Radiographic bone texture analysis is correlated with 3D microarchitecture in the femoral head, and improves the estimation of the femoral neck fracture risk when combined with bone mineral density

Texture Parameters Measured by UHF-MRI and CT Scan Provide Information on Bone Quality in Addition to BMD: A Biomechanical Ex Vivo Study

The current definition of osteoporosis includes alteration of bone quality. The assessment of bone quality is improved by the development of new texture analysis softwares. Our objectives were to assess if proximal femoral trabecular bone texture measured in Ultra high field (UHF) 7 Tesla MRI and CT scan were related to biomechanical parameters, and if the combination of texture parameters and areal bone mineral density (aBMD) measured by dual-energy X-ray absorptiometry provided a better prediction of femoral failure than aBMD alone. The aBMD of 16 proximal femur ends from eight cadavers were investigated. Nineteen textural parameters were computed in three regions or volumes of interest for each specimen on UHF MRI and CT scan. Then, the corresponding failure load and failure stress were calculated thanks to mechanical compression test. aBMD was not correlated to failure load (R² = 0.206) and stress (R² = 0.153). The failure load was significantly correlated with ten parameters in the greater trochanter using UHF MRI, and with one parameter in the neck and the greater trochanter using CT scan. Eight parameters in the greater trochanter using UHF MRI combined with aBMD improved the failure load prediction, and seven parameters improved the failure stress prediction. Our results suggest that textural parameters provide additional information on the fracture risk of the proximal femur when aBMD is not contributive.

Prediction Models for Prognosis of Femoral Neck–Fracture Patients 6 Months after Total Hip Arthroplasty

Purpose

To establish prediction models for 6-month prognosis in femoral neck–fracture patients receiving total hip arthroplasty (THA).

Patients and Methods

In total, 182 computed tomography image pairs from 85 patients were collected and divided into a training set (n=127) and testing set (n=55). Least absolute shrinkage–selection operator regression was used for selecting optimal predictors. A random-forest algorithm was used to establish the prediction models, which were evaluated for accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC).

Results

The best model in this study was constructed based on demographic data, preoperative laboratory indicators, and three preoperative radiomic features. In the random-forest model, activated partial thromboplastin time, a preoperative radiomic feature (maximum diameter), and fibrinogen were important variables correlating with patient outcomes. The AUC, sensitivity, specificity, PPV, NPV, and accuracy in the training set were 0.986 (95% CI 0.971–1), 0.925 (95% CI 0.862–0.988), 0.983 (95% CI 0.951–1.016), 0.984 (95% CI 0.953–1.014), 0.922 (95% CI 0.856–0.988), and 0.953 (95% CI 0.916–0.990), respectively. The AUC, sensitivity, specificity, PPV, NPV, and accuracy in the testing set were 0.949 (95% CI 0.885–1), 0.767 (95% CI 0.615–0.918), 1 (95% CI 1–1), 1 (95% CI 1–1), 0.781 (95% CI 0.638–0.924), and 0.873 (95% CI 0.785–0.961), respectively.

Conclusion

The model based on demographic, preoperative clinical, and preoperative radiomic data showed the best predictive ability for 6-month prognosis in the femoral neck–fracture patients receiving THA.

Computed Tomography Radiomics Kinetics as Early Imaging Correlates of Osteoradionecrosis in Oropharyngeal Cancer Patients

Osteoradionecrosis (ORN) is a major side-effect of radiation therapy in oropharyngeal cancer (OPC) patients. In this study, we demonstrate that early prediction of ORN is possible by analyzing the temporal evolution of mandibular subvolumes receiving radiation. For our analysis, we use computed tomography (CT) scans from 21 OPC patients treated with Intensity Modulated Radiation Therapy (IMRT) with subsequent radiographically-proven ≥ grade II ORN, at three different time points: pre-IMRT, 2-months, and 6-months post-IMRT. For each patient, radiomic features were extracted from a mandibular subvolume that developed ORN and a control subvolume that received the same dose but did not develop ORN. We used a Multivariate Functional Principal Component Analysis (MFPCA) approach to characterize the temporal trajectories of these features. The proposed MFPCA model performs the best at classifying ORN vs. Control subvolumes with an area under curve (AUC) = 0.74 [95% confidence interval (C.I.): 0.61–0.90], significantly outperforming existing approaches such as a pre-IMRT features model or a delta model based on changes at intermediate time points, i.e., at 2- and 6-month follow-up. This suggests that temporal trajectories of radiomics features derived from sequential pre- and post-RT CT scans can provide markers that are correlates of RT-induced mandibular injury, and consequently aid in earlier management of ORN.

