Radiographic texture analysis of densitometric calcaneal images: relationship to clinical characteristics and to bone fragility

A computed tomography-based radiomics nomogram for predicting osteoporotic vertebral fractures: A longitudinal study

CONTEXT

Fractures are serious consequence of osteoporosis in old adults. However, few longitudinal studies showed the role of computed tomography (CT)-based radiomics in predicting osteoporotic fractures.

OBJECTIVE

We evaluated the performance of CT radiomics-based model for osteoporotic vertebral fractures (OVF) in a longitudinal study.

METHODS

7906 subjects without OVF who were aged over 50 years, and underwent CT scans between 2016 and 2019 were enrolled and followed up until 2021. Seventy-two cases of new OVF were identified. One hundred and forty-four people without OVF during follow-up were selected as control. Radiomics features were extracted from baseline CT images. CT values of trabecular bone, and area and density of erector spinae were determined. Cox regression analysis was used to identify the independent associated factors. The predictive performance of the nomogram was assessed using the receiver operating characteristic (ROC) curve, calibration curve and decision curve.

RESULTS

CT value of vertebra (adjusted hazard ratio (aHR) = 2.04, 95% confidence interval (CI): 1.07, 3.89), radiomics score (aHR = 6.56, 95%CI:3.47, 12.38) and area of erector spinae (aHR = 1.68, 95%CI: 1.02, 2.78) were independently associated with OVF. Radscore was associated with severe OVF (aHR = 6.00, 95% CI:2.78-12.93). The nomogram showed good discrimination with a C-index of 0.82 (95%CI: 0.77, 0.87). The area under the curve of nomogram and radscore were both higher than osteoporosis + muscle area for 3-year and 4-year risk of fractures (p < 0.05). Decision curve also demonstrated that the radiomics nomogram was useful.

CONCLUSIONS

Bone radiomics is associated with OVF and the nomogram based on radiomics signature and muscle provides a tool for the prediction of OVF.

Novel Radiomics-Clinical Model for the Noninvasive Prediction of New Fractures After Vertebral Augmentation

PURPOSE

To investigate the noninvasive prediction model for new fractures after percutaneous vertebral augmentation (PVA) based on radiomics signature and clinical parameters.

METHODS

Data from patients who were diagnosed with osteoporotic vertebral compression fracture (OVCF) and treated with PVA in our hospital between May 2014 and April 2019 were retrospectively analyzed. Radiomics features were extracted from T1-weighted magnetic resonance imaging (MRI) of the T11-L5 segments taken before PVA. Different radiomics models was developed by using linear discriminant analysis (LDA), multilayer perceptron (MLP), and stochastic gradient descent (SGD) classifiers. A nomogram was constructed by integrating clinical parameters and Radscore that calculated by the best radiomics model. The model performance was quantified in terms of discrimination, calibration and clinical usefulness.

RESULT

Four clinical parameters and 16 selected radiomics features were used for model development. The clinical model showed poor discrimination capability with area under the curves (AUCs) yielding of 0.522 in the training dataset and 0.517 in the validation dataset. The LDA, MLP and SGD classifier-based radiomics model had achieved AUCs of 0.793, 0.810, and 0.797 in the training dataset, and 0.719, 0.704, and 0.725 in the validation dataset, respectively. The nomogram showed the best performance with AUCs achieving 0.810 and 0.754 in the training and validation datasets, respectively. The decision curve analysis demonstrated the net benefit of the nomogram was higher than that of other models.

CONCLUSION

Our findings indicate that combining clinical features with radiomics features from pre-augmentation T1-weighted MRI can be used to develop a nomogram that can predict new fractures in patients after PVA.

Deep learning and SURF for automated classification and detection of calcaneus fractures in CT images

BACKGROUND AND OBJECTIVES

The calcaneus is the most fracture-prone tarsal bone and injuries to the surrounding tissue are some of the most difficult to treat. Currently there is a lack of consensus on treatment or interpretation of computed tomography (CT) images for calcaneus fractures. This study proposes a novel computer-assisted method for automated classification and detection of fracture locations in calcaneus CT images using a deep learning algorithm.

