Improving bone strength prediction in human proximal femur specimens through geometrical characterization of trabecular bone microarchitecture and support vector regression.

We investigate the use of different trabecular bone descriptors and advanced machine learning tech niques to complement standard bone mineral density (BMD) measures derived from dual-energy x-ray absorptiometry (DXA) for improving clinical assessment of osteoporotic fracture risk. For this purpose, volumes of interest were extracted from the head, neck, and trochanter of 146 ex vivo proximal femur specimens on multidetector computer tomography. The trabecular bone captured was characterized with (1) statistical moments of the BMD distribution, (2) geometrical features derived from the scaling index method (SIM), and (3) morphometric parameters, such as bone fraction, trabecular thickness, etc. Feature sets comprising DXA BMD and such supplemental features were used to predict the failure load (FL) of the specimens, previously determined through biomechanical testing, with multiregression and support vector regression. Prediction performance was measured by the root mean square error (RMSE); correlation with measured FL was evaluated using the coefficient of determination R2. The best prediction performance was achieved by a combination of DXA BMD and SIM-derived geometric features derived from the femoral head (RMSE: 0.869 ± 0.121, R2: 0.68 ± 0.079), which was significantly better than DXA BMD alone (RMSE: 0.948 ± 0.119, R2: 0.61 ± 0.101) (p < 10-4). For multivariate feature sets, SVR outperformed multiregression (p < 0.05). These results suggest that supplementing standard DXA BMD measurements with sophisticated femoral trabecular bone characterization and supervised learning techniques can significantly improve biomechanical strength prediction in proximal femur specimens.

Cortical and trabecular bone structure analysis at the distal radius-prediction of biomechanical strength by DXA and MRI.

The purpose of this study was to investigate whether the combination of dual-energy X-ray absorptiometry (DXA)-based bone mass and magnetic resonance imaging (MRI)-based cortical and trabecular structural measures improves the prediction of radial bone strength. Thirty-eight left forearms were harvested from formalin-fixed human cadavers. Bone mineral content (BMC) and bone mineral density (BMD) of the distal radius were measured using DXA. Cortical and trabecular structural measures of the distal radius were computed in high-resolution 1.5T MR images. Cortical measures included average cortical thickness and cross-sectional area. Trabecular measures included morphometric and texture parameters. The forearms were biomechanically tested in a fall simulation to measure absolute radial bone strength (failure load). Relative radial bone strength was determined by dividing radial failure loads by age, body mass index, radius length, and average radius cross-sectional area, respectively. DXA derived BMC and BMD showed statistically significant (p < 0.05) correlations with absolute and relative radial bone strength (r ≤ 0.78). Correlation coefficients for cortical and trabecular structural measures with absolute and relative radial bone strength amounted up to r = 0.59 and r = 0.74, respectively, (p < 0.05). In combination with DXA-based bone mass, trabecular but not, cortical structural measures, added in multiple regression models significant (p < 0.05) information in predicting absolute and relative radial bone strength (up to R adj = 0.88). Thus, a combination of DXA-based bone mass and MRI-based trabecular structural measures most accurately predicted absolute and relative radial bone strength, whereas structural measures of the cortex did not provide significant additional information in combination with DXA.

Proximal femur specimens: automated 3D trabecular bone mineral density analysis at multidetector CT--correlation with biomechanical strength measurement.