Influence of thermodisinfection on microstructure of human femoral heads: duration of heat exposition and compressive strength

Allogeneic bone derived from living donors being necessary to match demand for bone transplantation and thermodisinfection of femoral heads is an established sterilization method. During the thermodisinfection the peripheral bone is exposed to maximum 86 °C for 94 min providing 82.5 °C within the center of the femoral head for at least 15 min. This study examined the compression force of the central and representative peripheral regions of native and thermodisinfected human femoral heads to observe wether different duration and intensity of heat exposure might alter mechanic behaviour. Slices from the equatorial region of human femoral heads were taken from each 14 native and thermodisinfected human femoral heads. The central area revealed a significantly higher compression force for native (p ≤ 0.001) and for thermodisinfected bone (p = 0.002 and p = 0.005) compared with peripheral regions since no relevant differences were found between the peripheral and intermediate areas themselves. A small reduction of compression force for thermodisinfected bone was shown since this did not appear significant due to the small number of specimens. The heat exposure did not alter the pre-existing anatomical changes of the microarchitecture of the native femoral heads from the center towards the peripheral regions. The heterogeneity of microstructure of the femoral head might be of interest concerning clinical applications of bone grafts since the difference between native and thermodisinfected bone appears moderate as shown previously. The different quantity of heat exposure did not reveal any significant influence on compression force which might enable thermodisinfection of preformed bone pieces for surgical indications.

Non-operative treatment is a reliable option in over two thirds of patients with Garden I hip fractures. Rates and risk factors for failure in 298 patients

BACKGROUND

Non-operative treatment for impacted femoral neck fractures is a now rarely used strategy whose indications are controversial. No outcome predictors have been convincingly identified, in part due to the small sizes of available studies. We conducted a large retrospective study with the following objectives: (1) to evaluate the percentage of patients older than 65 years of age with non-operatively treated Garden I femoral neck fractures who experience secondary displacement, (2) to identify predictors of secondary displacement, and (3) to determine the frequency of non-operative treatment failure due to any cause and requiring joint replacement surgery.

HYPOTHESIS

Non-operative treatment is reliable in patients older than 65 years of age with Garden I femoral neck fractures.

MATERIAL AND METHODS

Approval was obtained from the French data protection authority to conduct a retrospective observational study of information in the Marseille university hospitals database. Consecutive patients who were older than 65 years of age at traumatology department admission for Garden I femoral neck fractures managed non-operatively between January 2007 and December 2017 were included. Non-operative treatment consisted in a walking test on day 1 followed by radiographs on days 2, 7, 14, 21, and 45 and after 3 and 12 months. Patients with secondary displacement underwent hip arthroplasty. Demographic data, cognitive performance, and radiological parameters were collected for each patient. We evaluated the rates of secondary displacement avascular necrosis, and non-union.

RESULTS

We included 298 patients with a mean age of 82 years (range, 65-101). Mean follow-up was 5±3 years. Secondary displacement occurred in 91 (30%) patients, at a mean of 16 days (range 2-45 days) after the fracture. Avascular necrosis of the femoral head developed in 13 (4.3%) patients and non-union in 11 (3.7%) patients. Secondary displacement was significantly associated with hypnotic treatment (OR, 4.1; 95%CI, 2.2-7.5; p=0.039), institutionalisation (OR, 6.7; 95%CI, 3.1-14.8; p=0.028), a history of repeated falls (OR, 13.5; 95%CI, 6.3-8.4; p<0.0001), having three or more comorbidities (OR, 3.2; 95%CI, 1.7-5.8; p=0.046), and having dementia (OR, 3.5; 95%CI, 1.7-6.9; p=0.0003). Secondary displacement occurred in 18 (12%) of the 151 community-dwelling patients with normal cognition and no history of repeated falls compared to 37 (75%) of the 50 institutionalised patients with dementia.

