Search

Finite element models built from quantitative computed tomography images rely on element-wise mapping of material properties starting from Hounsfield Units (HU), which can be converted into mineral densities upon calibration. While calibration is preferably carried out by scanning a phantom with known-density components, conducting phantom-based calibration may not always be possible. In such cases, a phantomless procedure, where the scanned subject's tissues are used as a phantom, is an interesting alternative. The aim of this study was to compare a phantom-based and a phantomless calibration method on 41 postmenopausal women. The proposed phantomless calibration utilized air, adipose, and muscle tissues, with reference equivalent mineral density values of -797, -95, and 38 mg/cm3, extracted from a previously performed phantom-based calibration. A 9-slice volume of interest (VOI) centred between the femoral head and knee rotation centres was chosen. Reference HU values for air, adipose, and muscle tissues were extracted by identifying HU distribution peaks within the VOI, and patient-specific calibration was performed using linear regression. Comparison of FE models calibrated with the two methods showed average relative differences of 1.99% for Young's modulus1.30% for tensile and 1.34% for compressive principal strains. Excellent correlations (R2 > 0.99) were identified for superficial maximum tensile and minimum compressive strains. Maximum normalised root mean square relative error (RMSRE) values settled at 4.02% for Young's modulus, 2.99% for tensile, and 3.22% for compressive principal strains, respectively. The good agreement found between the two methods supports the adoption of the proposed methodology when phantomless calibration is needed. [ABSTRACT FROM AUTHOR]

Quantitative computed tomography (QCT)-based in silico models have demonstrated improved accuracy in predicting hip fractures with respect to the current gold standard, the areal bone mineral density. These models require that the femur bone is segmented as a first step. This task can be challenging, and in fact, it is often almost fully manual, which is time-consuming, operator-dependent, and hard to reproduce. This work proposes a semi-automated procedure for femur bone segmentation from CT images. The proposed procedure is based on the bone and joint enhancement filter and graph-cut algorithms. The semi-automated procedure performances were assessed on 10 subjects through comparison with the standard manual segmentation. Metrics based on the femur geometries and the risk of fracture assessed in silico resulting from the two segmentation procedures were considered. The average Hausdorff distance (0.03 ± 0.01 mm) and the difference union ratio (0.06 ± 0.02) metrics computed between the manual and semi-automated segmentations were significantly higher than those computed within the manual segmentations (0.01 ± 0.01 mm and 0.03 ± 0.02). Besides, a blind qualitative evaluation revealed that the semi-automated procedure was significantly superior (p < 0.001) to the manual one in terms of fidelity to the CT. As for the hip fracture risk assessed in silico starting from both segmentations, no significant difference emerged between the two (R2 = 0.99). The proposed semi-automated segmentation procedure overcomes the manual one, shortening the segmentation time and providing a better segmentation. The method could be employed within CT-based in silico methodologies and to segment large volumes of images to train and test fully automated and supervised segmentation methods. [ABSTRACT FROM AUTHOR]

Little is known about dynamic changes of femoral anatomy after total hip arthroplasty (THA), in particular about sagittal femoral bowing (SFB). A 3D CT study was designed to evaluate the chronological changes of SFB after cementless femoral stem implantation for primary THA. Ten patients who underwent unilateral primary THA with a cementless femoral stem, with 2 consecutive CT scans (extending from the fourth lumbar vertebra to the tibial plateaus), performed before THA and at least 3 years after THA, were enrolled. The 3D models of femurs were created using image segmentation software. Using the two CT scans, SFB values of the proximal and middle thirds were calculated on the replaced and untreated sides by two different observers. Eight anatomical stems and two conical stems were involved. The post-operative CT was performed at an average follow-up of 6.5 years after THA (range: 3–12.5). The measurements performed by the two observers did not differ in the proximal and middle regions. A significant difference between the pre-operative and post-operative SFB compared to the untreated side was found in the proximal femur segment (p = 0.004). Use of a cementless stem in THA induced chronological changes in SFB of the proximal femur, after a minimum timespan of 3 years. [ABSTRACT FROM AUTHOR]