METHODS

Two types of Convolutional Neural Network (CNN) architectures with different network depths, a Residual network (ResNet) and a Visual geometry group (VGG), were evaluated and compared for the classification performance of CT scans into fracture and non-fracture categories based on coronal, sagittal, and transverse views. The bone fracture detection algorithm incorporated fracture area matching using the speeded-up robust features (SURF) method, Canny edge detection, and contour tracing.

RESULTS

Results showed that ResNet was comparable in accuracy (98%) to the VGG network for bone fracture classification but achieved better performance for involving a deeper neural network architecture. ResNet classification results were used as the input for detecting the location and type of bone fracture using SURF algorithm.

CONCLUSIONS

Results from real patient fracture data sets demonstrate the feasibility using deep CNN and SURF for computer-aided classification and detection of the location of calcaneus fractures in CT images.

Early diagnosis of osteoporosis using radiogrammetry and texture analysis from hand and wrist radiographs in Indian population

We propose an automated low cost tool for early diagnosis of onset of osteoporosis using cortical radiogrammetry and cancellous texture analysis from hand and wrist radiographs. The trained classifier model gives a good performance accuracy in classifying between healthy and low bone mass subjects.

INTRODUCTION

We propose a low cost automated diagnostic tool for early diagnosis of reduction in bone mass using cortical radiogrammetry and cancellous texture analysis of hand and wrist radiographs. Reduction in bone mass could lead to osteoporosis, a disease observed to be increasingly occurring at a younger age in recent times. Dual X-ray absorptiometry (DXA), currently used in clinical practice, is expensive and available only in urban areas in India. Therefore, there is a need to develop a low cost diagnostic tool in order to facilitate large-scale screening of people for early diagnosis of osteoporosis at primary health centers.

METHODS

Cortical radiogrammetry from third metacarpal bone shaft and cancellous texture analysis from distal radius are used to detect low bone mass. Cortical bone indices and cancellous features using Gray Level Run Length Matrices and Laws' masks are extracted. A neural network classifier is trained using these features to classify healthy subjects and subjects having low bone mass.

RESULTS

In our pilot study, the proposed segmentation method shows 89.9 and 93.5% accuracy in detecting third metacarpal bone shaft and distal radius ROI, respectively. The trained classifier shows training accuracy of 94.3% and test accuracy of 88.5%.

CONCLUSION

An automated diagnostic technique for early diagnosis of onset of osteoporosis is developed using cortical radiogrammetric measurements and cancellous texture analysis of hand and wrist radiographs. The work shows that a combination of cortical and cancellous features improves the diagnostic ability and is a promising low cost tool for early diagnosis of increased risk of osteoporosis.

Epidemiology of Vertebral Fractures

Vertebral fractures are one of the most common fractures associated with skeletal fragility and can cause as much morbidity as hip fractures. However, the epidemiology of vertebral fractures differs from that of osteoporotic fractures at other skeletal sites in important ways, largely because only one quarter to one-third of vertebral fractures are recognized clinically at the time of their occurrence and otherwise require lateral spine imaging to be recognized. This article first reviews the prevalence and incidence of clinical and radiographic vertebral fractures in populations across the globe and secular trends in the incidence of vertebral fracture over time. Next, associations of vertebral fractures with measures of bone mineral density and bone microarchitecture are reviewed followed by associations of vertebral fracture with various textural measures of trabecular bone, including trabecular bone score. Finally, the article reviews clinical risk factors for vertebral fracture and the association of vertebral fractures with morbidity, mortality, and other subsequent adverse health outcomes.

Spine bone texture assessed by trabecular bone score (TBS) predicts osteoporotic fractures in men: the Manitoba Bone Density Program

INTRODUCTION

One quarter of osteoporotic fractures occur in men. TBS, a gray-level measurement derived from lumbar spine DXA image texture, is related to microarchitecture and fracture risk independently of BMD. Previous studies reported the ability of spine TBS to predict osteoporotic fractures in women. Our aim was to evaluate the ability of TBS to predict clinical osteoporotic fractures in men.

