Constitutive model of brain tissue suitable for finite element analysis of surgical procedures.

Realistic finite element modelling and simulation of neurosurgical procedures present a formidable challenge. Appropriate, finite deformation, constitutive model of brain tissue is a prerequisite for such development. In this paper, a large deformation, linear, viscoelastic model, suitable for direct use with commercially available finite element software packages such as ABAQUS is constructed. The proposed constitutive equation is of polynomial form with time-dependent coefficients. The model requires four material constants to be identified. The material constants were evaluated based on unconfined compression experiment results. The analytical as well as numerical solutions to the unconfined compression problem are presented. The agreement between the proposed theoretical model and the experiment is good for compression levels reaching 30% and for loading velocities varying over five orders of magnitude. The numerical solution using the finite element method matched the analytical solution very closely.     

Mechanical properties of brain tissue in-vivo: experiment and computer simulation

Realistic computer simulation of neurosurgical procedures requires incorporation of the mechanical properties of brain tissue in the mathematical model. Possible applications of computer simulation of neurosurgery include non-rigid registration, virtual reality training and operation planning systems and robotic devices to perform minimally invasive brain surgery. A number of constitutive models of brain tissue, both single-phase and bi-phasic, have been proposed in recent years. The major deficiency of most of them, however, is the fact that they were identified using experimental data obtained in vitro and there is no certainty whether they can be applied in the realistic in vivo setting. In this paper we attempt to show that previously proposed by us hyper-viscoelastic constitutive model of brain tissue can be applied to simulating surgical procedures. An in vivo indentation experiment is described. The force-displacement curve for the loading speed typical for surgical procedures is concave upward containing no linear portion from which a meaningful elastic modulus might be determined. In order to properly analyse experimental data, a three-dimensional, non-linear finite element model of the brain was developed. Magnetic resonance imaging techniques were used to obtain geometric information needed for the model. The shape of the force-displacement curve obtained using the numerical solution was very similar to the experimental one. The predicted forces were about 31% lower than those recorded during the experiment. Having in mind that the coefficients in the model had been identified based on experimental data obtained in vitro, and large variability of mechanical properties of biological tissues, such agreement can be considered as very good. By appropriately increasing material parameters describing instantaneous stiffness of the tissue one is able, without changing the structure of the model, to reproduce experimental curve almost perfectly. Numerical studies showed also that the linear, viscoelastic model of brain tissue is not appropriate for the modelling brain tissue deformation even for moderate strains.     

Continuum description of the Poisson׳s ratio of ligament and tendon under finite deformation

Ligaments and tendons undergo volume loss when stretched along the primary fiber axis, which is evident by the large, strain-dependent Poisson?s ratios measured during quasi-static tensile tests. Continuum constitutive models that have been used to describe ligament material behavior generally assume incompressibility, which does not reflect the volumetric material behavior seen experimentally. We developed a strain energy equation that describes large, strain dependent Poisson?s ratios and nonlinear, transversely isotropic behavior using a novel method to numerically enforce the desired volumetric behavior. The Cauchy stress and spatial elasticity tensors for this strain energy equation were derived and implemented in the FEBio finite element software (www.febio.org). As part of this objective, we derived the Cauchy stress and spatial elasticity tensors for a compressible transversely isotropic material, which to our knowledge have not appeared previously in the literature. Elastic simulations demonstrated that the model predicted the nonlinear, upwardly concave uniaxial stress–strain behavior while also predicting a strain-dependent Poisson?s ratio. Biphasic simulations of stress relaxation predicted a large outward fluid flux and substantial relaxation of the peak stress. Thus, the results of this study demonstrate that the viscoelastic behavior of ligaments and tendons can be predicted by modeling fluid movement when combined with a large Poisson?s ratio. Further, the constitutive framework provides the means for accurate simulations of ligament volumetric material behavior without the need to resort to micromechanical or homogenization methods, thus facilitating its use in large scale, whole joint models.     

Subject-specific non-linear biomechanical model of needle insertion into brain

The previous models for predicting the forces acting on a needle during insertion into very soft organs (such as, e.g. brain) relied on oversimplifying assumptions of linear elasticity and specific experimentally derived functions for determining needle-tissue interactions. In this contribution, we propose a more general approach in which the needle forces are determined directly from the equations of continuum mechanics using fully non-linear finite element procedures that account for large deformations (geometric non-linearity) and non-linear stress-strain relationship (material non-linearity) of soft tissues. We applied these procedures to model needle insertion into a swine brain using the constitutive properties determined from the experiments on tissue samples obtained from the same brain (i.e. the subject-specific constitutive properties were used). We focused on the insertion phase preceding puncture of the brain meninges and obtained a very accurate prediction of the needle force. This demonstrates the utility of non-linear finite element procedures in patient-specific modelling of needle insertion into soft organs such as, e.g. brain.     