Pathologies such as cancer metastasis and osteoporosis strongly affect the mechanical properties of the vertebral bone and increase the risk of fragility fractures. The prediction of the fracture risk with a patient-specific model, directly generated from the diagnostic images of the patient, could help the clinician in the choice of the correct therapy to follow. But before such models can be used to support any clinical decision, their credibility must be demonstrated through verification, validation, and uncertainty quantification. In this study we describe a procedure for the generation of such patient-specific finite element models and present a first validation of the kinematics of the spine segment. Quantitative computed tomography images of a cadaveric lumbar spine segment presenting vertebral metastatic lesions were used to generate the model. The applied boundary conditions replicated a specific experimental test where the spine segment was loaded in compression-flexion. Model predictions in terms of vertebral surface displacements were compared against the full-field experimental displacements measured with Digital Image Correlation. A good agreement was obtained from the local comparison between experimental data and simulation results (R2 > 0.9 and RMSE% <8%). In conclusion, this work demonstrates the possibility to apply the developed modelling pipeline to predict the displacement field of human spine segment under physiological loading conditions, which is a first fundamental step in the credibility assessment of these clinical decision-support technology. [ABSTRACT FROM AUTHOR]

Recently, coupled musculoskeletal-finite element modelling approaches have emerged as a way to investigate femoral neck loading during various daily activities. Combining personalised gait data with finite element models will not only allow us to study changes in motion/movement, but also their effects on critical internal structures, such as the femur. However, previous studies have been hampered by the small sample size and the lack of fully personalised data in order to construct the coupled model. Therefore, the aim of this study was to build a pipeline for a fully personalised multiscale (body-organ level) model to investigate the strain levels at the femoral neck during a normal gait cycle. Five postmenopausal women were included in this study. The CT and MRI scans of the lower limb, and gait data were collected for all participants. Muscle forces derived from the body level musculoskeletal models were used as boundary constraints on the finite element femur models. Principal strains were estimated at the femoral neck region during a full gait cycle. Considerable variation was found in the predicted peak strain among individuals with mean peak first principal strain of 0.24% ± 0.11% and mean third principal strain of -0.29% ± 0.24%. For four individuals, two overall peaks of the maximum strains were found to occur when both feet were in contact with the floor, while one individual had one peak at the toe-off phase. Both the joint contact forces and the muscular forces were found to substantially influence the loading at the femoral neck. A higher correlation was found between the predicted peak strains and the gluteus medius (R2 ranged between 0.95 and 0.99) than the hip joint contact forces (R2 ranged between 0.63 and 0.96). Therefore, the current findings suggest that personal variations are substantial, and hence it is important to consider multiple subjects before deriving general conclusions for a target population. [ABSTRACT FROM AUTHOR]

Fractures of bone account 25% of all paediatric injuries (Cooper et al. in J Bone Miner Res 19:1976-1981, 2004. 10.1359/JBMR.040902). These can be broadly categorised into accidental or inflicted injuries. The current clinical approach to distinguish between these two is based on the clinician’s judgment, which can be subjective. Furthermore, there is a lack of studies on paediatric bone to provide evidence-based information on bone strength, mainly due to the difficulties of obtaining paediatric bone samples. There is a need to investigate the behaviour of children’s bones under external loading. Such data will critically enhance our understanding of injury tolerance of paediatric bones under various loading conditions, related to injuries, such as bending and torsional loads. The aim of this study is therefore to investigate the response of paediatric femora under two types of loading conditions, bending and torsion, using a CT-based finite element approach, and to determine a relationship between bone strength and age/body mass of the child. Thirty post-mortem CT scans of children aged between 0 and 3 years old were used in this study. Two different boundary conditions were defined to represent four-point bending and pure torsional loads. The principal strain criterion was used to estimate the failure moment for both loading conditions. The results showed that failure moment of the bone increases with the age and mass of the child. The predicted failure moment for bending, external and internal torsions were 0.8-27.9, 1.0-31.4 and 1.0-30.7 Nm, respectively. To the authors’ knowledge, this is the first report on infant bone strength in relation to age/mass using models developed from modern medical images. This technology may in future help advance the design of child, car restrain system, and more accurate computer models of children. [ABSTRACT FROM AUTHOR]

