Computer Methods in Biomechanics and Biomedical Engineering最新文献

The COVID-19 pandemic has profoundly impacted health, emphasizing the need for timely disease detection. Blood tests have become key diagnostic tools due to the virus's effects on blood composition. Accurate COVID-19 prediction through machine learning requires selecting relevant features, as irrelevant features can lower classification accuracy. This study proposes Modified Mutual Information (MMI) for feature selection, ranking features by relevance and using backtracking to find the optimal subset. Support Vector Machines (SVM) are then used for classification. Results show that MMI with SVM achieves 95% accuracy, outperforming other methods, and demonstrates strong generalizability on various benchmark datasets.

To investigate the impact of patient-specific boundary conditions (BC) on blood flow and thrombosis modelling for patients with chronic thromboembolic pulmonary hypertension (CTEPH), three types of BCs were utilized to construct CTEPH models based on computed tomography pulmonary angiography images. First BC type is the patient-specific velocity profiles at the main pulmonary artery using phase contrast MRI (PC-MRI). The other two simplified types are the pulsatile BC and steady BC, which are obtained by spatially and temporally averaging the PC-MRI BC. Hemodynamic features including helical density, time-averaged wall shear stress (TAWSS) and oscillatory shear index (OSI), and thrombosis were compared for the three types BCs. The results indicated that, compared to the MRI BC, steady BC overestimated helical density and TAWSS in the pulmonary arteries by approximately 63.1% and 60%, respectively. The impact of simplified pulsatile BC on TAWSS and OSI in most regions of the pulmonary arteries was negligible with differences within 5%. Regarding thrombosis, the area predicted under pulsatile BC was approximately 80% smaller than that under PC-MRI BC. In conclusion, compared to PC-MRI BC, steady inlet BC tend to overestimate hemodynamic parameters, while pulsatile inlet BC yield similar wall shear stress based on parameters in most regions of the pulmonary artery. Patient-specific PC-MRI inlet BC should be used for accurate predictions of helical flow pattern and thrombus formation.

Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest diseases. Here are investigated two synthesized and two hypothetical coumarin derivatives, and their capacity to be used in the PDAC targeted treatment. The inhibitory activity of these four molecules against PARP, ATM, and CHK1 proteins responsible for DNA molecule repair was examined by docking and molecular dynamic analysis. ADMET analysis was applied to determine the pharmacokinetic properties of the tested compounds. The applied theoretical approach showed that the biomedical activity of the investigated coumarins is comparable to the inhibitory activity and pharmacokinetic properties of Olaparib, already used in the PDAC treatment.

Few studies consider the variability of the model parameters. This study aimed to perform a sensitivity analysis of a trapeziometacarpal joint model, by performing 675 finite element simulations built from the combination of different morphologies, joint passive stiffness, and grip strategies to estimate the joint pressure. Pressure variability was significantly more affected by morphology than grip strategy and joint passive stiffness. The effect of morphology and grip strategy on joint pressure was significant. A significant correlation between the trapezium dorso-volar curvature and the joint pressure was found. Morphology seems more important than the other parameters to estimate joint contact pressure correctly.

Breast cancer concerns 1 in 8 women in the world and is followed in 40% of cases by a mastectomy. Only 14% of women receive reconstructive surgery because of unfavorable clinical issues. The need of innovative tissue engineering devices leads Lattice Medical company to bring a new 3D-printed device, allowing the regeneration of soft tissue in order to replace the withdrawn breast. The implant, based on TEC (tissue engineering chamber) and fat-flat surgical technique, is constituted with bioresorbable thermosensitive materials to be fully absorbed by the body in several months, once the regeneration process is completed. In this industrial context, we need to assess some properties for predictive simulation: the TEC mechanical and biological properties over time, its sensitivity to implantation in the body temperature, its batch raw material variability and its structural 3D-printed behavior. This would lead to a more enlightened numerical design and topological optimization work. To do so, mechanical testing are conducted to gather necessaries information for fully border the behaviour of the material and eventually the impact of the process on the final prosthesis. Then, the G'sell Law is chosen to model the mechanical behaviour of the material taking into account all particularities of this medical case. Finally, the behaviour law is used in Finite Element Method (FEM) in a compression simulation to compare with experimental results which find good similarity in the mechanical response.