Prediction of brain deformations and risk of traumatic brain injury due to closed-head impact: quantitative analysis of the effects of boundary conditions and brain tissue constitutive model

In this study, we investigate the effects of modelling choices for the brain–skull interface (layers of tissues between the brain and skull that determine boundary conditions for the brain) and the constitutive model of brain parenchyma on the brain responses under violent impact as predicted using computational biomechanics model. We used the head/brain model from Total HUman Model for Safety (THUMS)—extensively validated finite element model of the human body that has been applied in numerous injury biomechanics studies. The computations were conducted using a well-established nonlinear explicit dynamics finite element code LS-DYNA. We employed four approaches for modelling the brain–skull interface and four constitutive models for the brain tissue in the numerical simulations of the experiments on post-mortem human subjects exposed to violent impacts reported in the literature. The brain–skull interface models included direct representation of the brain meninges and cerebrospinal fluid, outer brain surface rigidly attached to the skull, frictionless sliding contact between the brain and skull, and a layer of spring-type cohesive elements between the brain and skull. We considered Ogden hyperviscoelastic, Mooney–Rivlin hyperviscoelastic, neo–Hookean hyperviscoelastic and linear viscoelastic constitutive models of the brain tissue. Our study indicates that the predicted deformations within the brain and related brain injury criteria are strongly affected by both the approach of modelling the brain–skull interface and the constitutive model of the brain parenchyma tissues. The results suggest that accurate prediction of deformations within the brain and risk of brain injury due to violent impact using computational biomechanics models may require representation of the meninges and subarachnoidal space with cerebrospinal fluid in the model and application of hyperviscoelastic (preferably Ogden-type) constitutive model for the brain tissue.     

Identification of linear viscoelastic constitutive models

Accelerations induce in the brain mechanical stresses that may explain the loss of consciousness feared by fighter pilots. In this study, the brain is modelled as a multi-domain structure and a finite element method is used to identify the constitutive law parameters of each domain and then to analyse the stress level in the brain. The loading and observed strain rates induced by hypergravity seem to indicate a quasi-static behaviour of the brain structure. A general procedure has been developed to characterise the behaviour of a structure including several domains. Each of them is assumed to be isotropic and homogeneous with a linear viscoelastic behaviour. These constitutive laws were identified using only the displacements of several nodes on the envelope discarding the displacements between domains at the interaction surfaces. These interfaces may be buried inside the structure and not connected with the external surface. Two validation examples are proposed to show the reliability and effectiveness of the method.     

A transversely isotropic constitutive model of excised guinea pig spinal cord white matter

Narrowing of the spinal canal generates an amalgamation of stresses within the spinal cord parenchyma. The tissue’s stress state cannot be quantified experimentally; it must be described using computational methods, such as finite element analysis. The objective of this research was to propose a compressible, transversely isotropic constitutive model, an augmentation of the isotropic Mooney–Rivlin hyperelastic strain energy function, to describe the guinea pig spinal cord white matter. Model parameters were derived from a combination of inverse finite element analysis on transverse compression experiments and least squared error analysis applied to quasi-static longitudinal tensile tests. A comparison of the residual errors between the predicted response and the experimental measurements indicated that the transversely isotropic constitutive law that incorporates an offset stretch reduced the error by a factor of four when compared to other commonly used models.     

The non-linear response of a muscle in transverse compression: assessment of geometry influence using a finite element model

Most recent finite element models that represent muscles are generic or subject-specific models that use complex, constitutive laws. Identification of the parameters of such complex, constitutive laws could be an important limit for subject-specific approaches. The aim of this study was to assess the possibility of modelling muscle behaviour in compression with a parametric model and a simple, constitutive law. A quasi-static compression test was performed on the muscles of dogs. A parametric finite element model was designed using a linear, elastic, constitutive law. A multi-variate analysis was performed to assess the effects of geometry on muscle response. An inverse method was used to define Young's modulus. The non-linear response of the muscles was obtained using a subject-specific geometry and a linear elastic law. Thus, a simple muscle model can be used to have a bio-faithful, biomechanical response.     