The estimation of local and structural mechanical properties of bones with micro Finite Element (microFE) models based on Micro Computed Tomography images depends on the quality bone geometry is captured, reconstructed and modelled. The aim of this study was to validate microFE models predictions of local displacements for vertebral bodies and to evaluate the effect of the elastic tissue modulus on model’s predictions of axial forces. Four porcine thoracic vertebrae were axially compressed in situ, in a step-wise fashion and scanned at approximately 39μm resolution in preloaded and loaded conditions. A global digital volume correlation (DVC) approach was used to compute the full-field displacements. Homogeneous, isotropic and linear elastic microFE models were generated with boundary conditions assigned from the interpolated displacement field measured from the DVC. Measured and predicted local displacements were compared for the cortical and trabecular compartments in the middle of the specimens. Models were run with two different tissue moduli defined from microindentation data (12.0GPa) and a back-calculation procedure (4.6GPa). The predicted sum of axial reaction forces was compared to the experimental values for each specimen. MicroFE models predicted more than 87% of the variation in the displacement measurements (R2 = 0.87–0.99). However, model predictions of axial forces were largely overestimated (80–369%) for a tissue modulus of 12.0GPa, whereas differences in the range 10–80% were found for a back-calculated tissue modulus. The specimen with the lowest density showed a large number of elements strained beyond yield and the highest predictive errors. This study shows that the simplest microFE models can accurately predict quantitatively the local displacements and qualitatively the strain distribution within the vertebral body, independently from the considered bone types. [ABSTRACT FROM AUTHOR]

Non-destructive 3D micro-computed tomography (microCT) based finite element (microFE) models are used to estimate bone mechanical properties at tissue level. However, their validation remains challenging. Recent improvements in the quantification of displacements in bone tissue biopsies subjected to staged compression, using refined Digital Volume Correlation (DVC) techniques, now provide a full field displacement information accurate enough to be used for microFE validation. In this study, three specimens (two humans and one bovine) were tested with two different experimental set-ups, and the resulting data processed with the same DVC algorithm. The resulting displacement vector field was compared to that predicted by microFE models solved with three different boundary conditions (BC): nominal force resultant, nominal displacement resultant, distributed displacement. The first two conditions were obtained directly from the measurements provided by the experimental jigs, whereas in the third case the displacement field measured by the DVC in the top and bottom layer of the specimen was applied. Results show excellent relationship between the numerical predictions (x) and the experiments (y) when using BC derived from the DVC measurements (U X : y=1.07x−0.002, RMSE: 0.001 mm; U Y : y=1.03x−0.001, RMSE: 0.001 mm; U Z : y=x+0.0002, RMSE: 0.001 mm for bovine specimen), whereas only poor correlation was found using BCs according to experiment set-ups. In conclusion, microFE models were found to predict accurately the vectorial displacement field using interpolated displacement boundary condition from DVC measurement. [ABSTRACT FROM AUTHOR]

Subject-specific musculoskeletal models have become key tools in the clinical decision-making process. However, the sensitivity of the calculated solution to the unavoidable errors committed while deriving the model parameters from the available information is not fully understood. The aim of this study was to calculate the sensitivity of all the kinematics and kinetics variables to the inter-examiner uncertainty in the identification of the lower limb joint models. The study was based on the computer tomography of the entire lower-limb from a single donor and the motion capture from a body-matched volunteer. The hip, the knee and the ankle joint models were defined following the International Society of Biomechanics recommendations. Using a software interface, five expert anatomists identified on the donor's images the necessary bony locations five times with a three-day time interval. A detailed subject-specific musculoskeletal model was taken from an earlier study, and re-formulated to define the joint axes by inputting the necessary bony locations. Gait simulations were run using OpenSim within a Monte Carlo stochastic scheme, where the locations of the bony landmarks were varied randomly according to the estimated distributions. Trends for the joint angles, moments, and the muscle and joint forces did not substantially change after parameter perturbations. The highest variations were as follows: (a) 11° calculated for the hip rotation angle, (b) 1% BW × Hcalculated for the knee moment and (c) 0.33 BW calculated for the ankle plantarflexor muscles and the ankle joint forces. In conclusion, the identification of the joint axes from clinical images is a robust procedure for human movement modelling and simulation. [ABSTRACT FROM AUTHOR]

