CMF-Net: craniomaxillofacial landmark localization on CBCT images using geometric constraint and transformer.

Accurate and robust anatomical landmark localization is a mandatory and crucial step in deformation diagnosis and treatment planning for patients with craniomaxillofacial (CMF) malformations. In this paper, we propose a trainable end-to-end cephalometric landmark localization framework on Cone-beam computed tomography (CBCT) scans, referred to as CMF-Net, which combines the appearance with transformers, geometric constraint, and adaptive wing (AWing) loss. More precisely: (1) we decompose the localization task into two branches: the appearance branch integrates transformers for identifying the exact positions of candidates, while the geometric constraint branch at low resolution allows the implicit spatial relationships to be effectively learned on the reduced training data. (2) We use the AWing loss to leverage the difference between the pixel values of the target heatmaps and the automatic prediction heatmaps. We verify our CMF-Net by identifying the 24 most relevant clinical landmarks on 150 dental CBCT scans with complicated scenarios collected from real-world clinics. Comprehensive experiments show that it performs better than the state-of-the-art deep learning methods, with an average localization error of 1.108 mm (the clinically acceptable precision range being 1.5 mm) and a correct landmark detection rate equal to 79.28%. Our CMF-Net is time-efficient and able to locate skull landmarks with high accuracy and significant robustness. This approach could be applied in 3D cephalometric measurement, analysis, and surgical planning.

Free and Forced Vibration Analyses of Functionally Graded Graphene-Nanoplatelet-Reinforced Beams Based on the Finite Element Method.

The finite element method (FEM) is used to investigate the free and forced vibration characteristics of functionally graded graphene-nanoplatelet-reinforced composite (FG-GPLRC) beams. The weight fraction of graphene nanoplatelets (GPLs) is assumed to vary continuously along the beam thickness according to a linear, parabolic, or uniform pattern. For the FG-GPLRC beam, the modified Halpin-Tsai micromechanics model is used to calculate the effective Young's modulus, and the rule of mixture is used to determine the effective Poisson's ratio and mass density. Based on the principle of virtual work under the assumptions of the Euler-Bernoulli beam theory, finite element formulations are derived to analyze the free and forced vibration characteristics of FG-GPLRC beams. A two-node beam element with six degrees of freedom is adopted to discretize the beam, and the corresponding stiffness matrix and mass matrix containing information on the variation of material properties can be derived. On this basis, the natural frequencies and the response amplitudes under external forces are calculated by the FEM. The performance of the proposed FEM is assessed, with some numerical results obtained by layering method and available in published literature. The comparison results show that the proposed FEM is capable of analyzing an FG-GPLRC beam. A detailed parametric investigation is carried out to study the effects of GPL weight fraction, distribution pattern, and dimensions on the free and forced vibration responses of the beam. Numerical results show that the above-mentioned effects play an important role with respect to the vibration behaviors of the beam.

Landmark Localization for Cephalometric Analysis Using Multiscale Image Patch-Based Graph Convolutional Networks.

Accurate and robust cephalometric image analysis plays an essential role in orthodontic diagnosis, treatment assessment and surgical planning. This paper proposes a novel landmark localization method for cephalometric analysis using multiscale image patch-based graph convolutional networks. In detail, image patches with the same size are hierarchically sampled from the Gaussian pyramid to well preserve multiscale context information. We combine local appearance and shape information into spatialized features with an attention module to enrich node representations in graph. The spatial relationships of landmarks are built with the incorporation of three-layer graph convolutional networks, and multiple landmarks are simultaneously updated and moved toward the targets in a cascaded coarse-to-fine process. Quantitative results obtained on publicly available cephalometric X-ray images have exhibited superior performance compared with other state-of-the-art methods in terms of mean radial error and successful detection rate within various precision ranges. Our approach performs significantly better especially in the clinically accepted range of 2 mm and this makes it suitable in cephalometric analysis and orthognathic surgery.

Comparison of Pain-Generated Functional Outcomes in Experimental Models of Delayed-Onset Muscle Soreness and Nerve Growth Factor Injection of the Masticatory Muscles.

AIMS: To compare two pain models of myalgic TMD, delayed-onset muscle soreness (DOMS) and injections of nerve growth factor (NGF), in terms of pain-related and motor function outcomes, as well as activity-related temporal summation. METHODS: Fifty age- and gender-matched healthy participants were recruited and randomized into one of three groups: to a repeated eccentric contraction task to cause DOMS (n = 20), to receive NGF injections into the masseter muscle (n = 20), or to a control group (n = 10). Mechanical sensitivity of masticatory muscles, chewing parameters, jaw function limitation, maximum bite force, and activity-related temporal summation were assessed at baseline and at days 1, 2, and 7 following the intervention. RESULTS: Compared to baseline, both model groups showed increased mechanical sensitivity, jaw function limitation, pain on chewing, and decreased chewing efficiency, lasting longer in the NGF group than in the DOMS group (P < .05). Furthermore, also compared to baseline, the NGF group showed increased pain on maximum bite and decreased pain-free maximum opening (P < .05). No increases in activity-related temporal summation were shown for any of the model groups when compared to baseline or the control group (P > .05). CONCLUSION: Both models produced similar pain-related outcomes, with the NGF model having a longer effect. Furthermore, the NGF model showed a more substantial effect on motor function, which was not seen for the DOMS model. Finally, neither of the models were able to provoke activity-related temporal summation of pain.

