Longitudinal bone microarchitectural changes are best detected using image registration

A 2D-registration algorithm for the correction of motion-induced misalignments of consecutive image stacks in multi-stack high-resolution peripheral quantitative CT scans

Multi-stack imaging using high-resolution peripheral quantitative CT (HR-pQCT) can involve misalignments of consecutive image stacks ('stack shift') due to subject movement during scan acquisition. We developed a simple, 2D-registration algorithm for the correction of stack shifts in multi-stack HR-pQCT scans and investigated 1) the differences in standard HR-pQCT parameters and repeatability between before and after stack-shift correction; and 2) the correlation between the transformation needed for the stack-shift correction and corresponding difference in HR-pQCT parameters. The algorithm generates an artificial stack overlap of two slices, then rigidly registers the overlapping region (only in-plane translation allowed), and subsequently applies the resulting translation to the proximal stack. The algorithm was applied to data of 23 men and women with three same-day repeated scans (69 radius and 63 tibia scans, Dataset 1) and of 48 postmenopausal women with 78 radius scans taken at two time points with 12-week interval (Dataset 2). In both datasets, median differences in HR-pQCT parameters between before and after stack-shift correction were mostly significant yet small (≤0.53 %). The differences could vary considerably between subjects and ranged between -12.1 % and +35.8 % for cortical porosity, stiffness, and failure load. For the other HR-pQCT parameters, the differences ranged between ±0.8 % (Dataset 1) and between -4.5 % and +0.9 % (Dataset 2) among subjects. Spearman correlations between the magnitude of the translation and corresponding difference in HR-pQCT parameters were significant for most parameters in both datasets and strongest for stiffness and failure load (ρ = 0.687-0.947; p < 0.01). Based on Dataset 1, coefficients of variation differed between ±0.3 percentage points after stack-shift correction as compared to before. To conclude, correction of stack misalignments in two-stack HR-pQCT scans using our algorithm resulted in significant but negligible median differences in HR-pQCT parameters and precision, but differences could exceed least-significant differences and thereby be clinically relevant in individual subjects. The translation needed for the stack-shift correction correlated significantly with the difference in most HR-pQCT parameters, thereby potentially serving as objective measure for stack-shift severity. The algorithm can be applied directly after scan reconstruction, at low computational cost and without negative effects from image interpolation.

Bone-wise rigid registration of femur, tibia, and fibula for the tracking of temporal changes

BACKGROUND

Multiple myeloma (MM) induces temporal alterations in bone structure, such as osteolytic bone lesions, which are challenging to identify through manual image interpretation. The large variation in radiologists' assessments, even at expert centers, further complicates diagnosis. Automatic image analysis methods, including segmentation and registration, can expedite detecting and tracking these bone changes.

PURPOSE

This study presents an automated pipeline for accurately tracking temporal changes in the femurs, tibiae, and fibulae of MM patients using 3D whole-body CT images. The pipeline leverages image segmentation, rigid registration, and temporal subtraction to accelerate disease monitoring and support clinical decision-making.

METHODS

A convolutional neural network (CNN) was trained to segment bones in 3D CT images of 30 MM patients. Nine patients with pre- and post-diagnosis CT scans were used to validate the segmentation and registration process. A two-phase bone-wise rigid registration was applied, followed by temporal subtraction to generate difference images. Segmentation and registration accuracy were assessed using the Dice similarity coefficient (DSC) and mean surface distance (MSD). The proposed method was compared to a non-rigid registration method.

RESULTS

The neural network segmentation resulted in a mean DSC of 0.93 across all bone types and all test data. The registration accuracy measured by the mean DSC across the test data was at least 0.94 for all bone types. The second phase of rigid registration improved the registration fibulae. Metric-wise, the nonrigid method performed better but diminished lesion visibility in difference images.

CONCLUSIONS

An automated pipeline for the longitudinal tracking of bone alterations was presented. Both segmentation and registration demonstrated high accuracy as measured by DSC and MSD. In the difference images produced by temporal subtraction, osteolytic lesions were clearly visible in the femurs. The methodology lays a solid foundation for future improvements, such as inclusion of the axial spine.