PURPOSE: To prospectively evaluate an automated volume of interest (VOI)-fitting algorithm for quantitative computed tomography (CT) of proximal femur specimens, correlate bone mineral density (BMD) with biomechanically determined bone strength in vitro, and compare that correlation with those observed at dual-energy x-ray absorptiometry (DXA) measurement of BMD. MATERIALS AND METHODS: The study was compliant with institutional and legislative requirements; donors had dedicated their body for education and research before death. Multidetector CT and DXA scans were acquired in 178 proximal femur specimens harvested from human cadavers (91 women, 87 men; mean age at death, 79 years +/- 10.2; range, 52-100 years). An automated VOI-fitting algorithm was used to calculate BMD and bone mineral content (BMC) in the head, neck, and trochanter from CT findings and pixel distribution parameters. The femur failure load (FL) was determined by using a mechanical test. Quantitative CT BMD, quantitative CT pixel distribution parameters, DXA BMD, and FL were correlated at multiple regression analysis. RESULTS: Mean precision errors in quantitative CT BMD measurements at segmentation with repositioning were 0.56%, 2.26%, and 0.61% for the head, neck, and trochanter, respectively. For the head, neck, and trochanter, respectively, r values were 0.77, 0.53, and 0.59 for the correlation between quantitative CT BMD and FL and 0.74, 0.55, and 0.65 for the correlation between quantitative CT BMC and FL (P < .001). Values ranged from 0.77 to 0.80 for correlations between DXA BMD and FL and from 0.73 to 0.82 for correlations between DXA BMC and FL (P < .001). In a multiple regression model that included quantitative CT pixel distributions, adjusted multivariate correlation coefficient values for correlations with FL increased to up to 0.88. CONCLUSION: Regional BMD of the proximal femur can be determined in vitro from quantitative CT data with high precision by using an automated VOI-fitting algorithm. The best multiple regression model for predicting FL included DXA BMD and regional quantitative CT BMD measurements.

The role of fabric in the quasi-static compressive mechanical properties of human trabecular bone from various anatomical locations.

Osteoporosis leads to an increased risk of bone fracture. While bone density and architecture can be assessed in vivo with increasing accuracy using CT and MRI, their relationship with the critical mechanical properties at various anatomical sites remain unclear. The objective of this study was to quantify the quasi-static compressive mechanical properties of human trabecular bone among different skeletal sites and compare their relationships with bone volume fraction and a measure of microstructural anisotropy called fabric. Over 600 trabecular bone samples from six skeletal sites were assessed by microCT and tested in uniaxial compression. Bone volume fraction correlated positively with elastic modulus, yield stress, ultimate stress, and the relationships depended strongly on skeletal site. The account of fabric improved these correlations substantially, especially when the data of all sites were pooled together, but the fabric-mechanical property relationships remained somewhat distinct among the anatomical sites. The study confirms that, beyond volume fraction, fabric plays an important role in determining the mechanical properties of trabecular bone and should be exploited in mechanical analysis of clinically relevant sites of the human skeleton.

Sex differences of human trabecular bone microstructure in aging are site-dependent.

UNLABELLED: In this study, we characterize bone microstructure, specifically sex differences, at multiple skeletal sites in 165 subjects >52 yr of age, using microCT technology in vitro. Significant sex differences are observed at the distal radius, femoral neck, and femoral trochanter, but not at the iliac crest, calcaneus, and lumbar vertebral body. Correlations in BV/TV between sites ranged from r = 0.13 to 0.56. INTRODUCTION: The goals of this study were (1) to assess potential sex differences of bone microstructure and their difference between skeletal sites and (2) to explore the relationship of trabecular microstructural properties between relevant skeletal sites. MATERIALS AND METHODS: Trabecular bone microstructural properties were measured in vitro in 165 subjects 52-99 yr of age using microCT. Defined volumes of interest (cylinders with 6 mm diameter and 6 mm length) were scanned at a resolution of 26 microm (isotropic) in six different anatomical sites: distal radius, femoral neck and trochanter, iliac crest, calcaneus, and second lumbar vertebral body. RESULTS: At the radius and femoral neck, trabecular bone displayed a more plate-like structure, thicker trabeculae, smaller separation/higher trabecular number, higher connectivity, and a higher degree of anisotropy in men than in women (p < 0.05). At the trochanter, men displayed more plate-like structure and thicker trabeculae (p < 0.05), but no differences in trabecular separation or other parameters compared with the women. At the calcaneus, iliac crest, and second lumbar vertebra none of the bone parameters displayed significant differences between sexes. The BV/TV at one site explained a range of only 2-32% of the variability at other sites. CONCLUSIONS: These results suggest that trabecular bone microstructural properties are remarkably heterogeneous throughout the skeleton. Significant differences between men and women are observed at some, but not at all, sites. The magnitude of sex differences in trabecular microstructure coincides with that of fracture incidence observed for some of the sites in epidemiological studies.