METHODS

3620 men aged ≥50 (mean 67.6years) at the time of baseline DXA (femoral neck, spine) were identified from a database (Province of Manitoba, Canada). Health service records were assessed for the presence of non-traumatic osteoporotic fracture after BMD testing. Lumbar spine TBS was derived from spine DXA blinded to clinical parameters and outcomes. We used Cox proportional hazard regression to analyze time to first fracture adjusted for clinical risk factors (FRAX without BMD), osteoporosis treatment and BMD (hip or spine).

RESULTS

Mean followup was 4.5years. 183 (5.1%) men sustain major osteoporotic fractures (MOF), 91 (2.5%) clinical vertebral fractures (CVF), and 46 (1.3%) hip fractures (HF). Correlation between spine BMD and spine TBS was modest (r=0.31), less than correlation between spine and hip BMD (r=0.63). Significantly lower spine TBS were found in fracture versus non-fracture men for MOF (p<0.001), HF (p<0.001) and CVF (p=0.003). Area under the receiver operating characteristic curve (AUC) for incident fracture discrimination with TBS was significantly better than chance (MOF AUC=0.59, p<0.001; HF AUC=0.67, p<0.001; CVF AUC=0.57, p=0.032). TBS predicted MOF and HF (but not CVF) in models adjusted for FRAX without BMD and osteoporosis treatment. TBS remained a predictor of HF (but not MOF) after further adjustment for hip BMD or spine BMD.

CONCLUSION

We observed that spine TBS predicted MOF and HF independently of the clinical FRAX score, HF independently of FRAX and BMD in men. Studies with more incident fractures are needed to confirm these findings.

Vertebral microarchitecture and fragility fracture in men: a TBS study

INTRODUCTION

Although osteoporosis is considered a disease of women, 25% of the individuals with osteoporosis are men. BMD measurement by DXA is the gold standard used to diagnose osteoporosis and assess fracture risk. Nevertheless, BMD does not take into account alterations of microarchitecture. TBS is an index of bone microarchitecture extracted from the spine DXA. Previous studies have reported the ability of the spine TBS to predict osteoporotic fractures in women. This is the first case-controlled study in men to evaluate the potential diagnostic value of TBS as a complement to bone mineral density (BMD), by comparing men with and without fractures.

METHODS

To be eligible for this study, subjects had to be non-Hispanic US white men aged 40 and older. Furthermore, subjects were excluded if they have or have had previously any treatment or illness that may influence bone metabolism. Fractured subjects were included if the presence of at least one fracture was confirmed. Cases were matched for age (±3 years) and BMD (±0.04 g/cm(2)) with three controls. BMD and TBS were first retrospectively evaluated at AP spine (L1-L4) with a Prodigy densitometer (GE-Lunar, Madison, USA) and TBS iNsight® (Med-Imaps, France) in Lausanne University Hospital blinded from clinical outcome. Inter-group comparisons were undertaken using Student's t-tests or Wilcoxon signed rank tests. Odds ratios were calculated per one standard deviation decrease as well as areas under the receiver operating curve (AUC).

RESULTS

After applying inclusion/exclusion criteria, a group of 180 male subjects was obtained. This group consists of 45 fractured subjects (age=63.3±12.6 years, BMI=27.1±4.2 kg/m(2)) and 135 control subjects (age=62.9±11.9 years, BMI=26.7±3.9 kg/m(2)) matched for age (p=0.86) and BMD (p=0.20). A weak correlation was obtained between TBS and BMD and between TBS and BMI (r=0.27 and r=-0.28, respectively, p<0.01). Subjects with fracture have a significant lower TBS compared to control subjects (p=0.013), whereas no differences were obtained for BMI, height and weight (p>0.10). TBS OR per standard deviation is 1.55 [1.09-2.20] for all fracture type. When considering vertebral fracture only TBS OR reached 2.07 [1.14-3.74].