The aim of this paper is to analyze how the uncertainties in modelling the geometry and the material properties of a human bone affect the predictions of a finite-element model derived from computed tomography (CT) data. A sensitivity analysis, based on a Monte Carlo method, was performed using three femur models generated from in vivo CT datasets, each subjected to two different loading conditions. The geometry, the density and the mechanical properties of the bone tissue were considered as random input variables. Finite-element results typically used in biomechanics research were considered as statistical output variables, and their sensitivity to the inputs variability assessed. The results showed that it is not possible to define a priori the influence of the errors related to the geometry definition process and to the material assignment process on the finite-element analysis results. The errors in the geometric representation of the bone are always the dominant variables for the stresses, as was expected. However, for all the variables, the results seemed to be dependent on the loading condition and to vary from subject to subject. The most interesting result is, however, that using the proposed method to build a finite-element model of a femur from a CT dataset of the quality typically achievable in the clinical practice, the coefficients of variation of the output variables never exceed the 9%. The presented method is hence robust enough to be used for investigating the mechanical behavior of bones with subject-specific finite-element models derived from CT data taken in vivo. [ABSTRACT FROM AUTHOR]

Most of the finite element models of bones used in orthopaedic biomechanics research are based on generic anatomies. However, in many cases it would be useful to generate from CT data a separate finite element model for each subject of a study group. In a recent study a hexahedral mesh generator based on a grid projection algorithm was found very effective in terms of accuracy and automation. However, so far the use of this method has been documented only on data collected in vitro and only for long bones. The present study was aimed at verifying if this method represents a procedure for the generation of finite element models of human bones from data collected in vivo, robust, accurate, automatic and general enough to be used in clinical studies. Robustness, automation and numerical accuracy of the proposed method were assessed on five femoral CT data sets of patients affected by various pathologies. The generality of the method was verified by processing a femur, an ileum, a phalanx, a proximal femur reconstruction, and the micro-CT of a small sample of spongy bone. The method was found robust enough to cope with the variability of the five femurs, producing meshes with a numerical accuracy and a computational weight comparable to those found in vitro. Even when the method was used to process the other bones the levels of mesh conditioning remained within acceptable limits. Thus, it may be concluded that the method presents a generality sufficient to cope with almost any orthopaedic application. [Copyright &y& Elsevier]

The assignment of bone tissue material properties is a fundamental step in the generation of subject-specific finite element models from computed tomography data. Aim of the present work is to investigate the influence of the material mapping algorithm on the results predicted by the finite element analysis. Two models, a coarse and a refined one, of a human ileum, femur and tibia, were generated from CT data and used for the tests. In addition a convergence analysis was carried out for the femur model, using six refinement levels, to verify whether the inclusion of the material properties would significantly alter the convergence behaviour of the mesh. The results showed that the choice of the mapping algorithm influences the material distribution. However, this did not always propagate into the finite element results. The difference between the maximum Von Mises stress remained always lower than 10%, apart one case when it reached the 13%. However, the global behaviour of the meshes showed more marked differences between the two algorithms: in the finer meshes of the two long bones 20−30% of the bone volume showed differences in the predicted Von Mises stresses greater than 10%. The convergence behaviour of the model was not worsened by the introduction of inhomogeneous material properties. The software was made available in the public domain. [Copyright &y& Elsevier]