Strength training for osteoporosis prevention during early menopause (STOP-EM): a pilot study protocol for a single centre randomised waitlisted control trial in Canada

INTRODUCTION

Women lose up to 10% of their bone mass around menopause and the decade following. There is a need for proactive approaches to preserve bone mass and quality around menopause. Existing work has found that high-intensity resistance and impact training (HiRIT) can improve bone and muscle measures in late postmenopausal women. However, this has not been investigated in perimenopausal and early postmenopausal women who are in the midst of the menopausal transition.

METHODS AND ANALYSIS

This study is a 9-month randomised controlled feasibility trial evaluating a HiRIT programme in perimenopausal and early postmenopausal women. The primary objective of this study is to determine the feasibility of HiRIT in 40 perimenopausal and early postmenopausal women (45-60 years). Participants will be randomised 1:1 into a supervised HiRIT exercise intervention and waitlisted control. The primary outcomes are recruitment, retention and adherence to the exercise intervention. Secondary outcomes include bone (bone mineral density, microarchitecture and strength), muscle (mass, strength and power), physical function (balance and aerobic fitness) and quality of life measures. Feasibility will be assessed based on a priori criterion for success and secondary outcomes will be assessed via multiple linear regressions. The study will be considered feasible if>50% of interested and eligible participants are recruited, if there is>60% adherence to the two times per week, 9-month exercise intervention and if at least 65% of the sample complete the final study visit. Feasibility outcomes will be used to inform a larger, future trial aimed at identifying the efficacy of the exercise intervention for improving various health outcomes, including bone density and muscle mass. Secondary exploratory outcomes will provide insight into the effect of exercise on muscle and bone in perimenopausal and early postmenopausal women.

ETHICS AND DISSEMINATION

This study has been approved by the Conjoint Health Research Ethics Board of the University of Calgary REB22-1632. The results of this study will be disseminated at national and international conferences and published in academic journals.

TRAIL REGISTRATION NUMBER

ClinicalTrials.gov: NCT05961371. (Protocol V.1.2, 28 September 2023).

Bone imaging modality precision and agreement between DXA, pQCT, and HR-pQCT

Quantifying precision error for DXA, peripheral QCT (pQCT), and HR-pQCT is crucial for monitoring longitudinal changes in body composition and musculoskeletal outcomes. Agreement and associations between bone variables assessed using pQCT and second-generation HR-pQCT are unclear. This study aimed to determine the precision of, and agreement and associations between, bone variables assessed via DXA, pQCT, and second-generation HR-pQCT. Thirty older adults (mean age: 64.2 ± 8.0 yr; women: 67%) were recruited. DXA scans were performed at the total hip, lumbar spine, and whole body. Distal (4%) and proximal (30%/33%/66%) skeletal sites at the radius and tibia were scanned with pQCT and/or HR-pQCT. Root-mean-squared coefficients of variation (%CVRMS) were calculated to define precision errors, and Bland-Altman plots assessed agreement between densitometric estimates. Pearson correlations and linear regression explored relationships between bone variables at different skeletal sites and proportional bias, respectively. Precision errors ranged between 0.55% and 1.6% for DXA, 0.40% and 4.8% for pQCT, and 0.13% and 30.7% for HR-pQCT. Systematic bias was identified between pQCT- and HR-pQCT-determined radius and tibia volumetric BMD (vBMD) estimates (all p<.001). Proportional bias was not observed between vBMD measures at any skeletal site (all p>.05). pQCT- and HR-pQCT-determined total, trabecular, and cortical vBMD and estimates of bone strength at the radius and tibia were strongly correlated (all p<.05). Precision error was low for most bone variables and within the expected range for all imaging modalities. We observed significant systematic bias, but no proportional bias, between pQCT- and second-generation HR-pQCT-determined vBMD estimates at the radius and tibia. Nevertheless, measures of bone density and strength were strongly correlated at all skeletal sites. These findings suggest that although bone density and strength estimates from both imaging modalities are not interchangeable, they are strongly related and likely have similar fracture prediction capabilities.

PPI use is not associated with bone microarchitecture and strength assessed with HR-pQCT after three-years follow-up in patients visiting the Fracture Liaison Service

BACKGROUND

The use of proton pump inhibitors (PPIs) has been associated with an increased fracture risk in observational studies. However, the reported association between PPI use and bone mineral density (BMD), bone microarchitecture, and bone strength is inconsistent. This study aims to assess the association between PPI use and bone microarchitecture and strength using high-resolution peripheral quantitative CT (HR-pQCT) in a three-year follow-up study in patients with a recent fracture visiting the Fracture Liaison Service (FLS).