CONCLUSION

This study showed the potential use of TBS in men. TBS revealed a significant difference between fractured and age- and spine BMD-matched nonfractured subjects. These results are consistent with those previously reported on for men of other nationalities.

Radiographical texture analysis improves the prediction of vertebral fracture: an ex vivo biomechanical study

STUDY DESIGN

Compression biomechanical tests using fresh cadaveric thoracolumbar motion segments.

OBJECTIVE

The purpose of this study was to determine if the combination of bone texture parameters using bone microarchitecture, and bone mineral density (BMD) measurement by dual-energy x-ray absorptiometry provided a better prediction of vertebral fracture than BMD evaluation alone.

SUMMARY OF BACKGROUND DATA

Bone strength is routinely evaluated using BMD, as measured by dual-energy x-ray absorptiometry. Currently, there is an ongoing debate about the strengths and limitations of bone densitometry in clinical practice. To assess the fracture risk properly, other factors are important to be taken into account such as the macro- and microarchitecture of the bone. Recently, a new high-resolution x-ray device with direct digitization, named bone microarchitecture (BMA, D3A Medical Systems), has been developed to provide a better precision of texture parameters than those previously obtained on digitized films.

METHODS

Twenty-seven 3-level thoracolumbar motion segments (T11, T12, L1, and L2, L3, L4) of excised spines, obtained at the Anatomy Department of Marseille, were studied using bone microarchitecture to estimate 3 textural parameters: fractal parameter Hmean, co-occurrence matrix, and run-length matrix, dual-energy x-ray absorptiometry to measure BMD, and mechanical compression tests to failure. All specimens were examined by computed tomography before and after compression. The prediction of the vertebral failure load was evaluated using multiple regression analyses.

RESULTS

Twenty-seven vertebral fractures were observed with a mean failure load of 2636.3 N (standard deviation, 996 N). Fractal parameter Hmean, co-occurrence matrix, and run-length matrix were each significantly correlated with BMD (P< 0.01) and bone strength (P< 0.01). Combining bone texture parameters and BMD significantly improved the fracture load prediction from adjusted r = 0.701 to adjusted r = 0.806 (P< 0.01).

CONCLUSION

In these excised vertebrae, the combination of bone texture parameters with BMD demonstrated a better performance in the failure load prediction than that of BMD alone.

LEVEL OF EVIDENCE

N/A.

Bone texture analysis is correlated with three-dimensional microarchitecture and mechanical properties of trabecular bone in osteoporotic femurs

Fracture of the proximal femur is a major public health problem in elderly persons. It has recently been suggested that combining texture analysis and bone mineral density measurement improves the failure load prediction in human femurs. In this study, we aimed to compare bone texture analysis with three-dimensional (3D) microarchitecture and mechanical properties of trabecular bone in osteoporotic femurs. Eight femoral heads from osteoporotic patients who fractured their femoral neck provided 31 bone cores. Bone samples were studied using a new high-resolution digital X-ray device (BMA™, D3A Medical Systems) allowing for texture analysis with fractal parameter H (mean), and were examined using micro-computed tomography (microCT) for 3D microarchitecture. Finally, uniaxial compression tests to failure were performed to estimate failure load and apparent modulus of bone samples. The fractal parameter H (mean) was strongly correlated with bone volume fraction (BV/TV) (r = 0.84) and trabecular thickness (Tb.Th) (r = 0.91) (p < 0.01). H (mean) was also markedly correlated with failure load (r = 0.84) and apparent modulus (r = 0.71) of core samples (p < 0.01). Bone volume fraction (BV/TV) and trabecular thickness (Tb.Th) demonstrated significant correlations with failure load (r = 0.85 and 0.72, respectively) and apparent modulus (r = 0.72 and 0.64, respectively) (p < 0.01). Overall, the best predictors of failure load were H (mean), bone volume fraction, and trabecular thickness, with r (2) coefficients of 0.83, 0.76, and 0.80 respectively. This study shows that the fractal parameter H (mean) is correlated with 3D microCT parameters and mechanical properties of femoral head bone samples, which suggests that radiographic texture analysis is a suitable approach for trabecular bone microarchitecture assessment in osteoporotic femurs.