Fitted hyperelastic parameters for Human brain tissue from reported tension,compression, and shear tests

The mechanical properties of human brain tissue are the subject of interest because of their use in understanding brain trauma and in developing therapeutic treatments and procedures. To represent the behavior of the tissue, we have developed hyperelastic mechanical models whose parameters are fitted in accordance with experimental test results. However, most studies available in the literature have fitted parameters with data of a single type of loading, such as tension, compression, or shear. Recently, Jin et al. (Journal of Biomechanics 46:2795−2801, 2013) reported data from ex vivo tests of human brain tissue under tension, compression, and shear loading using four strain rates and four different brain regions. However, they do not report parameters of energy functions that can be readily used in finite element simulations. To represent the tissue behavior for the quasi-static loading conditions, we aimed to determine the best fit of the hyperelastic parameters of the hyperfoam, Ogden, and polynomial strain energy functions available in ABAQUS for the low strain rate data, while simultaneously considering all three loading modes. We used an optimization process conducted in MATLAB, calling iteratively three finite element models developed in ABAQUS that represent the three loadings. Results showed a relatively good fit to experimental data in all loading modes using two terms in the energy functions. Values for the shear modulus obtained in this analysis (897−1653 Pa) are in the range of those presented in other studies. These energy-function parameters can be used in brain tissue simulations using finite element models.     

Modified Bilston nonlinear viscoelastic model for finite element head injury studies

This paper proposes a modified nonlinear viscoelastic Bilston model (Bilston et al., 2001, Biorheol., 38, pp. 335-345). for the modeling of brain tissue constitutive properties. The modified model can be readily implemented in a commercial explicit finite element (FE) code, PamCrash. Critical parameters of the model have been determined through a series of rheological tests on porcine brain tissue samples and the time-temperature superposition (TTS) principle has been used to extend the frequency to a high region. Simulations by using PamCrash are compared with the test results. Through the use of the TTS principle, the mechanical and rheological behavior at high frequencies up to 10(4) rads may be obtained. This is important because the properties of the brain tissue at high frequencies and impact rates are especially relevant to studies of traumatic head injury. The averaged dynamic modulus ranges from 130 Pa to 1500 Pa and loss modulus ranges from 35 Pa to 800 Pa in the frequency regime studied (0.01 rads to 3700 rads). The errors between theoretical predictions and averaged relaxation test results are within 20% for strains up to 20%. The FEM simulation results are in good agreement with experimental results. The proposed model will be especially useful for application to FE analysis of the head under impact loads. More realistic analysis of head injury can be carried out by incorporating the nonlinear viscoelastic constitutive law for brain tissue into a commercial FE code.     

An orthotropic viscoelastic model for the passive myocardium: continuum basis and numerical treatment

This study deals with the viscoelastic constitutive modeling and the respective computational analysis of the human passive myocardium. We start by recapitulating the locally orthotropic inner structure of the human myocardial tissue and model the mechanical response through invariants and structure tensors associated with three orthonormal basis vectors. In accordance with recent experimental findings the ventricular myocardial tissue is assumed to be incompressible, thick-walled, orthotropic and viscoelastic. In particular, one spring element coupled with Maxwell elements in parallel endows the model with viscoelastic features such that four dashpots describe the viscous response due to matrix, fiber, sheet and fiber-sheet fragments. In order to alleviate the numerical obstacles, the strictly incompressible model is altered by decomposing the free-energy function into volumetric-isochoric elastic and isochoric-viscoelastic parts along with the multiplicative split of the deformation gradient which enables the three-field mixed finite element method. The crucial aspect of the viscoelastic formulation is linked to the rate equations of the viscous overstresses resulting from a 3-D analogy of a generalized 1-D Maxwell model. We provide algorithmic updates for second Piola–Kirchhoff stress and elasticity tensors. In the sequel, we address some numerical aspects of the constitutive model by applying it to elastic, cyclic and relaxation test data obtained from biaxial extension and triaxial shear tests whereby we assess the fitting capacity of the model. With the tissue parameters identified, we conduct (elastic and viscoelastic) finite element simulations for an ellipsoidal geometry retrieved from a human specimen.     