Understanding bone mechanics at different hierarchical levels is fundamental to improve preclinical and clinical assessments of bone strength. Digital Volume Correlation (DVC) is the only experimental measurement technique used for measuring local displacements and calculating local strains within bones. To date, its combination with laboratory source micro-computed tomography (LS-microCT) data typically leads to high uncertainties, which limit its application. Here, the benefits of synchrotron radiation micro-computed tomography (SR-microCT) for DVC are reported. Specimens of cortical and trabecular bovine bone and murine tibiae, were each scanned under zero-strain conditions with an effective voxel size of 1.6 μm. In order to consider the effect of the voxel size, analyses were also performed on downsampled images with voxel size of 8 μm. To evaluate displacement and strain uncertainties, each pair of tomograms was correlated using a global DVC algorithm (ShIRT-FE). Displacement random errors for original SR-microCT ranged from 0.024 to 0.226 μm, depending on DVC nodal spacing. Standard deviation of strain errors was below 200 microstrain (ca. 1/10 of the strain associated with physiological loads) for correlations performed with a measurement spatial resolution better than 40 μm for cortical bovine bone (240 μm for downsampled images), 80 μm for trabecular bovine bone (320 μm for downsampled images) and murine tibiae (120 μm for downsampled images). This study shows that the uncertainties of SR-microCT-based DVC, estimated from repeated scans, are lower than those obtained from LS-microCT-based DVC on similar specimens and low enough to measure accurately the local deformation at the tissue level. [ABSTRACT FROM AUTHOR]

In vivo micro-computed tomography (µCT) scanning of small rodents is a powerful method for longitudinal monitoring of bone adaptation. However, the life-time bone growth in small rodents makes it a challenge to quantify local bone adaptation. Therefore, the aim of this study was to develop a protocol, which can take into account large bone growth, to quantify local bone adaptations over space and time. The entire right tibiae of eight 14-week-old C57BL/6J female mice were consecutively scanned four times in an in vivo µCT scanner using a nominal isotropic image voxel size of 10.4 µm. The repeated scan image datasets were aligned to the corresponding baseline (first) scan image dataset using rigid registration. 80% of tibia length (starting from the endpoint of the proximal growth plate) was selected as the volume of interest and partitioned into 40 regions along the tibial long axis (10 divisions) and in the cross-section (4 sectors). The bone mineral content (BMC) was used to quantify bone adaptation and was calculated in each region. All local BMCs have precision errors (PE %CV ) of less than 3.5% (24 out of 40 regions have PE %CV of less than 2%), least significant changes (LSCs) of less than 3.8%, and 38 out of 40 regions have intraclass correlation coefficients (ICCs) of over 0.8. The proposed protocol allows to quantify local bone adaptations over an entire tibia in longitudinal studies, with a high reproducibility, an essential requirement to reduce the number of animals to achieve the necessary statistical power. [ABSTRACT FROM AUTHOR]

In vivo micro-computed tomography (µCT) scanning is an important tool for longitudinal monitoring of the bone adaptation process in animal models. However, the errors associated with the usage of in vivo µCT measurements for the evaluation of bone adaptations remain unclear. The aim of this study was to evaluate the measurement errors using the bone surface distance approach. The right tibiae of eight 14-week-old C57BL/6 J female mice were consecutively scanned four times in an in vivo µCT scanner using a nominal isotropic image voxel size (10.4 µm) and the tibiae were repositioned between each scan. The repeated scan image datasets were aligned to the corresponding baseline (first) scan image dataset using rigid registration and a region of interest was selected in the proximal tibia metaphysis for analysis. The bone surface distances between the repeated and the baseline scan datasets were evaluated. It was found that the average (±standard deviation) median and 95th percentile bone surface distances were 3.10 ± 0.76 µm and 9.58 ± 1.70 µm, respectively. This study indicated that there were inevitable errors associated with the in vivo µCT measurements of bone microarchitecture and these errors should be taken into account for a better interpretation of bone adaptations measured with in vivo µCT. [ABSTRACT FROM AUTHOR]