[Microarchitecture assessment of human trabecular bone: description of methods]

Trabecular bone microarchitecture changes in relation to mechanical stress, effects of age, osteoporosis and anti-osteoporotic drugs. In vivo, these anomalies can be evaluated using textural parameters on high resolution radiographs and images of DXA. It is possible to extract morphological and topological parameters: apparent on MRI images and 3D with a dedicated device called High resolution peripheral quantitative computed tomography (HR-pQCT) with a resolution close to the size of the trabeculae. In vitro, it is possible to obtain on bone samples a 2D analysis by histomorphometry and a 3D analysis from 10 µm images obtained by synchrotron radiation or conventional micro-CT.

Bone texture analysis of human femurs using a new device (BMA™) improves failure load prediction

We measured bone texture parameters of excised human femurs with a new device (BMA™). We also measured bone mineral density by DXA and investigated the performance of these parameters in the prediction of failure load. Our results suggest that bone texture parameters improve failure load prediction when added to bone mineral density.

INTRODUCTION

Bone mineral density (BMD) is a strong determinant of bone strength. However, nearly half of the fractures occur in patients with BMD which does not reach the osteoporotic threshold. In order to assess fracture risk properly, other factors are important to be taken into account such as clinical risk factors as well as macro- and microarchitecture of bone. Bone microarchitecture is usually assessed by high-resolution QCT, but this cannot be applied in routine clinical settings due to irradiation, cost and availability concerns. Texture analysis of bone has shown to be correlated to bone strength.

METHODS

We used a new device to get digitized X-rays of 12 excised human femurs in order to measure bone texture parameters in three different regions of interest (ROIs). We investigated the performance of these parameters in the prediction of the failure load using biomechanical tests. Texture parameters measured were the fractal dimension (Hmean), the co-occurrence matrix, and the run length matrix. We also measured bone mineral density by DXA in the same ROIs as well as in standard DXA hip regions.

RESULTS

The Spearman correlation coefficient between BMD and texture parameters measured in the same ROIs ranged from -0.05 (nonsignificant (NS)) to 0.57 (p = 0.003). There was no correlation between Hmean and co-occurrence matrix nor Hmean and run length matrix in the same ROI (r = -0.04 to 0.52, NS). Co-occurrence matrix and run length matrix in the same ROI were highly correlated (r = 0.90 to 0.99, p < 0.0001). Univariate analysis with the failure load revealed significant correlation only with BMD results, not texture parameters. Multiple regression analysis showed that the best predictors of failure load were BMD, Hmean, and run length matrix at the femoral neck, as well as age and sex, with an adjusted r (2) = 0.88. Added to femoral neck BMD, Hmean and run length matrix at the femoral neck (without the effect of age and sex) improved failure load prediction (compared to femoral neck BMD alone) from adjusted r (2) = 0.67 to adjusted r (2) = 0.84.

CONCLUSION

Our results suggest that bone texture measurement improves failure load prediction when added to BMD.

Combination of texture analysis and bone mineral density improves the prediction of fracture load in human femurs

Twenty-one excised femurs were studied using (1) a high-resolution digital X-ray device to estimate three textural parameters, (2) dual-energy X-ray absorptiometry (DXA) to measure bone mineral density (BMD), and (3) mechanical tests to failure. Textural parameters significantly correlated with BMD (p < 0.05) and bone strength (p < 0.05). Combining texture parameters and BMD significantly improved the fracture load prediction from adjusted r(2) = 0.74 to adjusted r(2) =0.82 (p < 0.05).

INTRODUCTION

The purpose of this study is to determine if the combination of bone texture parameters using a new high-resolution X-ray device and BMD measurement by DXA provided a better prediction of femoral failure load than BMD evaluation alone.