DISCUSSION

Non-operative treatment was effective in 196 (66%) of 298 patients with Garden I femoral neck fractures. Significant risk factors for secondary displacement were dementia, institutionalisation, hypnotic treatment, multiple comorbidities, and a history of repeated falls. Of 151 community-dwelling patients with normal cognition and no repeated falls, 133 (88%) achieved a full recovery with non-operative treatment alone.

LEVEL OF EVIDENCE

IV, retrospective cohort study.

Bone texture analysis for prediction of incident radiographic hip osteoarthritis using machine learning: data from the Cohort Hip and Cohort Knee (CHECK) study

OBJECTIVE

To assess the ability of radiography-based bone texture variables in proximal femur and acetabulum to predict incident radiographic hip osteoarthritis (rHOA) over a 10 years period.

DESIGN

Pelvic radiographs from CHECK at baseline (987 hips) were analyzed for bone texture using fractal signature analysis (FSA) in proximal femur and acetabulum. Elastic net (machine learning) was used to predict the incidence of rHOA (including Kellgren-Lawrence grade (KL) ≥ 2 or total hip replacement (THR)), joint space narrowing score (JSN, range 0-3), and osteophyte score (OST, range 0-3) after 10 years. Performance of prediction models was assessed using the area under the receiver operating characteristic curve (ROC AUC).

RESULTS

Of the 987 hips without rHOA at baseline, 435 (44%) had rHOA at 10-year follow-up. Of the 667 hips with JSN grade 0 at baseline, 471 (71%) had JSN grade ≥ 1 at 10-year follow-up. Of the 613 hips with OST grade 0 at baseline, 526 (86%) had OST grade ≥ 1 at 10-year follow-up. AUCs for the models including age, gender, and body mass index (BMI) to predict incident rHOA, JSN, and OST were 0.59, 0.54, and 0.51, respectively. The inclusion of bone texture variables in the models improved the prediction of incident rHOA (ROC AUC 0.68 and 0.71 when baseline KL was also included in the model) and JSN (ROC AUC 0.62), but not incident OST (ROC AUC 0.52).

CONCLUSION

Bone texture analysis provides additional information for predicting incident rHOA or THR over 10 years.

Comparison of bone texture between normal individuals and patients with Kashin-Beck disease from plain radiographs in knee

To compare tibial bone texture between Kashin-Beck disease (KBD) patients and normal individuals from plain radiographs using an advanced image analysis. Plain knee radiographs were obtained from KBD patients (n = 49) and age-matched healthy controls (n = 98). KBD were graded with diagnostic criteria WS/T 207-2010. The textural values related to bone structure from medial and lateral tibial subchondral and trabecular bones were evaluated using entropy of Laplacian-based image (E_Lap), entropy of local binary patterns (E_LBP), homogeneity indices (HI) of local angles (HI_Mean, HI_Perp and HI_Paral), and fractal dimensions from horizontal (FD_Hor) and vertical (FD_Ver) structures. KBD patients were shorter in height and lighter in weight, and their tibial width was wider than controls. Anatomical angle of KBD patients showed more genu valgus. Total KBD patients and subgroups had higher E_Lap, HI_Mean, HI_Perp and HI_Paral in detected tibial subchondral and trabecular bones than controls, except E_Lap in lateral subchondral bone. E_LBP, FD_Hor and FD_Ver from the detected tibial bone in KBD patients and subgroups were lower than controls, except FD_Ver in lateral trabecular bone. Our results indicate that micro-scale in bone texture in KBD-affected knees can be quantitatively examined from plain radiographs using an advanced image analysis.

Quantification of Volumetric Bone Mineral Density of Proximal Femurs Using a Two-Compartment Model and Computed Tomography Images

Objectives

Dual-energy X-ray absorptiometry (DXA) is frequently used to measure the areal bone mineral density (aBMD) in clinical practice. However, DXA measurements are affected by the bone thickness and the body size and are unable to indicate nonosseous areas within the trabecular bone. This study aims to quantify the volumetric bone mineral density (vBMD) using computed tomography (CT) images and the two-compartment model (TCM) methods.

Methods

The TCM method was proposed and validated by dipotassium phosphate (K₂HPO₄) phantoms and a standard forearm phantom. 28 cases with DXA scans and pelvic CT scans acquired within six months were retrospectively collected. The vBMD calculated by TCM was compared with the aBMD obtained from DXA.