Association of geometric factors and failure load level with the distribution of cervical vs. trochanteric hip fractures.

UNLABELLED: We experimentally studied the distribution of hip fracture types at different structural mechanical strength. Femoral neck fractures were dominant at the lowest structural strength levels, whereas trochanteric fractures were more common at high failure loads. The best predictor of fracture type across all failure loads and in both sexes was the neck-shaft angle. INTRODUCTION: Bone geometry has been shown to be a potential risk factor for osteoporotic fractures. Risk factors have been shown to differ between cervical and trochanteric hip fractures. However, the determinants of cervical and trochanteric fractures at different levels of structural mechanical strength are currently unknown. In addition, it is not known if the distribution of fracture types differs between sexes. The aim of this experimental study on excised femora was to investigate whether there exist differences in the distribution of cervical and trochanteric fractures between different structural mechanical strength levels and different sexes and to identify the geometric determinants that predict a fracture type. MATERIALS AND METHODS: The sample was comprised of 140 cadavers (77 females: mean age, 81.7 years; 63 males: mean age, 79.1 years) from whom the left femora were excised for analysis. The bones were radiographed, and geometrical parameters were determined from the digitized X-rays. The femora were mechanically tested in a side impact configuration, simulating a sideways fall. After the mechanical test, the fracture patterns were classified into cervical and trochanteric. RESULTS: The overall proportion of cervical fractures was higher in females (74%) than in males (49%) (p = 0.002). The fracture type distribution differed significantly across load quartiles in females (p = 0.025), but not in males (p = 0.205). At the lowest load quartiles, 94.7% of fractures in female and 62.5% in males were femoral neck fractures. At the highest quartiles, in contrast, only 52.6% of fractures in females and 33.3% in males were cervical fractures. Among geometric variables, the neck-shaft angle was the best predictor of fracture type, with higher values in subjects with cervical fractures. This finding was made in females (p < 0.001) and males (p = 0.02) and was consistent across all failure load quartiles. CONCLUSIONS: Femoral neck fractures predominate at the lowest structural mechanical strength levels, whereas trochanteric fractures are more common at high failure loads. Females are more susceptible to femoral neck fractures than males. The best predictor of fracture type across all structural strength levels and both sexes was the neck-shaft angle.

Radius bone strength in bending, compression, and falling and its correlation with clinical densitometry at multiple sites.

This study comprehensively analyzes the ability of site-specific and nonsite-specific clinical densitometric techniques for predicting mechanical strength of the distal radius in different loading configurations. DXA of the distal forearm, spine, femur, and total body and peripheral quantitative computed tomography (pQCT) measurements of the distal radius (4, 20, and 33%) were obtained in situ (with soft tissues) in 129 cadavers, aged 80.16 +/- 9.8 years. Spinal QCT and calcaneal quantitative ultrasound (QUS) were performed ex situ in degassed specimens. The left radius was tested in three-point bending and axial compression, and the right forearm was tested in a fall configuration, respectively. Correlation coefficients with radius DXA were r = 0.89, 0.84, and 0.70 for failure in three-point bending, axial compression, and the fall simulation, respectively. The correlation with pQCT (r = 0.75 for multiple regression models with the fall) was not significantly higher than for DXA. Nonsite-specific measurements and calcaneal QUS displayed significantly (p < 0.01) lower correlation coefficients, and QUS did only contribute to the prediction of axial failure stress but not of failure load. We conclude that a combination of pQCT parameters involves only marginal improvement in predicting mechanical strength of the distal radius, nonsite-specific measurements are less accurate for this purpose, and QUS adds only little independent information to site-specific bone mass. Therefore, the noninvasive diagnosis of loss of strength at the distal radius should rely on site-specific measurements with DXA or pQCT and may be the earliest chance to detect individuals at risk of osteoporotic fracture.