Different digital volume correlation (DVC) approaches are currently available or under development for bone tissue micromechanics. The aim of this study was to compare accuracy and precision errors of three DVC approaches for a particular three-dimensional (3D) zero-strain condition. Trabecular and cortical bone specimens were repeatedly scanned with a micro-computed tomography (CT). The errors affecting computed displacements and strains were extracted for a known virtual translation, as well as for repeated scans. Three DVC strategies were tested: two local approaches, based on fast-Fourier-transform (DaVis-FFT) or direct-correlation (DaVis-DC), and a global approach based on elastic registration and a finite element (FE) solver (ShIRT-FE). Different computation subvolume sizes were tested. Much larger errors were found for the repeated scans than for the virtual translation test. For each algorithm, errors decreased asymptotically for larger subvolume sizes in the range explored. Considering this particular set of images, ShIRT-FE showed an overall better accuracy and precision (a few hundreds microstrain for a subvolume of 50 voxels). When the largest subvolume (50-52 voxels) was applied to cortical bone, the accuracy error obtained for repeated scans with ShIRT-FE was approximately half of that for the best local approach (DaVis-DC). The difference was lower (250 microstrain) in the case of trabecular bone. In terms of precision, the errors shown by DaVis-DC were closer to the ones computed by ShIRT-FE (differences of 131 microstrain and 157 microstrain for cortical and trabecular bone, respectively). The multipass computation available for DaVis software improved the accuracy and precision only for the DaVis-FFT in the virtual translation, particularly for trabecular bone. The better accuracy and precision of ShIRT-FE, followed by DaVis-DC, were obtained with a higher computational cost when compared to DaVis-FFT. The results underline the importance of performing a quantitative comparison of DVC methods on the same set of samples by using also repeated scans, other than virtual translation tests only. ShIRT-FE provides the most accurate and precise results for this set of images. However, both DaVis approaches show reasonable results for large nodal spacing, particularly for trabecular bone. Finally, this study highlights the importance of using sufficiently large subvolumes, in order to achieve better accuracy and precision. [ABSTRACT FROM AUTHOR]

Computed tomography (CT)-based finite element (FE) reconstructions describe shape and density distribution of bones. Both shape and density distribution, however, can vary a lot between individuals. Shape/density indexation (usually achieved by principal component analysis—PCA) can be used to synthesize realistic models, thus overcoming the shortage of CT-based models, and helping e.g. to study fracture determinants, or steer prostheses design. The aim of this study was to describe a PCA-based statistical modelling algorithm, and test it on a large CT-based population of femora, to see if it can accurately describe and reproduce bone shape, density distribution, and biomechanics. To this aim, 115 CT-datasets showing normal femoral anatomy were collected and characterized. Isotopological FE meshes were built. Shape and density indexation procedures were performed on the mesh database. The completeness of the database was evaluated through a convergence study. The accuracy in reconstructing bones not belonging to the indexation database was evaluated through (i) leave-one-out tests (ii) comparison of calculated vs. in-vitro measured strains. Fifty indexation modes for shape and 40 for density were necessary to achieve reconstruction errors below pixel size for shape, and below 10% for density. Similar errors for density, and slightly higher errors for shape were obtained when reconstructing bones not belonging to the database. The in-vitro strain prediction accuracy of the reconstructed FE models was comparable to state-of-the-art studies. In summary, the results indicate that the proposed statistical modelling tools are able to accurately describe a population of femora through finite element models. [ABSTRACT FROM AUTHOR]

Abstract: Background: Biological massive skeletal reconstructions in tumours adopt a long rehabilitation protocol aimed at minimising the fracture risk. To improve rehabilitation and surgical procedures it is important to fully understand the biomechanics of the reconstructed limb. The aim of the present study was to develop a subject-specific musculoskeletal model of a patient with a massive biological skeletal reconstruction, to investigate the loads acting on the femur during gait, once the rehabilitation protocol was completed. Methods: A personalised musculoskeletal model of the patient''s lower limbs was built from a CT exam and registered with the kinematics recorded in a gait analysis session. Predicted activations for major muscles were compared to EMG signals to assess the model''s predictive accuracy. Findings: Gait kinematics showed only minor discrepancies between the two legs and was compatible with normality data. External moments showed slightly higher differences and were almost always lower on the operated leg exhibiting a lower variability. In the beginning of the stance phase, the joint moments were, conversely, higher on the operated side and showed a higher variability. This pattern was reflected and amplified on the femoral forces where the differences became important: on the hip, a maximum difference of 1.6BW was predicted. The variability of the forces seemed, generally, lower on the operated leg than on the contralateral one. Interpretation: Small asymmetries in kinematic patterns might be associated, in massive skeletal reconstruction, to significant difference in the skeletal loads (up to 1.6BW for the hip joint reaction) during gait. [Copyright &y& Elsevier]

