. In internal and external tests, the ADD classification model achieved AUROC values of 0.985 and 0.960, sensitivities of 0.950 and 0.926, and specificities of 0.919 and 0.892, respectively.

CONCLUSIONS

The proposed explainable deep learning-based engine provides clinicians with the predictive result and its visualized rationale. The clinicians can make the final diagnosis by integrating primary diagnostic prediction obtained from the proposed engine with the patient's clinical examination findings.

Temporomandibular joint segmentation in MRI images using deep learning

OBJECTIVES

Temporomandibular joint (TMJ) internal derangements (ID) represent the most prevalent temporomandibular joint disorder (TMD) in the population and its diagnosis typically relies on magnetic resonance imaging (MRI). TMJ articular discs in MRIs usually suffer from low resolution and contrast, and it is difficult to identify them. In this study, we applied two convolutional neural networks (CNN) to delineate mandibular condyle, articular eminence, and TMJ disc in MRI images.

METHODS

The models were trained on MRI images from 100 patients and validated on images from 40 patients using 2D slices and 3D volume as input, respectively. Data augmentation and five-fold cross-validation scheme were applied to further regularize the models. The accuracy of the models was then compared with four raters having different expertise in reading TMJ-MRI images to evaluate the performance of the models.

RESULTS

Both models performed well in segmenting the three anatomical structures. A Dice coefficient of about 0.7 for the articular disc, more than 0.9 for the mandibular condyle, and Hausdorff distance of about 2mm for the articular eminence were achieved in both models. The models reached near-expert performance for the segmentation of TMJ articular disc and performed close to the expert in the segmentation of mandibular condyle and articular eminence. They also surpassed non-experts in segmenting the three anatomical structures.

CONCLUSION

This study demonstrated that CNN-based segmentation models can be a reliable tool to assist clinicians identifying key anatomy on TMJ-MRIs. The approach also paves the way for automatic diagnosis of TMD.

CLINICAL SIGNIFICANCE

Accurately locating the articular disc is the hardest and most crucial step in the interpretation of TMJ-MRIs and consequently in the diagnosis of TMJ-ID. Automated software that assists in locating the articular disc and its surrounding structures would improve the reliability of TMJ-MRI interpretation, save time and assist in reader training. It will also serve as a foundation for additional automated analysis of pathology in TMJ structures to aid in TMD diagnosis.

Diagnostic efficacy of CBCT, MRI, and CBCT-MRI fused images in distinguishing articular disc calcification from loose body of temporomandibular joint

OBJECTIVES

To evaluate the diagnostic efficacy of CBCT-MRI fused images for articular disc calcification of temporomandibular joint (TMJ).

MATERIALS AND METHODS

Twenty patients (24 TMJs) whose image examinations showed dense bodies in the TMJ space were included in the study. The locations of dense bodies evaluated by the three experts were used as a reference standard. Three oral and maxillofacial radiology residents evaluated whether the dense bodies were disc calcification or not, with a five-point scale for four sets of images (CBCT alone, MRI alone, both CBCT and MRI observed at a time, and CBCT-MRI fused images) randomly and independently. Each set of images was observed at least 1 week apart. A second evaluation was performed after 4 weeks. Intraclass correlation coefficients were calculated to assess the intra- and inter-observer agreement. The areas under the ROC curves (AUCs) were compared between the four image sets using Z test.

RESULTS

Ten cases were determined as articular disc calcifications, and fourteen cases were recognized as loose bodies in the TMJ spaces. The average AUC index for the CBCT-MRI fused images was 0.95 and significantly higher than the other sets (p < 0.01). The intra- and inter-observer agreement in the CBCT-MRI fused images (0.90-0.91, 0.93) was excellent and higher than those in the other images.

CONCLUSIONS

CBCT-MRI fused images can significantly improve the observers' reliability and accuracy in determining articular disc calcification of the TMJ.

CLINICAL RELEVANCE

The multimodality image fusion is feasible in detecting articular disc calcification of the TMJ which are hard to define by CBCT or MRI alone. It can be utilized especially for inexperienced residents to shorten the learning curve and improve diagnostic accuracy.