Effect of transcutaneous electrical nerve stimulation on jaw movement-evoked pain in patients with TMJ disc displacement without reduction and healthy controls.

Objective: Transcutaneous electrical nerve stimulation (TENS) may serve as non-invasive intervention for painful temporomandibular disorders (TMD) to improve jaw motor function, but its efficacy is still debated. This parallel study evaluated the effect of TENS on pain and movement patterns after repeated jaw movements in patients with painful temporomandibular joints (TMJ) and disc displacement without reduction (DDwoR), and compared with healthy controls.Material and Methods: 20 patients with TMJ pain and DDwoR and 20 age- and gender-matched healthy volunteers were randomly assigned to TENS/sham TENS (sTENS) intervention groups in a block design (10 in each group). Participants performed 20 repeated jaw movements (4 x 5 sessions), and reported pain intensity on a 0-10 Numerical Rating Scale (NRS) subsequently both before and after the intervention. Data were tested by repeated measures analysis of variance (ANOVA).Results: Significant increase of pain intensity and reduction of opening range were shown within repeated jaw movements in TMJ pain patients in contrast to healthy participants (p ≤ .001). Pain was significantly reduced during repeated open-close (p = .007), fast open-close (p = .016) and horizontal movements (p = .023), accompanied with increased opening range (p = .033) and open-close velocity (p = .019) with TENS intervention when compared with sTENS group (p > .05) in TMJ pain patients.Conclusions: This study indicated that movement-evoked pain was reduced either spontaneously or by sTENS in TMJ pain patients with DDwoR, and interestingly, that TENS could attenuate movement-evoked pain and improve jaw motor function during repeated jaw movements. The findings may have implications for TENS treatment in TMJ pain patients with DDwoR.

Temporal summation and motor function modulation during repeated jaw movements in patients with temporomandibular disorder pain and healthy controls.

Temporal summation of nociceptive inputs may be an important pathophysiological mechanism in temporomandibular disorders (TMD) pain; however, it remains unknown how natural jaw function relates to underlying pain mechanisms. This study evaluated changes in pain and movement patterns during repeated jaw movements in patients with painful temporomandibular joints (TMJ) compared with healthy controls. Twenty patients with TMD with TMJ pain, and an anterior disk displacement without reduction and 20 age- and gender-matched healthy volunteers were included. Participants performed 20 trials (4 × 5 sessions) of standardized and repeated mandibular movements, and scored the movement-associated pain intensity on 0 to 10 numeric rating scale in addition to measurements of jaw movements. Patients with TMJ pain reported higher baseline pain compared to the control group for all types of jaw movements (P = 0.001) and significant increases in numeric rating scale pain scores by repetition of jaw movements (P < 0.001), which was not observed in the control group (P > 0.05). Jaw total opening distance (P = 0.030), maximum opening velocity (P = 0.043) and average closing velocity (P = 0.044) in the TMJ pain group were significantly reduced during the repeated movements. In the control group, however, total opening distance (P = 0.499), maximum opening velocity (P = 0.064), and average closing velocity (P = 0.261) remained unchanged, whereas average opening velocity (P = 0.040) and maximum closing velocity (P = 0.039) increased. The study demonstrates that repeated jaw movements constitute a sufficient and adequate stimulation for triggering temporal summation effects associated with significant inhibition of motor function in painful TMJs. These findings have practical implications for diagnosis of TMD pain and for more mechanism-driven management protocols in the future.

Detection and quantification of intracerebral and intraventricular hemorrhage from computed tomography images with adaptive thresholding and case-based reasoning.

PURPOSE:    Hemorrhage within the brain space (HWBS) involves the brain parenchyma and ventricle systems, and is associated with high morbidity and mortality. Computed tomography (CT) head scans are the recommended modality for diagnosis and treatment for HWBS. However, HWBS detection may be difficult when the hemorrhage is inconspicuous, while quantification is hard as hemorrhage can have very variable intensity that overlaps with normal brain tissue. An algorithm is proposed to detect and quantify HWBS. METHODS: Adaptive thresholding and case-based reasoning (CBR) were applied to HWBS in four steps: preprocessing to extract the brain, adaptive thresholding based on local contrast with varied window sizes to derive candidate HWBS regions, case representation to represent each candidate HWBS region by parameters on context as well as intensity and geometrical characteristics, and classification of HWBS by taking each candidate HWBS region as a case and applying CBR. Additionally, case base indexing and weights optimization were used to increase retrieval speed and improve performances. Refinement of each recognized HWBS was performed for quantifying HWBS. RESULTS: Validation on 426 clinical CT data indicates that the proposed algorithm achieved a detection rate of 94.4 % and recall of 79.2 % for detecting HWBS regions. Visually, the HWBS regions calculated from adaptive thresholding plus refinement agreed well with expert delineation. For 10 representative data with small to large hemorrhage, the algorithm quantitatively yielded a segmentation accuracy of [Formula: see text]. Case base indexing increased the retrieval speed by 41.1 times at the expense of decreasing detection rate of 0.5 % and recall of 2.6 %. Genetic algorithm optimization enhanced the detection rate and recall to, respectively, 94.9 and 83.5 %. CONCLUSIONS: We developed and tested an algorithm that combined adaptive thresholding and CBR for detecting and quantifying HWBS. Experiments showed that adaptive thresholding could provide suitable candidates, while CBR was able to identify HWBS regions. The proposed method has potential as a new tool for accurately detecting and quantifying HWBS.