METHODS

The proximal ends of 21 excised femurs were studied using (1) a high-resolution digital X-ray device (BMA, D3A Medical Systems) to estimate three textural parameters: fractal parameter Hmean, co-occurrence, and run-length matrices, (2) DXA to measure BMD, and (3) mechanical tests to failure in a side-impact configuration. Regions of interest in the femoral neck, intertrochanteric region, and greater trochanter were selected for DXA and bone texture analysis. Every specimen was scanned twice with repositioning before mechanical testing to assess reproducibility using intraclass correlation coefficient (ICC) with 95% confidence interval. The prediction of femoral failure load was evaluated using multiple regression analysis.

RESULTS

Thirteen femoral neck and 8 intertrochanteric fractures were observed with a mean failure load of 2,612 N (SD, 1,382 N). Fractal parameter Hmean, co-occurrence, and run-length matrices each significantly correlated with site-matched BMD (p < 0.05) and bone strength (p < 0.05). The ICC of the textural parameters varied between 0.65 and 0.90. Combining bone texture parameters and BMD significantly improved the fracture load prediction from adjusted r(2) =0.74 to adjusted r(2) = 0.82 (p < 0.05).

CONCLUSION

In these excised femurs, the combination of bone texture parameters with BMD demonstrated a better performance in the failure load prediction than that of BMD alone.

Bone microarchitecture assessed by TBS predicts osteoporotic fractures independent of bone density: the Manitoba study

The measurement of BMD by dual-energy X-ray absorptiometry (DXA) is the "gold standard" for diagnosing osteoporosis but does not directly reflect deterioration in bone microarchitecture. The trabecular bone score (TBS), a novel gray-level texture measurement that can be extracted from DXA images, correlates with 3D parameters of bone microarchitecture. Our aim was to evaluate the ability of lumbar spine TBS to predict future clinical osteoporotic fractures. A total of 29,407 women 50 years of age or older at the time of baseline hip and spine DXA were identified from a database containing all clinical results for the Province of Manitoba, Canada. Health service records were assessed for the incidence of nontraumatic osteoporotic fracture codes subsequent to BMD testing (mean follow-up 4.7 years). Lumbar spine TBS was derived for each spine DXA examination blinded to clinical parameters and outcomes. Osteoporotic fractures were identified in 1668 (5.7%) women, including 439 (1.5%) spine and 293 (1.0%) hip fractures. Significantly lower spine TBS and BMD were identified in women with major osteoporotic, spine, and hip fractures (all p < 0.0001). Spine TBS and BMD predicted fractures equally well, and the combination was superior to either measurement alone (p < 0.001). Spine TBS predicts osteoporotic fractures and provides information that is independent of spine and hip BMD. Combining the TBS trabecular texture index with BMD incrementally improves fracture prediction in postmenopausal women.

MRI texture analysis of femoral neck: Detection of exercise load-associated differences in trabecular bone

PURPOSE

To assess the ability of co-occurrence matrix-based texture parameters to detect exercise load-associated differences in MRI texture at the femoral neck cross-section.

MATERIALS AND METHODS

A total of 91 top-level female athletes representing five differently loading sports and 20 referents participated in this cross-sectional study. Axial T1-weighted FLASH and T2*-weighted MEDIC sequence images of the proximal femur were obtained with a 1.5T MRI. The femoral neck trabecular bone at the level of the insertion of articular capsule was divided manually into regions of interest representing four anatomical sectors (anterior, posterior, superior, and inferior). Selected co-occurrence matrix-based texture parameters were used to evaluate differences in apparent trabecular structure between the exercise loading groups and anatomical sectors of the femoral neck.

RESULTS

Significant differences in the trabecular bone texture, particularly at the superior femoral neck, were observed between athletes representing odd-impact (soccer and squash) and high-magnitude exercise loading (power-lifting) groups and the nonathletic reference group.

CONCLUSION

MRI texture analysis provides a quantitative method for detecting and classifying apparent structural differences in trabecular bone that are associated with specific exercise loading.