METHODS

This three-year prospective cohort study included FLS patients aged ≥ 50 years with a recent fracture (median age 62 [IQR 56-69] years, 68.7 % females) and without anti-osteoporosis treatment indication. HR-pQCT scans (distal radius and tibia) were obtained at baseline (T0) and three-year follow-up (T3). Volumetric bone mineral density and bone area, microarchitecture, and strength (micro-finite element analysis) were determined. The association between three-year continuous PPI use and the percentage change in HR-pQCT parameters between T0 and T3 was assessed using sex-stratified multivariate linear regression analyses. Covariates included age, BMI, vitamin-D deficiency (< 50 nmol/l), glucocorticoid use, and cardiovascular co-morbidity (males and females) fracture type (major/hip vs. all others, only males) and probable sarcopenia (only females).

RESULTS

In total, 282 participants had available medication data throughout follow-up, of whom 20.6 % were continuous PPI users. In both males and females with complete HR-pQCT follow-up data (males: N = 69 radius, N = 84 tibia; females: N = 147 radius, N = 168 tibia), PPI use was not associated with the percentage change of any of the bone microarchitecture or strength parameters between T0 and T3 at the radius and tibia as compared to non-use.

CONCLUSION

Compared to non-use, PPI use was not associated with the change of bone microarchitecture and strength in FLS patients at three years of follow-up. These results do not support that an altered bone microarchitecture or strength may contribute to the increased fracture risk associated with PPI use, as reported in observational studies.

Modelling the Effects of Growth and Remodelling on the Density and Structure of Cancellous Bone

A two-stage model is proposed for investigating remodelling characteristics in bone over time and distance to the growth plate. The first stage comprises a partial differential equation (PDE) for bone density as a function of time and distance from the growth plate. This stage clarifies the contributions to changes in bone density due to remodelling and growth processes and tracks the rate at which new bone emanates from the growth plate. The second stage consists of simulating the remodelling process to determine remodelling characteristics. Implementing the second stage requires the rate at which bone moves away from the growth plate computed during the first stage. The second stage is also needed to confirm that remodelling characteristics predicted by the first stage may be explained by a realistic model for remodelling and to compute activation frequency. The model is demonstrated on microCT scans of tibia of juvenile female rats in three experimental groups: sham-operated control, oestrogen deprived, and oestrogen deprived followed by treatment. Model predictions for changes in bone density and remodelling characteristics agree with the literature. In addition, the model provides new insight into the role of treatment on the density of new bone emanating from the growth plate and provides quantitative descriptions of changes in remodelling characteristics beyond what has been possible to ascertain by experimentation alone.

Predicting Bone Adaptation in Astronauts during and after Spaceflight

A method was previously developed to identify participant-specific parameters in a model of trabecular bone adaptation from longitudinal computed tomography (CT) imaging. In this study, we use these numerical methods to estimate changes in astronaut bone health during the distinct phases of spaceflight and recovery on Earth. Astronauts (N = 16) received high-resolution peripheral quantitative CT (HR-pQCT) scans of their distal tibia prior to launch (L), upon their return from an approximately six-month stay on the international space station (R+0), and after six (R+6) and 12 (R+12) months of recovery. To model trabecular bone adaptation, we determined participant-specific parameters at each time interval and estimated their bone structure at R+0, R+6, and R+12. To assess the fit of our model to this population, we compared static and dynamic bone morphometry as well as the Dice coefficient and symmetric distance at each measurement. In general, modeled and observed static morphometry were highly correlated (R2> 0.94) and statistically different (p < 0.0001) but with errors close to HR-pQCT precision limits. Dynamic morphometry, which captures rates of bone adaptation, was poorly estimated by our model (p < 0.0001). The Dice coefficient and symmetric distance indicated a reasonable local fit between observed and predicted bone volumes. This work applies a general and versatile computational framework to test bone adaptation models. Future work can explore and test increasingly sophisticated models (e.g., those including load or physiological factors) on a participant-specific basis.

Recommendations for High-resolution Peripheral Quantitative Computed Tomography Assessment of Bone Density, Microarchitecture, and Strength in Pediatric Populations

PURPOSE OF REVIEW

The purpose of this review is to summarize current approaches and provide recommendations for imaging bone in pediatric populations using high-resolution peripheral quantitative computed tomography (HR-pQCT).

