Biomechanics and Modeling in Mechanobiology最新文献

The tympanic membrane (TM) plays a pivotal role in auditory transduction by converting acoustic signals into mechanical energy. Damage to the TM, whether from chronic perforations, infections, trauma, or pathological changes such as tympanosclerosis, can significantly impair its biomechanical function and lead to conductive hearing loss. Tympanoplasty remains the standard intervention, traditionally utilising autologous grafts such as temporalis fascia or cartilage. However, these biological materials present limitations, including donor site morbidity, variable mechanical properties, and potential for long-term resorption. To overcome these challenges, synthetic scaffolds engineered from polymers such as polylactic acid (PLA), polycaprolactone (PCL), and silk fibroin have emerged as promising alternatives. This review provides a comprehensive examination of the mechanical characteristics of both healthy and diseased tympanic membranes, offering insights into their structural integrity and mechanical performance. Emphasis is placed on key parameters such as thickness, Young’s modulus, tensile strength, strain tolerance, and viscoelastic behaviour, drawing from ex vivo and in vivo studies. The discussion extends to computational strategies, particularly finite element modelling (FEM) and inverse FEM, which enable accurate simulation of TM responses under physiological and pathological conditions. By linking empirical mechanical data with computational analyses, this review supports the rational design of synthetic grafts that closely mimic native tissue properties. Additionally, emerging trends in TM tissue engineering, including 3D printing and biomimetic scaffolds, are also highlighted for their potential to improve surgical outcomes in tympanoplasty.

The abnormal accumulation of low-density lipoprotein (LDL) can lead to aortic atherosclerosis. However, the aortic LDL transfer and its relationship with hemodynamics are still not fully explored. This study aims to reveal the mechanism of heat and LDL transfer in healthy aortas, leveraging a fluid–structure interaction (FSI) model. Two healthy aortic geometry models were reconstructed based on clinical computed tomography angiography images. The flow rate used as the inlet boundary condition was taken from another previous publication, and non-invasive blood pressure measurement data were exploited to determine the parameters of the three-element Windkessel model for boundary conditions of the aortic outlets. The aortic wall was assumed to be uniform in thickness, and the hyperelastic material was simulated by Yeoh second-order model. Our two-way FSI method was further developed to predict the heat and LDL transfer. Results show that the correlation coefficients of time-averaged LDL, temperature, and wall shear stress (WSS)-related indices between the rigid and hyperelastic aortic wall are high (> 0.914) except for the topological shear variation index (TSVI). The interaction between the blood flow and the aortic wall is suggested to be considered to accurately capture the distribution of oscillatory shear index, relative residual time (RRT), and TSVI. Besides, we find that there is a positive correlation (> 0.596) between the concentration of LDL and aortic wall temperature. The long RRT region also coincides with the high LDL area, which negatively correlates with WSS and TSVI. This study demonstrates the heat and LDL transfer in healthy aortas using the FSI model, and the findings would inform novel strategies to measure LDL concentration and regulate its accumulation.

Large-scale axonal dynamic simulation is critical to study white matter injury but is prohibitive in computational cost. We solve this challenge by training a convolutional neural network (CNN) that takes fiber strain profiles as inputs to instantly estimate multimodal axonal injury parameters. First, tractography-based fiber strains are derived based on subject-specific simulations of N = 46 head impacts from a male ice hockey player. To generate the minimum training dataset, the brain is subdivided into coarse cubes (isotropic resolution of 6 mm; N = 4979 voxels). A stratified (one sample per cube) and adaptive (by controlling a similarity threshold) sampling strategy is devised to iteratively identify the most distinct profiles from N = 45 head impacts used for training (with the remaining one reserved for independent validation). They serve as the input to a male axonal injury model for simulation. A CNN is then trained to estimate the peak strains in microtubule and axolemma as well as the failure percentages of tau proteins and neurofilaments. The CNN is cross-validated to determine the minimum training samples of N = 2000 to reach ({R}^{2})>0.90. Under the “worst case scenario” for independent validation (N = 75 testing samples identified), the CNN achieves an ({R}^{2}) of 0.91–0.98 and a normalized root mean-squared error (NRMSE) of 2.7–5.0%. Finally, we showcase the CNN by generating high-resolution multimodal axonal responses for the entire white matter within 12 s (isotropic resolution of 2 mm with ~ 92,500 voxels), vs. an estimated ~ 12 years using conventional direct simulations (~ 31.5-million-fold efficiency gain). This study demonstrates the potential of deep learning to enable large-scale mechanistic investigations of white matter injury in the future.

Aortic valve replacement is a cornerstone treatment for severe aortic valve diseases, including stenosis and regurgitation. Suboptimal valve seating can elevate the transvalvular pressure gradient, while valve orientation and size may produce flow jets that impinge on the ascending aorta, potentially weakening the vessel wall. Such hemodynamic complications can compromise valve performance and patient outcomes. This study presents a computational fluid dynamics framework, derived from medical CT images, for preprocedural hemodynamic assessment of aortic valve replacement. The framework minimizes user input and delivers rapid results, enabling efficient evaluation of valve types, orientations, and their hemodynamic impact. The results demonstrate that non-optimal implantation angles substantially increase pressure drop across the valve, thereby imposing higher workload on the heart. This automated and efficient simulation framework demonstrates strong potential for clinical application, supporting precise planning and execution of valve implantation procedures to improve patient care.