In vivo modeling of interstitial pressure in the brain under surgical load using finite elements

Current brain deformation models have predominantly reflected solid constitutive relationships generated from empirical ex vivo data and have largely overlooked interstitial hydrodynamic effects. In the context of a technique to update images intraoperatively for image-guided neuronavigation, we have developed and quantified the deformation characteristics of a three-dimensional porous media finite element model of brain deformation in vivo. Results have demonstrated at least 75-85 percent predictive capability, but have also indicated that interstitial hydrodynamics are important. In this paper we investigate interstitial pressure transient behavior in brain tissue when subjected to an acute surgical load consistent with neurosurgical events. Data are presented from three in vivo porcine experiments where subsurface tissue deformation and interhemispheric pressure gradients were measured under conditions of an applied mechanical deformation and then compared to calculations with our three-dimensional brain model. Results demonstrate that porous-media consolidation captures the hydraulic behavior of brain tissue subjected to comparable surgical loads and that the experimental protocol causes minimal trauma to porcine brain tissue. Working values for hydraulic conductivity of white and gray matter are also reported and an assessment of transient pressure gradient effects with respect to deformation is provided.     

Patient-specific non-linear finite element modelling for predicting soft organ deformation in real-time; Application to non-rigid neuroimage registration

Long computation times of non-linear (i.e. accounting for geometric and material non-linearity) biomechanical models have been regarded as one of the key factors preventing application of such models in predicting organ deformation for image-guided surgery. This contribution presents real-time patient-specific computation of the deformation field within the brain for six cases of brain shift induced by craniotomy (i.e. surgical opening of the skull) using specialised non-linear finite element procedures implemented on a graphics processing unit (GPU). In contrast to commercial finite element codes that rely on an updated Lagrangian formulation and implicit integration in time domain for steady state solutions, our procedures utilise the total Lagrangian formulation with explicit time stepping and dynamic relaxation. We used patient-specific finite element meshes consisting of hexahedral and non-locking tetrahedral elements, together with realistic material properties for the brain tissue and appropriate contact conditions at the boundaries. The loading was defined by prescribing deformations on the brain surface under the craniotomy. Application of the computed deformation fields to register (i.e. align) the preoperative and intraoperative images indicated that the models very accurately predict the intraoperative deformations within the brain. For each case, computing the brain deformation field took less than 4 s using an NVIDIA Tesla C870 GPU, which is two orders of magnitude reduction in computation time in comparison to our previous study in which the brain deformation was predicted using a commercial finite element solver executed on a personal computer.     

Material properties of the human calcaneal fat pad in compression: experiment and theory

The structural behaviour of the human heel pad has been studied extensively due to its ability to absorb shock, protect against excessive local stress, and reduce plantar pressures. However, the material properties of the tissue have not been adequately measured. These must be known in order to perform a finite element analysis of the effect of factors such as foot geometry and shoe/surface construction on heel pad function. Therefore, the purposes of this study were to (a) measure the viscoelastic behaviour of the fat pad in compression, and (b) to determine an appropriate constitutive equation to model the tissue. A series of unconfined compression tests were performed on 8 mm diameter cylinders of fat pad tissue, consisting of quasi-static, 175, 350 mm/s and stress-relaxation tests to 50% deformation. The tissue exhibited nonlinear, viscoelastic behaviour. No significant difference was found in the quasi-static behaviour between samples from different locations and orientations in the heel. The stress-relaxation tests were used to determine the time constant (τ₁=0.5 s), the 175 mm/s test to determine the relaxation coefficient (g₁=28), and the 350 mm/s compression test to determine the material constants (C₁₀₀=C₀₁₀=0.01, C₂₀₀=C₀₂₀=0.1 Pa) of a single-phase, hyperelastic, linear viscoelastic strain energy function (r²=0.98).     

Nonlinear viscoelasticty plays an essential role in the functional behavior of spinal ligaments

Despite the significant role ligament viscoelasticity plays in functional spinal biomechanics, relatively few studies have been performed to develop constitutive models that explicitly characterize this complex behavior. Unfortunately, the application and interpretation of these previous models are limited due to the use of simplified (quasi-linear) viscoelastic formulations or characterization techniques that have been shown to affect the predictive accuracy of the fitted coefficients. In order to surmount these previous limitations, the current study presents the application of a novel fitting technique (applied to stress relaxation experiments) and nonlinear viscoelastic constitutive formulation to human cervical spine anterior longitudinal ligament (ALL), posterior longitudinal ligament (PLL) and ligamentum flavum (LF). The fitted coefficients were validated by quantifying the ability of the constitutive equation to predict an independent cyclic data set across multiple physiologic strain amplitudes and frequencies. The resulting validated constitutive formulation indicated that the strain-dependent viscoelastic behavior of the longitudinal ligaments (ALL and PLL) was dominated by both the short-term (t=0.1s) and the steady-state (as t→∞) behavior. Conversely, the LF exhibited consistent relaxation behavior across the investigated temporal spectrum. From these data, it can be hypothesized that the unique strain-dependent temporal behavior of these spinal ligaments may be a functional adaptation that minimizes muscular expenditure during quasi-static postures while maximizing structural stability of the spine during transient loading events.     