Assessing bone status beyond BMD: evaluation of bone geometry and porosity by quantitative ultrasound of human finger phalanges.

UNLABELLED: In an in vitro study, we found significant associations between QUS variables and properties and geometrical parameters of the compact bone of human finger phalanges. QUS variables were not only related to BMD but also to other skeletal properties, which explained 70% of the variability of speed of sound. INTRODUCTION: Transverse transmission quantitative ultrasound (QUS) measurements at the finger phalanges are known to be correlated with BMD and to predict osteoporotic fractures. To determine which other skeletal properties are affected by ultrasound, we investigated the impact of density, geometry, and porosity on QUS variables in vitro. MATERIALS AND METHODS: Ultrasound variables were correlated with density, porosity, and geometrical characteristics of cortical bone. Additionally, we tested which combinations of geometry and bone properties best predicted the ultrasound results observed. Forty-four proximal phalanges from the middle finger were investigated at their distal metaphysis, similar to the typical clinical measurement procedure. Donor age ranged from 52 to 98 years (15 males and 29 females; mean age, 80.9 +/- 9.4 years). QUS variables were measured on the metaphysis of the phalanges using the DBMSonic 1200. Quantitative CT was used for the measurement of BMD, and high-resolution MRI was used for the measurement of porosity and geometrical variables. RESULTS: Speed of sound (SOS) and the clinically used variable AD-SOS correlated significantly with area, relative area, density, and porosity of the compact bone (R2 = 0.28-0.58, p < 0.0001). Signal amplitude correlated significantly only with relative area of the compact bone and area of the medullary canal (R2 = 0.18-0.20, p < 0.01). The combination of cortical area, density, and porosity improved the determination of SOS to R2 = 0.70, with a residual unexplained variability of 54 m/s (3.2%). CONCLUSIONS: These results clarify the impact of skeletal properties on QUS variables. SOS is affected by cortical area, cortical bone density, and cortical porosity, whereas attenuation only depends on geometry of the medulla. AD-SOS, the variable routinely measured in clinical practice, is primarily affected by cortical area. QUS of the finger phalanges is not only related to BMD but also to other skeletal properties.

Reproducibility and side differences of mechanical tests for determining the structural strength of the proximal femur.

UNLABELLED: In this experimental study, we evaluated the reproducibility error of mechanical strength tests of the proximal femur when simulating a fall on the trochanter. Based on side differences in femoral failure loads in 55 pairs of femora, we estimated the upper limit of the precision error to be 15% for the side impact test, whereas the intersubject variability was >40%. INTRODUCTION: Mechanical tests are commonly used as the gold standard for determining one of the main functions of bones, that is, to provide mechanical strength. However, it is unknown what magnitude of error is associated with these tests. Here we investigate the precision error and side difference of a side impact test of the proximal femur. MATERIALS AND METHODS: BMC was measured using DXA in 54 pairs of femora from donors 79.0 +/- 10.6 years of age. Bones were tested to failure, simulating a fall on the greater trochanter. RESULTS: Failure loads were 3951 +/- 1659N (CV% = 42%) on the right and 3900 +/- 1652N (CV% = 42%) on the left (no significant side difference). The average random difference of femoral BMC was 7 +/- 7% and that of femoral failure loads was 17 +/- 12%. The correlation between BMC and failure load was 79% (r2), but the association between side differences in failure load with those in BMC was only 4%. When confining the analysis to pairs with less than 5% differences in BMC (n = 31), side differences in failure loads were 15 +/- 13%. When correcting failure loads for side differences of BMC, the difference was 16 +/- 15% CONCLUSIONS: These results suggest that the upper limit of the precision error for femoral strength tests is approximately 15% in a side impact configuration. Given the large intersubject variability of failure loads, this test provides an efficient tool for determining the structural strength of the proximal femur in a fall.

Can novel clinical densitometric techniques replace or improve DXA in predicting bone strength in osteoporosis at the hip and other skeletal sites?