RECENT FINDINGS

Imaging the growing skeleton is challenging and HR-pQCT protocols are not standardized across centers. Adopting a single-imaging protocol for all studies is unrealistic; thus, we present three established protocols for HR-pQCT imaging in children and adolescents and share advantages and disadvantages of each. Limiting protocol variation will enhance the uniformity of results and increase our ability to compare study results between different research groups. We outline special cases along with tips and tricks for acquiring and processing scans to minimize motion artifacts and account for growing bone. The recommendations in this review are intended to help researchers perform HR-pQCT imaging in pediatric populations and extend our collective knowledge of bone structure, architecture, and strength during the growing years.

Image Registration in Longitudinal Bone Assessment Using Computed Tomography

PURPOSE OF REVIEW

Rigid image registration is an important image processing tool for the assessment of musculoskeletal chronic disease. In this paper, we critically review applications of rigid image registration in terms of similarity measurement methods over the past three years (2019-2022) in the context of monitoring longitudinal changes to bone microstructure and mechanical properties using computed tomography. This review identifies critical assumptions and trade-offs underlying different similarity measurement methods used in image registration and demonstrates the effect of using different similarity measures on registration outcomes.

RECENT FINDINGS

Image registration has been used in recent studies for: correcting positional shifts between longitudinal scans to quantify changes to bone microstructural and mechanical properties over time, developing registration-based workflows for longitudinal assessment of bone properties in pre-clinical and clinical studies, and developing and validating registration techniques for longitudinal studies. In evaluating the recent literature, it was found that the assumptions at the root of different similarity measures used in rigid image registration are not always confirmed and reported. Each similarity measurement has its advantages and disadvantages, as well as underlying assumptions. Breaking these assumptions can lead to poor and inaccurate registration results. Thus, care must be taken with regards to the choice of similarity measurement and interpretation of results. We propose that understanding and verifying the assumptions of similarity measurements will enable more accurate and efficient quantitative assessments of structural changes over time.

Bone Microarchitecture and Strength Changes During Teriparatide and Zoledronic Acid Treatment in a Patient with Pregnancy and Lactation-Associated Osteoporosis with Multiple Vertebral Fractures

Pregnancy- and lactation-associated osteoporosis (PLO) is a rare form of osteoporosis, of which the pathogenesis and best treatment options are unclear. In this report, we describe the case of a 34-year old woman diagnosed with severe osteoporosis and multiple vertebral fractures after her first pregnancy, who was subsequently treated with teriparatide (TPTD) and zoledronic acid (ZA). We describe the clinical features, imaging examination, and genetic analysis. Substantial improvements were observed in areal and volumetric bone mineral density (BMD), microarchitecture, and strength between 7 and 40 months postpartum as assessed by dual-energy X-ray absorptiometry at the total hip and spine and by high-resolution peripheral quantitative CT at the distal radius and tibiae. At the hip, spine, and distal radius, these improvements were mainly enabled by treatment with TPTD and ZA, while at the distal tibiae, physiological recovery and postpartum physiotherapy due to leg pain after stumbling may have played a major role. Additionally, the findings show that, despite the improvements, BMD, microarchitecture, and strength remained severely impaired in comparison with healthy age- and gender-matched controls at 40 months postpartum. Genetic analysis showed no monogenic cause for osteoporosis, and it is suggested that PLO in this woman could have a polygenic origin with possible susceptibility based on familiar occurrence of osteoporosis.

Learning-based three-dimensional registration with weak bounding box supervision

Purpose: Image registration is the process of aligning images, and it is a fundamental task in medical image analysis. While many tasks in the field of image analysis, such as image segmentation, are handled almost entirely with deep learning and exceed the accuracy of conventional algorithms, currently available deformable image registration methods are often still conventional. Deep learning methods for medical image registration have recently reached the accuracy of conventional algorithms. However, they are often based on a weakly supervised learning scheme using multilabel image segmentations during training. The creation of such detailed annotations is very time-consuming. Approach: We propose a weakly supervised learning scheme for deformable image registration. By calculating the loss function based on only bounding box labels, we are able to train an image registration network for large displacement deformations without using densely labeled images. We evaluate our model on interpatient three-dimensional abdominal CT and MRI images. Results: The results show an improvement of ∼ 10 % (for CT images) and 20% (for MRI images) in comparison to the unsupervised method. When taking into account the reduced annotation effort, the performance also exceeds the performance of weakly supervised training using detailed image segmentations. Conclusion: We show that the performance of image registration methods can be enhanced with little annotation effort using our proposed method.