Cone Beam Computed Tomography in Oral and Maxillofacial Surgery: An Evidence-Based Review

Cone Beam Computed Tomography (CBCT) is a valuable imaging technique in oral and maxillofacial surgery (OMS) that can help direct a surgeon’s approach to a variety of conditions. A 3-dimensional analysis of head and neck anatomy allows practitioners to plan appropriately, operate with confidence, and assess results post-operatively. CBCT imaging has clear indications and limitations. CBCT offers the clinician 3-dimensional and multi-planar views for a more accurate diagnosis and treatment without the financial burden and radiation exposure of conventional computed tomography (CT) scans. Furthermore, CBCT overcomes certain limitations of 2-dimensional imaging, such as distortion, magnification, and superimposition. However, CBCT lacks the detailed depiction of soft tissue conditions for evaluation of pathologic conditions, head and neck infections, and temporomandibular joint (TMJ) disc evaluation. This review evaluates the evidence-based research supporting the application of CBCT in the various fields of oral and maxillofacial surgery, including dentoalveolar surgery, dental implants, TMJ, orthognathic surgery, trauma, and pathology, and will assess the value of CBCT in pre-operative assessment, surgical planning, and post-operative analysis when applicable. Additionally, the significant limitations of CBCT and potential areas for future research will be discussed.

In vivo prediction of temporomandibular joint disc thickness and position changes for different jaw positions

Temporomandibular joint disorders (TMD) are common dysfunctions of the masticatory region and are often linked to dislocation or changes of the temporomandibular joint (TMJ) disc. Magnetic resonance imaging (MRI) is the gold standard for TMJ imaging but standard clinical sequences do not deliver a sufficient resolution and contrast for the creation of detailed meshes of the TMJ disc. Additionally, bony structures cannot be captured appropriately using standard MRI sequences due to their low signal intensity. The objective of this study was to enable researchers to create high resolution representations of all structures of the TMJ and consequently investigate morphological as well as positional changes of the masticatory system. To create meshes of the bony structures, a single computed tomography (CT) scan was acquired. In addition, a high-resolution MRI sequence was produced, which is used to collect the thickness and position change of the disc for various static postures using bite blocks. Changes in thickness of the TMJ disc as well as disc translation were measured. The newly developed workflow successfully allows researchers to create high resolution models of all structures of the TMJ for various static positions, enabling the investigation of TMJ disc translation and deformation. Discs were thinnest in the lateral part and moved mainly anteriorly and slightly medially. The procedure offers the most comprehensive picture of disc positioning and thickness changes reported to date. The presented data can be used for the development of a biomechanical computer model of TMJ anatomy and to investigate dynamic and static loads on the components of the system, which could be useful for the prediction of TMD onset.

Evaluation of condyle position in patients with Angle Class I, II, and III malocclusion using cone-beam computed tomography panoramic reconstructions

OBJECTIVES

This study was performed to compare the positions of the right and left condyles between male and female patients with different Angle malocclusions using cone-beam computed tomography (CBCT) panoramic reconstructions.

METHODS

The CBCT images of 60 patients (age of 18-37 years) were retrospectively evaluated. The patients were divided according to their Angle malocclusion classifications (Angle Classes I, II, and III). The condyle-to-eminence, condyle-to-fossa, and condyle-to-meatus distances were measured digitally using i-CAT software.

RESULTS

The left and right condyle-to-fossa distances were the most variable parameters among the Angle classes. The right condyle-to-eminence and right condyle-to-fossa distances were significantly different among the classes. Male patients seemed to have a greater condyle-to-fossa distance on the right side in both the Class I and III groups. The mean distance from the condyle to eminence, condyle to fossa, and condyle to meatus on the right side was the greatest in the Angle Class II group.