New peripheral techniques are now available for the diagnosis of osteoporosis, but their value in the clinical management of the disease remains controversial. This study tests the hypothesis that peripheral quantitative computed tomography (pQCT) at the distal radius and/or quantitative ultrasound (QUS) at the calcaneus can serve as replacement or improvement of current methodology (QCT and DXA) for predicting bone strength at the hip and other sites. In 126 human cadavers (age, 80.2 +/- 10.4 years), DXA of the femur, spine, and radius and pQCT of the radius were acquired with intact soft tissues. QCT (spine) and QUS (calcaneus) were performed ex situ in degassed specimens. Femoral failure loads were assessed in side impact and vertical loading. Failure loads of the thoracolumbar spine were determined at three levels in compression and those of the radius by simulating a fall. Site-specific DXA explained approximately 55% of the variability in femoral strength, whereas pQCT and QUS displayed a lower association (15-40%). QUS did not provide additional information on mechanical strength of the femur, spine, or radius. All techniques displayed similar capability in predicting a combined index of failure strength at these three sites, with only QUS exhibiting significantly lower associations than other methods. These experimental results suggest that clinical assessment of femoral fracture risk should preferably rely on femoral DXA, whereas DXA, QCT, and pQCT display similar capability of predicting a combined index of mechanical strength at the hip, spine, and radius.

Bone strength at clinically relevant sites displays substantial heterogeneity and is best predicted from site-specific bone densitometry.

In this study we test the hypotheses that mechanical bone strength in elderly individuals displays substantial heterogeneity among clinically relevant skeletal sites, that ex situ dual-energy X-ray absorptiometry (DXA) provides better estimates of bone strength than in situ DXA, but that a site-specific approach of bone densitometry is nevertheless superior for optimal prediction of bone failure under in situ conditions. DXA measurements were obtained of the lumbar spine, the left femur, the left radius, and the total body in 110 human cadavers (age, 80.6 +/- 10.5 years; 72 female, 38 male), including the skin and soft tissues. The bones were then excised, spinal and femoral DXA being repeated ex situ. Mechanical failure tests were performed on thoracic vertebra 10 and lumbar vertebra 3 (compressive loading of a functional unit), the left and right femur (side impact and vertical loading configuration), and the left and right distal radius (fall configuration, axial compression, and 3-point-bending). The failure loads displayed only very moderate correlation among sites (r = 0.39 to 0.63). Ex situ DXA displayed slightly higher correlations with failure loads compared with those of in situ DXA, but the differences were not significant and relatively small. Under in situ conditions, DXA predicted 50-60% of the variability in bone failure loads at identical (or closely adjacent) sites, but only around 20-35% at distant sites, advocating a site-specific approach of densitometry. These data suggest that mechanical competence in the elderly is governed by strong regional variation, and that its loss in osteoporosis may not represent a strictly systemic process.

Determinants and heterogeneity of mechanical competence throughout the thoracolumbar spine of elderly women and men.

Vertebral fractures represent the hallmark of osteoporosis. Here, we test the hypotheses that (sub)cortical bone strength and density predict failure better than trabecular core strength and density, and that elderly women display lower failure stress of thoracic vertebrae than men. We examined the vertebral bodies T3 to L5 in 39 spines from elderly donors (23 women; 16 men; age 79 +/- 11 years). Peripheral quantitative computed tomography was used to measure total, trabecular, and (sub)cortical bone density. Mechanical tests were performed in functional spinal units, planoparallel sections of vertebrae, trabecular cores, and (sub)cortical ring specimens. The failure stress decreased with descending vertebral level. Failure stress was highest for the (sub)cortical rings and planoparallel sections and lowest for the trabecular core. The failure stress did not differ significantly between men and women. Mechanical strength of the functional unit was more strongly correlated with the strength of the (sub)cortical ring (r = 0.78) than with that of the trabecular core (r = 0.62). However, total density was more highly correlated with mechanical strength of the same and remote vertebrae (r = 0.63) than trabecular (r = 0.50) or (sub)cortical density (r = 0.36), respectively. The results show that vertebral strength is similar in elderly women and men. Strength of (sub)cortical bone provides significantly better prediction of strength of functional spinal units than that of the trabecular core. However, total density predicts functional segment failure stress with higher accuracy than (sub)cortical or trabecular density and is thus recommended for predicting fracture strength clinically.

