International Journal for Numerical Methods in Biomedical Engineering最新文献

Sacroiliac joints (SIJ) injury has been recognised as a crucial cause of low back pain (LBP), but investigations on SIJ are insufficient. The aim of this study was to reveal the mechanism of SIJ injury induced by different vibration conditions and body inclinations and thus to analyse the relationship between vibration and LBP. Based on whole-body finite element models, eight load cases were analysed, where the multi-axis vibration loading cases were closer to the typical vehicle environment. In addition, three different body inclinations were involved to analyse the effect of body inclinations on vibration transmission of SIJ. The results showed that the vertical vibration would induce larger loads on SIJ than the fore-and-aft vibration. Single-axis vibrations tended to cause large fore-and-aft loads of SIJ, and multi-axis vibrations tended to cause large vertical loads. For the three body inclinations, static loads on SIJ were minimal at 100° and dynamic loads were minimal at 110°, which indicated that the 100° inclination was a recommended sitting posture for occupational drivers to reduce their incidence of LBP. These findings might be helpful for understanding the association between different vibration conditions and LBP, and provide reasonable ergonomics recommendations for occupational drivers to reduce the incidence of LBP.

Bone remodeling refers to the physiological behavior of bone tissue changes with changes in the biomechanical environment. Researches of bone remodeling process are significant for bone tissue engineering. The use of numerical simulation technology is an important means to analyze the bone remodeling process. The research of computational methods can deeply reveal the growth law of bone tissue under external load and environmental effects. This work aims to develop the computational model using a meshless method, and simulate an evolution of the porous structure of trabecular bone. The main research objectives include: (1) proposing a novel bone remodeling model based on the radial point interpolation method (RPIM), which integrates the mechanical and biological aspects of bone remodeling; (2) analyzing the theoretical foundations of the meshless method, including the mechanical model driven by strain energy stimulation and the biological model regulated by cell growth, death, and proliferation. This work and the presented cases are limited to two-dimensional areas. The results demonstrate that the proposed bone remodeling model can reflect the bone remodeling process and reflect the dynamic changes inside the bone throughout its life cycle. The developed computational algorithm is highly efficient, and can form the porous structure of trabecular bone through image fitting.

Poor long-term implant–bone fixation remains a significant clinical problem for Total Ankle Replacement (TAR). Recent advancements in metal additive manufacturing technology facilitate the development of porous lattice-structured implants. However, the optimal porous structure parameter, especially porosity, for the designing of porous implants for TAR remains ambiguous. The objective of the present study is to design porous diamond-structured tibial implant for TAR using macro–micro FE and Machine Learning (ML) based approach to enhance biomechanical and osseointegration performance. Four porous diamond architectures were designed with porosity varying from 50% to 80%. The study entails the macro-microscale FE modelling of four different porous diamond structured tibial implants (PDSTI) to assess the biomechanical performance and to perform bone ingrowth simulations using a physics based progressive mechanoregulatory tissue differentiation algorithm. ML models were used to predict the value of site-specific bone ingrowth, and the average value was compared between these PDSTI. ML models were accurately able to predict the site-specific bone ingrowth for each PDSTI. Bone ingrowth results revealed that the PDSTI at 70% porosity is more conducive to bone formation as well as better load transfer (less stress shielding) to the bone in comparison to a traditional solid implant. The lowest value of bone ingrowth was noted for the PDSTI at 80% porosity, indicating that large porosity is not suitable for bone ingrowth. The findings ultimately contribute to improving the clinical outcomes for TAR by reducing the risk of aseptic loosening.

The mechanical properties of spongy bone are a function of the tissue's microstructure including the strength and microarchitecture of the trabeculae. The thickness and number of trabeculae (morphological data) are the most important parameters that can indicate the porosity of tissue. Osteoporosis is known as a decrease in the thickness or loss of trabeculae that causes a reduction in the density and strength of bone. This research aims to model osteoporosis using cellular (lattice) structures and also presents mechanical relationships to predict tissue behavior during osteoporosis using morphological data for spongy tissue. Experimental trendlines (Young's modulus and relative density) are developed and equivalent lattice structures for osteoporotic spongy bone are presented using eight different types of unit cells such as cubic, hexagonal, tesseract, diamond, BCC, truncated cube, octahedral, and rhombic dodecahedron. Also, structural damage modes including trabeculae thinning, trabeculae elimination, and hybrid mode are implemented for determining the optimum geometrical characteristics of lattice structures for modeling osteoporosis. Comparison between normalized elastic moduli for unit cell structures indicates that the diamond unit cell exhibits less than 5% difference with experimental values. Also, for a reduction of relative density from 0.3 to 0.1 as a result of the hybrid damage mode, the normalized elastic moduli of the diamond numerical model and experimental result were reduced by about 0.013 and 0.0186, respectively. Finally, results show that among all the unit cell types, the diamond unit cell when used alongside the hybrid damage mode can predict the osteoporosis behavior of bone with the best accuracy.