CONCLUSIONS

In all three types of malocclusion (Angle Classes I, II, and III), the condyles on both the right and left sides were not exactly symmetric or centrally located within the glenoid fossa. This work emphasizes the differences in the condyle position between male and female patients. Furthermore, the symmetry and centricity of the condyles are not dependent on the patient's sex or type of malocclusion.

Three-dimensional CBCT images registration method for TMJ based on reconstructed condyle and skull base

OBJECTIVES

A method was introduced for three-dimensional (3D) cone-beamCT (CBCT) images registration of temporomandibular joint (TMJ). This study aimed to provide quantitative and qualitative analysis of TMJ bone changes in two-dimensional (2D) and 3D and to provide the technique for computer-aided diagnosis of temporomandibular joint disorders in the future.

METHODS

10 TMJ samples of six patients were obtained from Peking University Hospital of Stomatology. Four of the six patients imaged bilateral TMJs and the other two patients only imaged unilateral TMJ. Each sample consisted of two images from the same TMJ taken at different times. First, condyle and skull base were segmented semi-automatically for 3D model reconstruction. Then the segmented condyle and skull base were registered separately. Registration process can be divided into two processes of rough registration and fine registration. Rough registration step was achieved by selecting corresponding points manually and initialized fine registration. Condyle and skull base were fine registered by minimizing mean square error of condyle (MSE_condyle) and skull base (MSE_skull) respectively. Qualitative assessment of osseous component changes utilized 2D color-fused model and 3D surface-fused model and quantitative analyses the convergence of this method used the mean square error of the model (MSE_model). Independent repeated experiments were carried out to test the stability of our 3D registration method.

RESULTS

Sufficiently alignment was achieved. Osseous abnormality and morphology changes were displayed using fusion model. MSE_model of condylar registration and skull base registration declined 51.80% and 64.58% compared with that before registration. Quantitative analysis verified the stability of the method.

CONCLUSIONS

The proposed method completed 3D TMJ registration for different physiological structure. The result of this method was accurate, reproducible and not relied on the experience of operators.

Three-dimensional morphological changes of the temporomandibular joint and functional effects after mandibulotomy

BACKGROUND

The midline and paramedian mandibulotomy are surgical procedures that divide the mandibular bone into two halves and disconnects the condylar heads of the TMJ from each other. This study aimed to prospectively evaluate the temporomandibular joint (TMJ) functional and morphological changes after mandibulotomy using a reconstructed 3D models of the TMJ.

METHODS

Sixteen adult patients diagnosed with oral and oropharyngeal tumors with planned surgical mandibulotomy (test group, 9 patients) or transoral (control group, seven patients) treatments were included in the study. MRI and CBCT images were obtained immediately preceeding surgery and 6-8 weeks after surgery. Using the MRI-CBCT registered images, TMJ tissues were segmented at the two occasions by the same operator and 3D models were reconstructed for morphological assessment. Changes across time were measured using the volume overlap and Hausdorff distance of the disc and condyle 3D models. Disc-condyle relationship was measured using point-based and color map analysis. To assess the early functional changes, the Jaw function limitation scale (JFLS) and the maximum mouth opening were measured. Two-sample Hotelling T² t-test was performed to determine the significance of the morphological and clinical outcomes' differences between the two groups.

RESULTS

The two-sample Hotelling T² t-test showed significant differences (T² (df1,df2) = 0.97 (5,26), p <0.01) between the mean values of all outcomes among the 2 groups. The change in disc displacement was significantly different between the two groups (p <0.05). However, the condylar displacement was not significantly different between the two groups (p =0.3). The average of the JFLS score was five times larger after mandibulotomy, and was 2 times larger after transoral surgery (p < 0.01). Patients showed decrease in the average of the maximum interincisal mouth opening by 11 mm after mandibulotomy, and by 5.4 mm after transoral surgery.

CONCLUSION

The quantitative assessment of the TMJ showed minimal changes of the condylar position and variable degrees of articular disc displacement associated with the paramedian split mandibulotomy. As well, limited jaw functions and vertical mouth opening were noticed more in the mandibultomy group compared to the transoral group in 6- weeks after surgery.