Trabecular bone structure of the distal radius, the calcaneus, and the spine: which site predicts fracture status of the spine best?

RATIONALE AND OBJECTIVES: To compare trabecular bone structure measures obtained in magnetic resonance images of the distal radius and the calcaneus as well as computed tomographic images of the spine versus bone mineral density (BMD) of the spine and the calcaneus in the prediction of osteoporotic spine fracture status. MATERIAL AND METHODS: High-resolution magnetic resonance images of the calcaneus and the distal radius and thin-section computed tomographic images of thoracic and lumbar vertebrae were obtained from 74 cadavers. Structure analysis was performed using parameters analogous to standard histomorphometry. BMD of the spine was determined by using quantitative computed tomography and of the calcaneus by using dual x-ray absorptiometry. Spine radiographs of these cadavers were assessed concerning vertebral deformities. RESULTS: The diagnostic performance in differentiating fracture and nonfracture subjects was highest for structure parameters in the spine and slightly lower for these parameters in the distal radius and for BMD of the spine. CONCLUSION: In this study structure parameters in the spine were best suited to predict the osteoporotic fracture status of the spine.

Multislice computed tomography of the distal radius metaphysis: relationship of cortical bone structure with gender, age, osteoporotic status, and mechanical competence.

We explore the relationship of region-specific densitometric and geometry-based (cortical) parameters at the distal radial metaphysis with gender, age, and osteoporotic status, using multislice computed tomography (CT). We specifically test the hypothesis that these parameters can improve the prediction of mechanical strength of the distal radius vs bone mass (bone mineral content [BMC]). The BMC was determined in 56 forearm specimens with peripheral dual-energy X-ray absorptiometry (DXA). Trabecular and cortical density and geometric properties of the metaphyseal cortex were determined using multislice CT and proprietary image analysis software. Specimens were tested to failure in a fall simulation, maintaining the integrity of the elbow joint and hand. Women displayed significantly lower failure strength (-34%), BMC (-35%), trabecular density (-26%), and cortical area (-12%) than men. The reduction of trabecular density with age and osteoporotic status was stronger than that of cortical density or thickness. DXA explained approx 50% (r2) of the variability in bone failure loads. This proportion was slightly increased (55%) when adding geometry-based parameters. The study suggests that high-resolution tomographic measurements with current clinical imaging methodology can marginally improve the prediction of mechanical failure strength. Further efforts are required to improve spatial resolution for determining metaphyseal cortical properties clinically.

Using Anisotropic 3D Minkowski Functionals for Trabecular Bone Characterization and Biomechanical Strength Prediction in Proximal Femur Specimens.

The ability of Anisotropic Minkowski Functionals (AMFs) to capture local anisotropy while evaluating topological properties of the underlying gray-level structures has been previously demonstrated. We evaluate the ability of this approach to characterize local structure properties of trabecular bone micro-architecture in ex vivo proximal femur specimens, as visualized on multi-detector CT, for purposes of biomechanical bone strength prediction. To this end, volumetric AMFs were computed locally for each voxel of volumes of interest (VOI) extracted from the femoral head of 146 specimens. The local anisotropy captured by such AMFs was quantified using a fractional anisotropy measure; the magnitude and direction of anisotropy at every pixel was stored in histograms that served as a feature vectors that characterized the VOIs. A linear multi-regression analysis algorithm was used to predict the failure load (FL) from the feature sets; the predicted FL was compared to the true FL determined through biomechanical testing. The prediction performance was measured by the root mean square error (RMSE) for each feature set. The best prediction performance was obtained from the fractional anisotropy histogram of AMF Euler Characteristic (RMSE = 1.01 ± 0.13), which was significantly better than MDCT-derived mean BMD (RMSE = 1.12 ± 0.16, p<0.05). We conclude that such anisotropic Minkowski Functionals can capture valuable information regarding regional trabecular bone quality and contribute to improved bone strength prediction, which is important for improving the clinical assessment of osteoporotic fracture risk.