Incomplete recovery of bone strength and trabecular microarchitecture at the distal tibia 1 year after return from long duration spaceflight

Determining the extent of bone recovery after prolonged spaceflight is important for understanding risks to astronaut long-term skeletal health. We examined bone strength, density, and microarchitecture in seventeen astronauts (14 males; mean 47 years) using high-resolution peripheral quantitative computed tomography (HR-pQCT; 61 μm). We imaged the tibia and radius before spaceflight, at return to Earth, and after 6- and 12-months recovery and assessed biomarkers of bone turnover and exercise. Twelve months after flight, group median tibia bone strength (F.Load), total, cortical, and trabecular bone mineral density (BMD), trabecular bone volume fraction and thickness remained − 0.9% to − 2.1% reduced compared with pre-flight (p ≤ 0.001). Astronauts on longer missions (> 6-months) had poorer bone recovery. For example, F.Load recovered by 12-months post-flight in astronauts on shorter (< 6-months; − 0.4% median deficit) but not longer (− 3.9%) missions. Similar disparities were noted for total, trabecular, and cortical BMD. Altogether, nine of 17 astronauts did not fully recover tibia total BMD after 12-months. Astronauts with incomplete recovery had higher biomarkers of bone turnover compared with astronauts whose bone recovered. Study findings suggest incomplete recovery of bone strength, density, and trabecular microarchitecture at the weight-bearing tibia, commensurate with a decade or more of terrestrial age-related bone loss.

3D Image Registration Marginally Improves the Precision of HR-pQCT Measurements Compared to Cross-Sectional-Area Registration in Adults With Osteogenesis Imperfecta

Repositioning error in longitudinal high-resolution peripheral-quantitative computed tomography (HR-pQCT) imaging can lead to different bone volumes being assessed over time. To identify the same bone volumes at each time point, image registration is used. While cross-sectional area image registration corrects axial misalignment, 3D registration additionally corrects rotations. Other registration methods involving matched angle analysis (MA) or boundary transformations (3D-TB) can be used to limit interpolation error in 3D-registering micro-finite-element data. We investigated the effect of different image registration methods on short-term in vivo precision in adults with osteogenesis imperfecta, a collagen-related genetic disorder resulting in low bone mass, impaired quality, and increased fragility. The radii and tibiae of 29 participants were imaged twice on the same day with full repositioning. We compared the precision error of different image registration methods for density, microstructural, and micro-finite-element outcomes with data stratified based on anatomical site, motion status, and scanner generation. Regardless of the stratification, we found that image registration improved precision for total and trabecular bone mineral densities, trabecular and cortical bone mineral contents, area measurements, trabecular bone volume fraction, separation, and heterogeneity, as well as cortical thickness and perimeter. 3D registration marginally outperformed cross-sectional area registration for some outcomes, such as trabecular bone volume fraction and separation. Similarly, precision of micro-finite-element outcomes was improved after image registration, with 3D-TB and MA methods providing greatest improvements. Our regression model confirmed the beneficial effect of image registration on HR-pQCT precision errors, whereas motion had a detrimental effect on precision even after image registration. Collectively, our results indicate that 3D registration is recommended for longitudinal HR-pQCT imaging in adults with osteogenesis imperfecta. Since our precision errors are similar to those of healthy adults, these results can likely be extended to other populations, although future studies are needed to confirm this. © 2022 American Society for Bone and Mineral Research (ASBMR).

