An effective way for dealing with element distortion by nearest‐nodes FEM

Chen

2012

Proc Inst Mech Eng H

Self Cite

The mechanism of cerebrospinal fluid in mitigating closed head injuries caused by mild impacts was investigated by finite-element modeling. Three biomechanical models were constructed. In these models, cerebrospinal fluid was considered as a soft solid material, an inviscid fluid without intracranial pressure, and an inviscid fluid with normal intracranial pressure, respectively, while other conditions such as the finite-element mesh, the impact, and the boundary conditions were kept the same. The recently developed nearest nodes finite-element method was adopted to deal with large deformations in brain tissue. Results obtained from the numerical studies showed that cerebrospinal fluid was able to remarkably reduce the maximum peak strains, especially the shear strains induced by impacts and transmitted to the brain. Cerebrospinal fluid with intracranial pressure was able to further buffer relative oscillations between the skull and the brain.

Section: Construction Of Finite-element Modelsmentioning

confidence: 99%

Finite-element study of cerebrospinal fluid in mitigating closed head injuries

Chen

2012

Proc Inst Mech Eng H

Self Cite

“…The current authors have implemented a preliminary version of the above procedure in MATLAB codes. The recently developed nearest-nodes finite element method [27] was adopted owing to its higher convergence rate and its capability in dealing with distorted finite elements [28].…”

Section: Construction Of Dxa-based Patient-specific Finite Element Modelmentioning

confidence: 99%

A preliminary dual-energy X-ray absorptiometry-based finite element model for assessing osteoporotic hip fracture risk

Ferdous

Leslie

2011

Proc Inst Mech Eng H

To more accurately assess osteoporotic hip fracture risk in a specific patient, a dual-energy X-ray absorptiometry (DXA)-based finite element model was constructed from the patient's femur DXA image. The outermost contour of the femur bone segmented from the DXA image was used to generate a finite element mesh. Bone mechanical properties, such as Young's modulus, are correlated with areal bone mineral density (BMD) captured in the DXA image. A quasi-static loading condition representing a sideway fall was applied to the finite element model. Three fracture risk indices were introduced and expressed as ratios of internal forces caused by impact forces occurring in sideway fall to bone ultimate cross-section strength at the three critical locations, i.e. the femoral neck, the intertrochanteric region, and the subtrochanteric region. The proposed finite element modelling procedure was validated against six representative clinical cases extracted from the Manitoba BMD database, where initial and follow-up DXA images have been taken to monitor longitudinal variation of areal BMD in individual patients. It was found from the clinical validation that variations in the proposed fracture risk indices have the same trends as those indicated by the conventional areal BMD and T-score. In addition, by the three proposed fracture risk indices it is possible to further identify the specific fracture location. It was also found that for the same subject, the variations in the three fracture risk indices have quite different magnitudes, with intertrochanteric region the largest and subtrochanteric region the smallest, which is probably owing to the different content of trabecular and cortical bones in the three regions. With further development, it is promising that the proposed DXA-based finite element model will be a useful tool for accurate assessment of osteoporosis development and for treatment monitoring.

“…by using a set of nodes that are the nearest to a concerned quadrature point in the element, and nodes from adjacent elements may be involved in the construction. The most attractive merit of the NN-FEM is that its performance is nearly not affected by element distortion [8]. The minimum requirement on finite element mesh by the NN-FEM is there are no overlapped or self-penetrated elements [7].…”

Section: Introductionmentioning

confidence: 99%

Improved Computational Efficiency of Nearest-Nodes Finite Element Method (NN-FEM)

2013

AMM

The recently developed nearest-nodes finite element method (NN-FEM) has a number of advantages over the conventional finite element method (FEM). The most attractive one is that its performance is nearly not affected by element distortion. However, low computational efficiency of NN-FEM is a major concern, as in the original NN-FEM a local problem has to be solved at each quadrature point to construct shape functions there. In this paper, a new strategy is introduced in NN-FEM for constructing shape functions aiming at improving its computational efficiency. The strategy is, for regular-shape elements, shape functions are constructed at the element center and the obtained shape functions are used at all quadrature points in the element; only for severely distorted elements, shape functions are constructed separately at each quadrature point. Numerical results show that computational efficiency of the NN-FEM can be significantly improved with the above strategy, while other performance of the method is not affected.