Predicting the Biomechanical Strength of Proximal Femur Specimens with Minkowski Functionals and Support Vector Regression.

Regional trabecular bone quality estimation for purposes of femoral bone strength prediction is important for improving the clinical assessment of osteoporotic fracture risk. In this study, we explore the ability of 3D Minkowski Functionals derived from multi-detector computed tomography (MDCT) images of proximal femur specimens in predicting their corresponding biomechanical strength. MDCT scans were acquired for 50 proximal femur specimens harvested from human cadavers. An automated volume of interest (VOI)-fitting algorithm was used to define a consistent volume in the femoral head of each specimen. In these VOIs, the trabecular bone micro-architecture was characterized by statistical moments of its BMD distribution and by topological features derived from Minkowski Functionals. A linear multi-regression analysis and a support vector regression (SVR) algorithm with a linear kernel were used to predict the failure load (FL) from the feature sets; the predicted FL was compared to the true FL determined through biomechanical testing. The prediction performance was measured by the root mean square error (RMSE) for each feature set. The best prediction result was obtained from the Minkowski Functional surface used in combination with SVR, which had the lowest prediction error (RMSE = 0.939 ± 0.345) and which was significantly lower than mean BMD (RMSE = 1.075 ± 0.279, p<0.005). Our results indicate that the biomechanical strength prediction can be significantly improved in proximal femur specimens with Minkowski Functionals extracted from on MDCT images used in conjunction with support vector regression.

Introducing Anisotropic Minkowski Functionals and Quantitative Anisotropy Measures for Local Structure Analysis in Biomedical Imaging.

The ability of Minkowski Functionals to characterize local structure in different biological tissue types has been demonstrated in a variety of medical image processing tasks. We introduce anisotropic Minkowski Functionals (AMFs) as a novel variant that captures the inherent anisotropy of the underlying gray-level structures. To quantify the anisotropy characterized by our approach, we further introduce a method to compute a quantitative measure motivated by a technique utilized in MR diffusion tensor imaging, namely fractional anisotropy. We showcase the applicability of our method in the research context of characterizing the local structure properties of trabecular bone micro-architecture in the proximal femur as visualized on multi-detector CT. To this end, AMFs were computed locally for each pixel of ROIs extracted from the head, neck and trochanter regions. Fractional anisotropy was then used to quantify the local anisotropy of the trabecular structures found in these ROIs and to compare its distribution in different anatomical regions. Our results suggest a significantly greater concentration of anisotropic trabecular structures in the head and neck regions when compared to the trochanter region (p < 10-4 ). We also evaluated the ability of such AMFs to predict bone strength in the femoral head of proximal femur specimens obtained from 50 donors. Our results suggest that such AMFs, when used in conjunction with multi-regression models, can outperform more conventional features such as BMD in predicting failure load. We conclude that such anisotropic Minkowski Functionals can capture valuable information regarding directional attributes of local structure, which may be useful in a wide scope of biomedical imaging applications.

Scaling relations between trabecular bone volume fraction and microstructure at different skeletal sites.