Response to High-Dose Vitamin D Supplementation Is Specific to Imaging Modality and Skeletal Site

High-dose vitamin D supplementation (4000 or 10,000 IU/d) in vitamin D-sufficient individuals results in a dose-dependent decrease in radius and tibia total bone mineral density (Tt.BMD) compared with 400 IU/d. This exploratory analysis examined whether the response to high-dose vitamin D supplementation depends on imaging modality and skeletal site. Participants were aged 55 to 70 years, not osteoporotic, with serum 25(OH)D 30 to 125 nM. Participants' radius and tibia were scanned on high-resolution peripheral quantitative computed tomography (HR-pQCT) to measure Tt.BMD, trabecular bone volume fraction (Tb.BV/TV), trabecular separation (Tb.Sp), cortical thickness (Ct.Th), and finite element analysis (FEA) estimated failure load. Three-dimensional image registration was used. Dual-energy X-ray absorptiometry (DXA) scans of the hip, spine, and radius measured areal BMD (aBMD) and trabecular bone score (TBS). Constrained linear mixed-effects models determined treatment group-by-time and treatment group-by-time-by-sex interactions. The treatment group-by-time interaction previously observed for radial Tt.BMD was observed at both ultradistal (UD, p < 0.001) and 33% (p < 0.001) aBMD sites. However, the treatment group-by-time-by-sex interaction observed for radial Tt.BMD was not observed with aBMD at either the UD or 33% site, and the 4000 and 400 groups did not differ. Registered radial FEA results mirrored Tt.BMD. An increase in Tb.Sp and decrease in Ct.Th underpinned dose-dependent changes in radial BMD and strength. We observed no effects in DXA-based aBMD at the hip or spine or TBS. At the tibia, we observed a time-by-treatment group effect for Tb.BV/TV. Given that DXA measures at the radius did not detect sex differences or differences between the 4000 and 400 groups, HR-pQCT at the radius may be more sensitive for examining bone changes after vitamin D supplementation. Although DXA did not reveal treatment effects at the hip or spine, whether that is a true skeletal site difference or a lack of modality sensitivity remains unclear. © 2022 The Authors. JBMR Plus published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research.

An inverse technique to identify participant-specific bone adaptation from serial CT measurements

Simulated bone adaptation is framed as an interface evolution problem. The interface is extracted from a high-resolution computed tomography (CT) image of trabecular bone microarchitecture and modified by the level set equation. A model and its parameters determine the bone adaptation rate and thus the bone structure at any future time. This study develops an inverse problem and solver to identify model parameters from multiple high-resolution CT images of bone within the level set framework. We demonstrate the technique on a model of advection and mean curvature flow, termed curvature-driven adaptation. The inverse solver uses two CT scans to estimate model parameters, which map the bone surface from one image to the next. The solver was tested with synthetic images of bone changing according to the curvature-driven model with known model parameters. The algorithm recovered known model parameters excellently (R²  > .99, p < .001). A grid search indicated that the estimated model parameters were insensitive to hyper-parameter selection for learning rate 1e^-5≤η≤ 5e^-5 and gradient scaling factor 5e^-5≤γ≤ 5e^-4 . Finally, we tested the algorithm's sensitivity to salt-and-pepper noise of probability P , where .0 ≤P≤ .9. Model parameter accuracy did not change for P < .7, corresponding to Dice coefficients greater than .7. The inverse problem estimates bone adaptation parameters from multiple CT images of changing bone microarchitecture. In the future, this technique could be used to determine participant-specific bone adaptation parameters in vivo, validate bone adaptation models, and predict bone health.

Using 3D image registration to maximize the reproducibility of longitudinal bone strength assessment by HR-pQCT and finite element analysis

We developed and validated a finite element (FE) approach for longitudinal high-resolution peripheral quantitative computed tomography (HR-pQCT) studies using 3D image registration to account for misalignment between images. This reduced variability in longitudinal FE estimates and improved our ability to measure in vivo changes in HR-pQCT studies of bone strength.

INTRODUCTION

We developed and validated a finite element (FE) approach for longitudinal high-resolution peripheral quantitative computed tomography (HR-pQCT) studies using 3D rigid-body registration (3DR) to maximize reproducibility by accounting for misalignment between images.

METHODS

In our proposed approach, we used the full common bone volume defined by 3DR to estimate standard FE parameters. Using standard HR-pQCT imaging protocols, we validated the 3DR approach with ex vivo samples of the distal radius (n = 10, four repeat scans) by assessing whether 3DR can reduce measurement variability from repositioning error. We used in vivo data (n = 40, five longitudinal scans) to assess the sensitivity of 3DR to detect changes in bone strength at the distal radius by the standard deviation of the rate of change (σ), where the ideal value of σ is minimized to define true change. FE estimates by 3DR were compared to estimates by no registration (NR) and slice-matching (SM).