This study investigates NiTi-CoCr hybrid carotid artery stents to enhance mechanical properties over NiTi-only designs. Different configurations (24NiTi, 20NiTi-4CoCr, 16NiTi-8CoCr, and 12NiTi-12CoCr) were evaluated through radial compression and bending simulations. The 12NiTi-12CoCr stent showed the highest radial support (39.37 N) and increased bending strength by 77.96%. When modeled in a stenotic artery, this stent reduced stenosis from 81.52% to 29.33% and improved blood flow dynamics, alleviating high-pressure zones and balancing wall shear stress. These results suggest that CoCr wires improve stent performance, with the 12NiTi-12CoCr stent offering significant biomechanical and hemodynamic benefits.

Different forms of rugby may pose distinct risks to head injury. Video of rugby match footage was analyzed using head impact magnitude, frequency, and time interval for 15s and 7s athletes. Impacts were reconstructed in laboratory, and finite element modeling was used to estimate maximum principal strain. No difference was found in impact frequency or time interval between the two forms. Significantly more head impacts of higher severity levels were documented during 7s. These findings provide objective comparisons between 7s and 15s which may guide risk mitigation strategies in managing brain trauma for specific forms of rugby.

We developed an original computational model for cell deformation and migration capable of accounting for the cell sensitivity to the environment and its appropriate adaptation. This cell model is ultimately intended to be used to address tissue morphogenesis. Hence it has been designed to comply with four requirements: (1) the cell should be able to probe and sense its environment and respond accordingly; (2) the model should be easy to parametrize to adapt to different cell types; (3) the model should be able to extend to 3D cases; (4) simulations should be fast enough to integrate many interacting cells. The simulations carried out focused on two aspects: first, the general behaviour of the cell on a homogeneous substrate, as observed experimentally, for model validation. This enabled us to decipher the mechanisms by which the cell can migrate, highlighting respective influences of the adhesions lifetimes and their sensitivity to traction; second, it predicts the sensitivity of the cell to an anisotropic patterned substrate, in agreement with recently published experiments. The results show that mechanosensors simulated by the model make it possible to reproduce such experiments in terms of migration bias generated by the substrate anisotropy. Here again, the model provides a biomechanical explanation of this phenomenon, depending on cell-matrix interactions and adhesion maturation rate.

We examine an SIRS reaction-diffusion model with local dispersal and spatial heterogeneity to study COVID-19 dynamics. Using the operator semigroup approach, we establish the existence of disease-free equilibrium (DFE) and endemic equilibrium (EE), and derive the basic reproduction number, $R_0$. Simulations show that without dispersal, reinfection and limited medical resources problems can cause a plateau in cases. Dispersal and spatial heterogeneity intensify localised outbreaks, while integrated control strategies (vaccination and treatment) effectively reduce infection numbers and epidemic duration. The possibility of reinfection demonstrates the need for adaptable health measures. These insights can guide optimised control strategies for enhanced public health preparedness.

Liujunzi decoction (LJZD) is commonly used to treat hepatitis B virus (HBV), though its active ingredients and mechanisms are not fully known. This study identified core targets and active components of LJZD for treating hepatitis B (HB) through network pharmacology, molecular docking, and molecular dynamics simulation. Screening from databases yielded 533 active components, 2619 targets for LJZD, and 2910 for HB, with 891 intersecting targets. STRING and CytoHubba analyses identified AR and VDR as core targets, with key pathways including PI3K-Akt and MAPK. The findings clarify LJZD's multicomponent, multitarget mechanisms, supporting its clinical application for HB treatment.