Results

For the K₂HPO₄ phantoms with vBMD ranging from 0.135 to 0.467 g/cm³, the average difference between the real and calculated vBMD was 0.009 g/cm³ and the maximum difference was 0.019 g/cm³. For the standard forearm phantom with vBMD of 0.194, 0.103, and 0.054 g/cm³, the average differences between the real and calculated vBMD were 0.017, 0.014, and 0.011 g/cm³. In the clinical CT image validation, a good linear relationship between vBMD and aBMD was observed with the Pearson correlation coefficient of 0.920 (p < 0.01).

Conclusions

The proposed TCM method in combination with the homemade cortical bone equivalent phantom provides accurate quantification and spatial distribution of bone mineral content.

Association between radiography-based subchondral bone structure and MRI-based cartilage composition in postmenopausal women with mild osteoarthritis

OBJECTIVE

Our aim was to investigate the relation between radiograph-based subchondral bone structure and cartilage composition assessed with delayed gadolinium enhanced magnetic resonance imaging of cartilage (dGEMRIC) and T₂ relaxation time.

DESIGN

Ninety-three postmenopausal women (Kellgren-Lawrence grade 0: n = 13, 1: n = 26, 2: n = 54) were included. Radiograph-based bone structure was assessed using entropy of the Laplacian-based image (E_Lap) and local binary patterns (E_LBP), homogeneity indices of the local angles (HI_Angles,mean, HI_Angles,Perp, HI_Angles,Paral), and horizontal (FD_Hor) and vertical fractal dimensions (FD_Ver). Mean dGEMRIC index and T₂ relaxation time of tibial cartilage were calculated to estimate cartilage composition.

RESULTS

HI_Angles,mean (r_s = -0.22) and HI_Angles,Paral (r_s = -0.24) in medial subchondral bone were related (P < 0.05) to dGEMRIC index of the medial tibial cartilage. E_Lap (r_s = -0.23), FD_Hor,0.34 mm (r = 0.21) and FD_Ver,0.68 mm (r = 0.24) in medial subchondral bone were related (P < 0.05) to T₂ relaxation time values of the medial tibial cartilage. FD_Hor at different scales in lateral subchondral bone were related (P < 0.01) to dGEMRIC index (r = 0.29-0.41) and T₂ values of lateral tibial cartilage (r = -0.28 to -0.36). FD_Ver at larger scales were related (P < 0.05) to dGEMRIC index (r = 0.24-0.25) and T₂ values of lateral tibial cartilage (r = -0.21). HI_Angles,Paral (r = -0.25) and FD_Ver,0.68 mm (r_s = 0.22) in the lateral tibial trabecular bone were related (P < 0.05) to dGEMRIC index of the lateral tibial cartilage.

CONCLUSION

Our results support the presumption that several tissues are affected in the early osteoarthritis (OA). Furthermore, they indicate that the detailed analysis of radiographs may serve as a complementary imaging tool for OA studies.

Differences in tibial subchondral bone structure evaluated using plain radiographs between knees with and without cartilage damage or bone marrow lesions - the Oulu Knee Osteoarthritis study

Objectives

To investigate whether subchondral bone structure from plain radiographs is different between subjects with and without articular cartilage damage or bone marrow lesions (BMLs).

Methods

Radiography-based bone structure was assessed from 80 subjects with different stages of knee osteoarthritis using entropy of Laplacian-based image (E_Lap) and local binary patterns (E_LBP), homogeneity index of local angles (HI_Angles,mean), and horizontal (FD_Hor) and vertical fractal dimensions (FD_Ver). Medial tibial articular cartilage damage and BMLs were scored using the magnetic resonance imaging osteoarthritis knee score. Level of statistical significance was set to p < 0.05.

Results

Subjects with medial tibial cartilage damage had significantly higher FD_Ver and E_LBP as well as lower E_Lap and HI_Angles,mean in the medial tibial subchondral bone region than subjects without damage. FD_Hor, FD_Ver, and E_LBP were significantly higher, whereas E_Lap and HI_Angles,mean were lower in the medial trabecular bone region. Subjects with medial tibial BMLs had significantly higher FD_Ver and E_LBP as well as lower E_Lap and HI_Angles,mean in medial tibial subchondral bone. FD_Hor, FD_Ver, and E_LBP were higher, whereas E_Lap and HI_Angles,mean were lower in medial trabecular bone.

