Explainable Deep Learning and Biomechanical Modeling for TMJ Disorder Morphological Risk Factors.

Clarifying multifactorial musculoskeletal disorder etiologies supports risk analysis and development of targeted prevention and treatment modalities. Deep learning enables comprehensive risk factor identification through systematic analysis of disease datasets but does not provide sufficient context for mechanistic understanding, limiting clinical applicability for etiological investigations. Conversely, multiscale biomechanical modeling can evaluate mechanistic etiology within the relevant biomechanical and physiological context. We propose a hybrid approach combining 3D explainable deep learning and multiscale biomechanical modeling; we applied this approach to investigate temporomandibular joint (TMJ) disorder etiology by systematically identifying risk factors and elucidating mechanistic relationships between risk factors and TMJ biomechanics and mechanobiology. Our 3D convolutional neural network recognized TMJ disorder patients through subject-specific morphological features in condylar, ramus, and chin. Driven by deep learning model outputs, biomechanical modeling revealed that small mandibular size and flat condylar shape were associated with increased TMJ disorder risk through increased joint force, decreased tissue nutrient availability and cell ATP production, and increased TMJ disc strain energy density. Combining explainable deep learning and multiscale biomechanical modeling addresses the "mechanism unknown" limitation undermining translational confidence in clinical applications of deep learning and increases methodological accessibility for smaller clinical datasets by providing the crucial biomechanical context.

Structure-function relationships of TMJ lateral capsule-ligament complex.

The human temporomandibular joint (TMJ) lateral capsule ligament (LCL) complex is debated as a fibrous capsule with distinct ligaments or ligamentous thickening, necessitating further evaluation of the complex and its role in TMJ anatomy and mechanics. This study explores the ultrastructural arrangement, biomechanical tensile properties, and biochemical composition of the human LCL complex including region-specific differences to explore the presence of a distinct temporomandibular ligament and sex-specific differences to inform evaluations of potential etiological mechanisms. LCL complex ultrastructural arrangement, biomechanical properties, and biochemical composition were determined using cadaveric samples. Statistical modeling assessed sex- and region-specific effects on LCL complex tissue properties. Collagen fiber coherency, collagen fiber bundle size, and elastin fiber count did not differ between sexes, but females trended higher in elastin fiber count. LCL complex water and sGAG content did not differ between sexes or regions, but collagen content was higher in the anterior region (311.0 ± 185.6 µg/mg) compared to the posterior region (221.0 ± 124.9 µg/mg) (p = 0.045) across sexes and in males (339.6 ± 170.6 µg/mg) compared to females (204.5 ± 130.7 µg/mg) (p = 0.006) across regions. Anterior failure stress (1.1 ± 0.7 MPa) was larger than posterior failure stress (0.6 ± 0.4 MPa) (p = 0.024). Regional differences confirm the presence of a mechanically and compositionally distinct temporomandibular ligament. Baseline sex-specific differences are critical for etiological investigations of sex disparities in TMJ disorders. These results have important biomechanical and clinical ramifications, providing critical baseline tissue material properties, informing the development of TMJ musculoskeletal models, and identifying new areas for etiologic investigations for temporomandibular disorders.