Abstract: Background: The Finite Element Method is at present the method of choice for strain prediction in bones from Computed Tomography data. However, accurate methods rely on the correct topological representation of the bone surface, which requires a massive operator effort, thus restricting their applicability to clinical practice. Meshless methods, which do not rely on a pre-defined topological discretisation of the domain, might greatly improve the numerical process automation, but currently their application to biomechanics is negligible. Methods: A meshless implementation of an innovative numerical approach based on a direct discrete formulation of physical laws, the Cell Method, was developed to predict strains in a cadaver femur from Computed Tomography data. The model accuracy was estimated by comparing the predicted strains with those experimentally measured on the same specimen in a previous study. As a reference, the results were compared to those obtained with a state-of-the-art finite element model. Findings: The Cell Method meshless model predicted strains highly correlated with the experimental measurements (R 2 =0.85) with a good global accuracy (RMSE=15.6%). The model performed slightly worse than the finite element one, but this was probably due to the need to sub-sample the original data, and the lower order of the interpolation used (linear vs parabolic). Interpretation: Although there is surely room for improvement, the accuracy already obtained with this meshless implementation of the Cell Method makes it a good candidate for some clinical applications, especially considering the full automation of the method, which does not require any data pre-processing. [Copyright &y& Elsevier]

Abstract: An experimental–numerical study was performed to investigate the relationships between computed tomography (CT)-density and ash density, and between ash density and apparent density for bone tissue, to evaluate their influence on the accuracy of subject-specific FE models of human bones. Sixty cylindrical bone specimens were examined. CT-densities were computed from CT images while apparent and ash densities were measured experimentally. The CT/ash-density and ash/apparent-density relationships were calculated. Finite element models of eight human femurs were generated considering these relationships to assess their effect on strain prediction accuracy. CT and ash density were linearly correlated (R 2=0.997) over the whole density range but not equivalent (intercep t <0, slope >1). A constant ash/apparent-density ratio (0.598±0.004) was found for cortical bone. A lower ratio, with a larger dispersion, was found for trabecular bone (0.459±0.100), but it became less dispersed, and equal to that of cortical tissue, when testing smaller trabecular specimens (0.598±0.036). This suggests that an experimental error occurred in apparent-density measurements for large trabecular specimens and a constant ratio can be assumed valid for the whole density range. Introducing the obtained relationships in the FE modelling procedure improved strain prediction accuracy (R 2=0.95, RMSE=7%). The results suggest that: (i) a correction of the densitometric calibration should be used when evaluating bone ash-density from clinical CT scans, to avoid ash-density underestimation and overestimation for low- and high-density bone tissue, respectively; (ii) the ash/apparent-density ratio can be assumed constant in human femurs and (iii) the correction improves significantly the model accuracy and should be considered in subject-specific bone modelling. [Copyright &y& Elsevier]

Abstract: The aim of the present study is to compare the results from subject-specific finite element analysis (FEA) of a human femur to experimental measurements, using two different methods for assigning material properties to the FE models. A modified material mapping strategy allowing for spatial variation of material properties within the elements and Young''s modulus surface corrections is presented and compared to a more conventional strategy, whereby constant material properties are assigned to each element. The accuracy of the superficial stress–strain predictions was evaluated against experimental results from 13 strain gauges and five different load cases. Both methods predicted stresses with acceptable accuracy (R 2 =0.92, root mean square error, RMSE<10%), with the conventional method performing slightly better. The modified method performed better in strain prediction (R 2 =0.85, RMSE=23% versus R 2 =0.79, RMSE=31%). [Copyright &y& Elsevier]