Local Viscoelastic Properties of Live Cells Investigated Using Dynamic and Quasi-Static Atomic Force Microscopy Methods

The measurement of viscoelasticity of cells in physiological environments with high spatio-temporal resolution is a key goal in cell mechanobiology. Traditionally only the elastic properties have been measured from quasi-static force-distance curves using the atomic force microscope (AFM). Recently, dynamic AFM-based methods have been proposed to map the local in vitro viscoelastic properties of living cells with nanoscale resolution. However, the differences in viscoelastic properties estimated from such dynamic and traditional quasi-static techniques are poorly understood. In this work we quantitatively reconstruct the local force and dissipation gradients (viscoelasticity) on live fibroblast cells in buffer solutions using Lorentz force excited cantilevers and present a careful comparison between mechanical properties (local stiffness and damping) extracted using dynamic and quasi-static force spectroscopy methods. The results highlight the dependence of measured viscoelastic properties on both the frequency at which the chosen technique operates as well as the interactions with subcellular components beyond certain indentation depth, both of which are responsible for differences between the viscoelasticity property maps acquired using the dynamic AFM method against the quasi-static measurements.     

Local Viscoelastic Properties of Live Cells Investigated Using Dynamic and Quasi-Static Atomic Force Microscopy Methods

The measurement of viscoelasticity of cells in physiological environments with high spatio-temporal resolution is a key goal in cell mechanobiology. Traditionally only the elastic properties have been measured from quasi-static force-distance curves using the atomic force microscope (AFM). Recently, dynamic AFM-based methods have been proposed to map the local in vitro viscoelastic properties of living cells with nanoscale resolution. However, the differences in viscoelastic properties estimated from such dynamic and traditional quasi-static techniques are poorly understood. In this work we quantitatively reconstruct the local force and dissipation gradients (viscoelasticity) on live fibroblast cells in buffer solutions using Lorentz force excited cantilevers and present a careful comparison between mechanical properties (local stiffness and damping) extracted using dynamic and quasi-static force spectroscopy methods. The results highlight the dependence of measured viscoelastic properties on both the frequency at which the chosen technique operates as well as the interactions with subcellular components beyond certain indentation depth, both of which are responsible for differences between the viscoelasticity property maps acquired using the dynamic AFM method against the quasi-static measurements.     

Patterning of brain organoids derived from human pluripotent stem cells

The emerging technology of brain organoids deriving from human pluripotent stem cells provides unprecedented opportunities to study human brain development and associated disorders. Various brain organoid protocols have been developed that can recapitulate some key features of cell type diversity, cytoarchitectural organization, developmental processes, functions, and pathologies of the developing human brain. In this review, we focus on patterning of human stem cell-derived brain organoids. We start with an overview of general procedures to generate brain organoids. We then highlight some recently developed brain organoid protocols and chemical cues involved in modeling development of specific human brain regions, subregions, and multiple regions together. We also discuss limitations and potential future improvements of human brain organoid technology.     

A novel two-layer, coupled finite element approach for modeling the nonlinear elastic and viscoelastic behavior of human erythrocytes

A novel finite element approach is presented to simulate the mechanical behavior of human red blood cells (RBC, erythrocytes). As the RBC membrane comprises a phospholipid bilayer with an intervening protein network, we propose to model the membrane with two distinct layers. The fairly complex characteristics of the very thin lipid bilayer are represented by special incompressible solid shell elements and an anisotropic viscoelastic constitutive model. Properties of the protein network are modeled with an isotropic hyperelastic third-order material. The elastic behavior of the model is validated with existing optical tweezers studies with quasi-static deformations. Employing material parameters consistent with literature, simulation results are in excellent agreement with experimental data. Available models in literature neglect either the surface area conservation of the RBC membrane or realistic loading conditions of the optical tweezers experiments. The importance of these modeling assumptions, that are both included in this study, are discussed and their influence quantified. For the simulation of the dynamic motion of RBC, the model is extended to incorporate the cytoplasm. This is realized with a monolithic fully coupled fluid-structure interaction simulation, where the fluid is described by the incompressible Navier–Stokes equations in an arbitrary Lagrangian Eulerian framework. It is shown that both membrane viscosity and cytoplasm viscosity have significant influence on simulation results. Characteristic recovery times and energy dissipation for varying strain rates in dynamic laser trap experiments are calculated for the first time and are found to be comparable with experimental data.