In this study, we investigated the scaling relations between trabecular bone volume fraction (BV/TV) and parameters of the trabecular microstructure at different skeletal sites. Cylindrical bone samples with a diameter of 8mm were harvested from different skeletal sites of 154 human donors in vitro: 87 from the distal radius, 59/69 from the thoracic/lumbar spine, 51 from the femoral neck, and 83 from the greater trochanter. µCT images were obtained with an isotropic spatial resolution of 26µm. BV/TV and trabecular microstructure parameters (TbN, TbTh, TbSp, scaling indices (< > and σ of α and αz), and Minkowski Functionals (Surface, Curvature, Euler)) were computed for each sample. The regression coefficient ß was determined for each skeletal site as the slope of a linear fit in the double-logarithmic representations of the correlations of BV/TV versus the respective microstructure parameter. Statistically significant correlation coefficients ranging from r=0.36 to r=0.97 were observed for BV/TV versus microstructure parameters, except for Curvature and Euler. The regression coefficients ß were 0.19 to 0.23 (TbN), 0.21 to 0.30 (TbTh), -0.28 to -0.24 (TbSp), 0.58 to 0.71 (Surface) and 0.12 to 0.16 (<α>), 0.07 to 0.11 (<αz>), -0.44 to -0.30 (σ(α)), and -0.39 to -0.14 (σ(αz)) at the different skeletal sites. The 95% confidence intervals of ß overlapped for almost all microstructure parameters at the different skeletal sites. The scaling relations were independent of vertebral fracture status and similar for subjects aged 60-69, 70-79, and >79years. In conclusion, the bone volume fraction-microstructure scaling relations showed a rather universal character.

Predicting the Biomechanical Strength of Proximal Femur Specimens through High Dimensional Geometric Features and Support Vector Regression.

Estimating local trabecular bone quality for purposes of femoral bone strength prediction is important for improving the clinical assessment of osteoporotic hip fracture risk. In this study, we explore the ability of geometric features derived from the Scaling Index Method (SIM) in predicting the biomechanical strength of proximal femur specimens as visualized on multi-detector computed tomography (MDCT) images. MDCT scans were acquired for 50 proximal femur specimens harvested from human cadavers. An automated volume of interest (VOI)-fitting algorithm was used to define a consistent volume in the femoral head of each specimen. In these VOIs, the non-linear micro-structure of the trabecular bone was characterized by statistical moments of its BMD distribution and by local scaling properties derived from SIM. Linear multi-regression analysis and support vector regression with a linear kernel (SVRlin) were used to predict the failure load (FL) from the feature sets; the predicted FL was compared to the FL values determined through biomechanical testing. The prediction performance was measured by the root mean square error (RMSE) for each image feature on independent test set. The best prediction result was obtained from the SIM feature set with SVRlin, which had the lowest prediction error (RMSE = 0.842 ± 0.209) and which was significantly lower than the conventionally used mean BMD (RMSE = 1.103 ± 0.262,, p<0.005). Our results indicate that the biomechanical strength prediction can be significantly improved in proximal femur specimens on MDCT images by using high-dimensional geometric features derived from SIM with support vector regression.

Computational finite element bone mechanics accurately predicts mechanical competence in the human radius of an elderly population.

High-resolution peripheral quantitative computed tomography (HR-pQCT) is clinically available today and provides a non-invasive measure of 3D bone geometry and micro-architecture with unprecedented detail. In combination with microarchitectural finite element (µFE) models it can be used to determine bone strength using a strain-based failure criterion. Yet, images from only a relatively small part of the radius are acquired and it is not known whether the region recommended for clinical measurements does predict forearm fracture load best. Furthermore, it is questionable whether the currently used failure criterion is optimal because of improvements in image resolution, changes in the clinically measured volume of interest, and because the failure criterion depends on the amount of bone present. Hence, we hypothesized that bone strength estimates would improve by measuring a region closer to the subchondral plate, and by defining a failure criterion that would be independent of the measured volume of interest. To answer our hypotheses, 20% of the distal forearm length from 100 cadaveric but intact human forearms was measured using HR-pQCT. µFE bone strength was analyzed for different subvolumes, as well as for the entire 20% of the distal radius length. Specifically, failure criteria were developed that provided accurate estimates of bone strength as assessed experimentally. It was shown that distal volumes were better in predicting bone strength than more proximal ones. Clinically speaking, this would argue to move the volume of interest for the HR-pQCT measurements even more distally than currently recommended by the manufacturer. Furthermore, new parameter settings using the strain-based failure criterion are presented providing better accuracy for bone strength estimates.