Abstract: Aim of the present study was to evaluate the influence on the global model''s accuracy of the strategy adopted to define the average element Young''s modulus in subject-specific finite element models of bones from computed tomography data. The classic strategy of calculating the Young''s modulus from an average element density and the one that averages the Young''s moduli directly derived from each voxel Hounsfield Unit were considered. These strategies were applied to the finite element model of a real human femur. The accuracy of the superficial stress and strain predictions was evaluated against experimentally measured values in 13 strain-gauge locations for five different loading conditions. The results obtained for the two material distributions were statistically different. Both models predicted very accurately the superficial stresses, with regression coefficients higher than 0.9 and slopes not significantly different from unity. The second strategy definitely improved the strains prediction accuracy: the regression coefficient raised from 0.69 to 0.79; the average and peak errors decreased from 45.1% to 31.3% and from 228% to 134% of the maximum measured strain, respectively. The stress fields predicted inside the bone were also significantly different. A new software implementing the second strategy was made available in the public domain. [Copyright &y& Elsevier]

Abstract: The prediction of the stress-state and fracture risk induced in bones by various loading conditions in individual patients using subject-specific finite element models still represents a challenge in orthopaedic biomechanics. The accuracy of the strain predictions reported in the literature is variable and generally not satisfactory. The aim of the present study was to evaluate if a proper choice of the density–elasticity relationship can lead to accurate strain predictions in the frame of an automatic subject-specific model generation strategy. To this aim, a combined numerical–experimental study was performed comparing finite element predicted strains with strain-gauges measurements obtained on eight cadaver proximal femurs, each instrumented with 15 rosettes mostly concentrated in the bone metaphyses, tested non-destructively in vitro under six different loading scenarios. Three different density–elasticity power relationships were selected from the literature and implemented in the finite element models derived from computed tomography data. The results of the present study confirm the great influence of the density–elasticity relationship used on the accuracy of numerical predictions. One of the tested constitutive laws provided a very good agreement (R 2=0.91, RMSE lower than 10% of the maximum measured value) between numerical calculations and experimental measurements. The presented results show, in addition, that the adoption of a single density–elasticity relationship over the whole bone density range is adequate to obtain an accuracy that is already suitable for many applications. [Copyright &y& Elsevier]

Abstract: The determination of the mechanical stresses induced in human bones is of great importance in both research and clinical practice. Since the stresses in bones cannot be measured non-invasively in vivo, the only way to estimate them is through subject-specific finite element modelling. Several methods exist for the automatic generation of these models from CT data, but before bringing them in the clinical practice it is necessary to assess their accuracy in the predictions of the bone stresses. Particular attention should be paid to those regions, like the epiphyseal and metaphyseal parts of long bones, where the automatic methods are typically less accurate. Aim of the present study was to implement a general procedure to automatically generate subject-specific finite element models of bones from CT data and estimate the accuracy of this general procedure by applying it to one real femur. This femur was tested in vitro under five different loading scenarios and the results of these tests were used to verify how the adoption of a simplified two-material homogeneous model would change the accuracy with respect to the density-based inhomogeneous one, with special attention paid to the epiphyseal and metaphyseal proximal regions of the bone. The results showed that the density-based inhomogeneous model predicts with a very good accuracy the measured stresses (, RMSE=8.6%, peak error=27%), while the two-material model was less accurate (, RMSE=9.6%, peak error=35%). The results showed that it is possible to automatically generate accurate finite element models of bones from CT data and that the strategy of material properties mapping has a significant influence on its accuracy. [Copyright &y& Elsevier]

X-ray computer tomography (CT) provides an accurate source of information in orthopaedics. Many computer aided orthopaedic surgery systems are based on CT images; thus, obtaining high resolution images is important. However, this may result in an excessive radiation dose for the patient. This study is aimed at developing a special CT scanning protocol for the hip region that can be adopted in clinical practise for 3-D pre-operative planning of total hip replacement surgery. Optimisation of CT acquisition parameters is investigated for both axial and spiral CT and the two resulting protocols are compared in terms of effective radiation dose to the patient. Results show that spiral CT with D=3 and P=1.5 in regions with higher morphological and density gradients and D=5 and P=1.5 in regions where the morphology is more regular degrades the image quality slightly but allows acquisition of a higher number of images at comparable costs, increasing the longitudinal resolution of the acquired data set. The effective dose is comparable to that of a standard pelvic CT exam. Adjusting the axial CT scan parameters the effective dose can be reduced, however lowering the accuracy of 3-D bone geometry reconstruction. [Copyright &y& Elsevier]