Conclusions

Our results support the use of bone structural analysis from radiographs when examining subjects with osteoarthritis or at risk of having it.

Key points

• Knee osteoarthritis causes changes in articular cartilage and subchondral bone

• Magnetic resonance imaging is a comprehensive imaging modality for knee osteoarthritis

• Radiography-based bone structure analysis can provide additional information of osteoarthritic subjects

Correlation of Subchondral Bone Density and Structure from Plain Radiographs with Micro Computed Tomography Ex Vivo

Osteoarthritis causes changes in the subchondral bone structure and composition. Plain radiography is a cheap, fast, and widely available imaging method. Bone tissue can be well seen from plain radiograph, which however is only a 2D projection of the actual 3D structure. Therefore, the aim was to investigate the relationship between bone density- and structure-related parameters from 2D plain radiograph and 3D bone parameters assessed from micro computed tomography (µCT) ex vivo. Right tibiae from eleven cadavers without any diagnosed joint disease were imaged using radiography and with µCT. Bone density- and structure-related parameters were calculated from four different locations from the radiographs of proximal tibia and compared with the volumetric bone microarchitecture from the corresponding regions. Bone density from the plain radiograph was significantly related with the bone volume fraction (r = 0.86; n = 44; p < 0.01). Mean homogeneity index for orientation of local binary patterns (HI_angle,mean) and fractal dimension of vertical structures (FD_Ver) were related (p < 0.01) with connectivity density (HI_angle,mean: r = −0.73, FD_Ver: r = 0.69) and trabecular separation (HI_angle,mean: r = 0.73, FD_Ver: r = −0.70) when all ROIs were pooled together (n = 44). Bone density and structure in tibia from standard clinically available 2D radiographs are significantly correlated with true 3D microstructure of bone.

Porosity imaged by a vector projection algorithm correlates with fractal dimension measured on 3D models obtained by microCT

Porosity is an important factor to consider in a large variety of materials. Porosity can be visualized in bone or 3D synthetic biomaterials by microcomputed tomography (microCT). Blocks of porous poly(2-hydroxyethyl methacrylate) were prepared with polystyrene beads of different diameter (500, 850, 1160 and 1560 μm) and analysed by microCT. On each 2D binarized microCT section, pixels of the pores which belong to the same image column received the same pseudo-colour according to a look up table. The same colour was applied on the same column of a frontal plane image which was constructed line by line from all images of the microCT stack. The fractal dimension Df of the frontal plane image was measured as well as the descriptors of the 3D models (porosity, 3D fractal dimension D3D, thickness, density and separation of material walls. Porosity, thickness Df and D3D increased with the size of the porogen beads. A linear correlation was observed between Df and D3D. This method provides quantitative and qualitative analysis of porosity on a single frontal plane image of a porous object.

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.

Three-dimensional solid texture analysis in biomedical imaging: review and opportunities

Three-dimensional computerized characterization of biomedical solid textures is key to large-scale and high-throughput screening of imaging data. Such data increasingly become available in the clinical and research environments with an ever increasing spatial resolution. In this text we exhaustively analyze the state-of-the-art in 3-D biomedical texture analysis to identify the specific needs of the application domains and extract promising trends in image processing algorithms. The geometrical properties of biomedical textures are studied both in their natural space and on digitized lattices. It is found that most of the tissue types have strong multi-scale directional properties, that are well captured by imaging protocols with high resolutions and spherical spatial transfer functions. The information modeled by the various image processing techniques is analyzed and visualized by displaying their 3-D texture primitives. We demonstrate that non-convolutional approaches are expected to provide best results when the size of structures are inferior to five voxels. For larger structures, it is shown that only multi-scale directional convolutional approaches that are non-separable allow for an unbiased modeling of 3-D biomedical textures. With the increase of high-resolution isotropic imaging protocols in clinical routine and research, these models are expected to best leverage the wealth of 3-D biomedical texture analysis in the future. Future research directions and opportunities are proposed to efficiently model personalized image-based phenotypes of normal biomedical tissue and its alterations. The integration of the clinical and genomic context is expected to better explain the intra class variation of healthy biomedical textures. Using texture synthesis, this provides the exciting opportunity to simulate and visualize texture atlases of normal ageing process and disease progression for enhanced treatment planning and clinical care management.