In idiopathic pulmonary fibrosis (IPF), the juxtaposition of preserved regions and fibrotic areas creates heterogeneous lung mechanics and abnormal stress environments, which are hypothesized to activate mechanotransduction pathways and drive fibrosis progression. This study uses the “squishy ball lung” concept to quantify the potentially injurious mechanical stimuli arising from this heterogeneity. We developed a mechanical model that simulates the static inflation of an alveolus. This is described as a hyperelastic membrane with surface tension that is partially confined by springs representing fibrotic tissue. Finite element analysis (FEA) was used to assess the mechanical state under various confinement conditions. FEA revealed bulging deformation and significant meridian stress/strain peaks at transitions between confined and unconfined zones, which could potentially exceed safe physiological limits. To rigorously evaluate the predicted stress and strain environment, as well as its sensitivity to parameter uncertainty, such as material properties and the extent of confinement, we performed comprehensive uncertainty quantification (UQ) and quantitative sensitivity analysis (QSA). UQ confirmed the robustness of these localized stress peaks across parameter variations, while QSA identified the angle of confinement and spring stiffness as the primary determinants of peak stress magnitude. By quantifying these potentially injurious stress peaks, this study provides insights into the mechanical environment hypothesized to initiate mechanotransduction pathways in idiopathic pulmonary fibrosis (IPF), laying the groundwork for future studies that incorporate biological responses such as growth and remodeling.

Atherosclerosis remains the leading cause of cardiovascular morbidity worldwide. However, its onset and progression are still difficult to predict, and it remains unclear why certain arterial segments develop lesions while others remain unaffected. Recent findings highlight a prominent role of the vasa vasorum (VV)—the small blood vessels embedded within the walls of larger arteries—in driving disease development through an outside-in mechanism. In this view, perfusion deficits caused by vascular dysfunction may trigger chronic inflammation and promote plaque formation. To investigate this mechanism, we propose a novel multi-field model that combines tissue turnover, inflammation, and kinematics-based tissue growth. Perfusion is governed by a diffusion–reaction equation and accounts for VV dysfunction in supplying the outer arterial layers. Inflammation is captured through a phase-field representation that tracks the evolving interface between non-inflamed and inflamed tissue. A multiplicative decomposition of the deformation gradient then combines the inflammation-driven swelling and the homeostasis-driven tissue turnover, which itself is regulated by mechanical stress. The numerical implementation is realized using the standard finite element method. We assess model performance and plausibility through well-designed numerical case studies. The acquired simulation results highlight the coupled interaction among transport of blood-borne factors, inflammation, and mechanics, ultimately emphasizing how compromised VV can initiate a vicious cycle of ischemia, inflammation, and plaque growth in an outside-in fashion. In addition, we show how a moderate increase in blood pressure may result in a progressive increase in peak stress within atherosclerotic plaque tissue. Although our atherosclerosis model yields plausible predictions and allows deep insights into the interaction of mechanics, inflammation and tissue turnover, it is based on multiple modeling approximations, assumptions that would need sound validation in the future.

Machine learning (ML) models are becoming increasingly valuable for cardiovascular prediction and simulation, offering critical support for medical decision-making. These models are particularly useful for predicting disease progression and evaluating potential treatments. A major challenge in these models is to preserve the geometric fidelity of meshes while optimizing parameter efficiency to reduce memory usage, computational resources and execution time. In this paper, we present innovative approaches to abdominal aortic aneurysm (AAA) mesh compression, utilizing both statistical and deep learning models, with a focus on unsupervised learning techniques. We explore principal component analysis (PCA) as a statistical method and compare it with several deep learning models, including a simple autoencoder, an enhanced autoencoder based on PCA, a convolutional neural network (CNN), and a graph neural network (GNN). Human aortas are compressed using different statistical and deep learning methods to get the most relevant features. The mesh is reconstructed using the computed features and the error of the reconstructed meshes is compared. Our results indicate that PCA, using 64 principal components, outperforms deep learning models with a comparable latent space of 64, achieving the best overall performance. Among the deep learning approaches, the PCA-based autoencoder demonstrates the highest effectiveness.

Patients with lower limb lymphedema experience lymphatic fluid accumulation and swelling, which can progress to fibrosis and fat deposition in the soft tissues, impacting patients physically, socially, and psychologically. Compression therapy is one of the main treatments for lymphedema, but its effects on lymphedematous soft tissues are not yet fully understood. In this study, we developed a finite element model of a lymphedematous leg including subcutaneous and muscle tissues, as well as skin and fascia cruris, to investigate the hydrostatic pressure distribution resulting from the interface pressure applied by a compression stocking. The results highlight the significant influence of the leg's external geometry on the interface pressure, and demonstrate the importance of modeling skin to accurately predict hydrostatic pressure distribution in the subcutaneous tissue, with a 3.5% reduction in leg volume observed after compression. The outcomes improve the understanding of the effects of compression therapy on lower limbs affected by lymphedema and support the development of adapted treatment strategies for patients.