Osteoarthritis induces profound structural degeneration of articular cartilage, with existing treatments remaining largely ineffective. This study pioneers a mechanoregulatory model utilizing histology-based finite element analysis to predict depth-dependent glycosaminoglycan (GAG) adaptations in both intact and damaged human cartilage under mechanical loading. Uniquely calibrated through rigorous one-week longitudinal in vitro experiments in intact cartilage, our model correctly predicts depth-dependent GAG content adaptation, also in damaged cartilage. Notably, the model reveals potential effects of fluid velocity and dissipated energy on an increase in GAG content, while highlighting the degenerative effects of maximum shear strain under physiological loading conditions. Interestingly, it replicates enhanced GAG production in damaged cartilage, consistent with our experimental observations. Beyond advancing the fundamental understanding of mechanical loading in cartilage homeostasis, this innovative model offers a robust platform for in silico trials, enabling the development of personalized rehabilitation protocols to optimize mechanical loading strategies for degenerative joint diseases. Our work represents a significant leap forward in leveraging computational tools to address the challenges of osteoarthritis treatment. All findings are based on human explants from one donor and should be interpreted as preliminary proof-of-concept.

Glioblastoma multiforme (GBM) remains a formidable challenge due to its aggressive proliferation, heterogeneity, and invasiveness. This review synthesizes biomechanical models for GBM prediction, from classic proliferation–invasion (PI) frameworks—based on reaction–diffusion equations—to continuum biomechanical models that quantify tumor-induced stress and tissue interactions. We highlight multiphysics approaches integrating fluid dynamics, nutrient transport, and solid mechanics to simulate the tumor microenvironment, alongside numerical methods like FEM and meshless techniques. Treatment modeling, including radiotherapy and emerging therapies, is critically evaluated for optimizing clinical strategies. Challenges in validation and parameterization are addressed, with a forward-looking emphasis on hybrid physics-informed and machine learning models to enable personalized prediction. By bridging biophysics, computation, and clinical needs, this work aims to guide future research toward improved GBM therapeutics.

The number of patients undergoing hemodialysis has been steadily increasing in recent decades. Arteriovenous fistula (AVF) is the gold standard for ensuring vascular access in these patients. Despite the prominent role of AVFs in hemodialysis treatment, their maturation and long-term functionality continue to pose challenges as less than a third of fistulas remain patent without further interventions in a 3-year follow-up. Computational biomechanics has become an essential tool for clarifying mechanical conditions accompanying the pathogenesis of various vascular complications, including suboptimal maturation and AVF stenosis. Constitutive description plays a crucial role in the design of computational models and without it simulations remain only at the rigid tube level. However, literature on the mechanical properties and constitutive modeling of upper extremity veins is lacking. This study aims to fill this gap by characterizing the mechanical properties of the human basilic vein (BV) and comparing it to the great saphenous vein (GSV). Uniaxial tensile tests in two perpendicular directions were used to obtain the mechanical response of the tissue. The results suggest that BVs do not significantly differ from GSVs in their elastic properties expressed by means of the tangent modulus. Overall anisotropy, understood as the difference in elastic moduli obtained in different directions, seems to be reduced in BVs. The 4-fiber family exponential model of the strain energy density function was adopted to fit the experimental data. The model fitted the data well, as suggested by the coefficients of determination R², which ranged from 0.97 to 0.99 for the majority of the average curves. The resulting parameter values can be used within the modeling of the mechanical behavior of veins in computational simulations of vascular access performance.

Venoarterial extracorporeal membrane oxygenation (VA ECMO) is an advanced life-saving therapy for patients with severe cardiopulmonary failure. Understanding the performance of the drainage cannula is critical to minimizing complications such as thrombosis formation, platelet activation, and circuit failure. This study utilizes computational fluid dynamics (CFD) to analyze the flow characteristics within the drainage cannula under both normal vessel conditions and vessel collapse scenarios. The simulations focus on flow behavior, shear stress distribution, and regions prone to platelet accumulation and thrombus formation. In the collapsed vessel scenario, significant alterations in flow patterns were observed, including elevated shear stress, increased velocities near the cannula tip, and flow redistribution along the cannula holes. While the collapsed condition exhibited higher mechanical platelet activation due to increased shear forces, improved washout resulted in a lower accumulation of activated platelets compared to the normal condition. Additionally, thrombosis-prone regions were identified, particularly near the cannula tip for normal drainage condition. The findings of this study highlight the fluid flow mechanisms contributing to thrombosis risk in the drainage cannula during VA ECMO. These insights can inform cannula design improvements to minimize thrombosis and optimize ECMO performance.