RESULTS

Group-wise comparisons of ex vivo variation (CV_RMS, %) found that FE measurement precision was improved by SM (CV_RMS < 0.80%) and 3DR (CV_RMS < 0.62%) compared to NR (CV_RMS~2%), and 3DR was advantageous as repositioning error increased. Longitudinal in vivo reproducibility was minimized by 3DR for failure load estimates (σ = 0.008 kN/month).

CONCLUSION

Although 3D registration cannot negate motion artifacts, it plays an important role in detecting and reducing variability in FE estimates for longitudinal HR-pQCT data and is well suited for estimating effects of interventions in in vivo longitudinal studies of bone strength.

A Novel HR-pQCT Image Registration Approach Reveals Sex-Specific Changes in Cortical Bone Retraction With Aging

During aging, changes in endosteal and periosteal boundaries of cortical bone occur that differ between men and women. We here develop a new procedure that uses high-resolution peripheral quantitative CT (HR-pQCT) imaging and 3D registration to identify such changes within the timescale of longitudinal studies. A first goal was to test the sensitivity of the approach. A second goal was to assess differences in periosteal/endosteal expansion over time between men and women. Rigid 3D registration was used to transform baseline and all follow-up (FU) images to a common reference configuration for which the region consisting of complete slices (largest common height) was determined. Periosteal and endosteal contours were transformed to the reference position to determine the net periosteal and endosteal expansion distances. To test the sensitivity, images from a short-term reproducibility study were used (15 female, aged 21 to 47 years, scanned three times). To test differences between men and women, images from a subset of the Geneva Retirees Cohort were used (248 female, 61 male, average age 65 years, 3.5 and 7 years FU). The sensitivity study indicated a least significant change for detecting periosteal/endosteal expansion of 41/31 microns for the radius and 17/26 microns for the tibia. Results of the cohort study showed significant net endosteal retraction only in females at the radius and tibia after 3.5 years (38.0 and 38.4 microns, respectively) that further increased at 7 years FU (70.4 and 70.8 microns, respectively). No significant net periosteal changes were found for males or females at 7 years. The results demonstrate that it is possible to measure changes in endosteal contours in longitudinal studies within several years. For the investigated cohort, significant endosteal retraction was found in females but not in males. Whether these changes in cortical geometry are related to fracture risk remains to be investigated in larger cohorts © 2021 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Multisite longitudinal calibration of HR-pQCT scanners and precision in osteogenesis imperfecta

BACKGROUND

For high-resolution peripheral quantitative computed tomography (HR-pQCT) to be used in longitudinal multi-center studies to assess disease and treatment effects, data must be aggregated across multiple timepoints and scanners. This requires an understanding of the factors contributing to scanner precision, and multi-scanner cross-calibration procedures, especially for clinical populations with severe phenotypes, like osteogenesis imperfecta (OI).

METHODS

To address this, we first evaluated single- and multi-center short- and long-term precision errors of standard HR-pQCT parameters. Two imaging phantoms were circulated among 13 sites (7 XtremeCT and 6 XtremeCT2) and scanned in triplicate at 3 timepoints/site. Additionally, duplicate in vivo radial and tibial scans were acquired in 29 individuals with OI. Secondly, we investigated subject- and scanner-related factors that contribute to precision errors using regression analysis. Thirdly, we proposed a reference site selection criterion for multisite cross-calibration and demonstrated the external validity of phantom-based calibrations.

RESULTS

Our results show excellent short-term single-site precision in both phantoms (CV % < 0.5%) and in density, microarchitecture and finite element parameters of OI participants (CV % = 0.75 to 1.2%). In vivo reproducibility significantly improved with (i) cross sectional area image registration versus no registration and (ii) scans with no motion artifacts. While reproducibility was similar across OI subtypes and anatomical sites, XtremeCT2 scanners achieved ~2.5% better precision than XtremeCT for trabecular parameters. Finally, we demonstrate that multisite longitudinal precision errors resulting from inconsistencies between scanners can be partially corrected through scanner cross-calibration.

CONCLUSIONS

This study is the first to assess long-term reproducibility and cross-calibration in a study using first and second generation HR-pQCT scanners. The results presented in this context provide timely guidelines for future use of this powerful clinical imaging modality in multi-center longitudinal clinical trials.

Formation Dominates Resorption With Increasing Mineralized Density and Time Postfracture in Cortical but Not Trabecular Bone: A Longitudinal HRpQCT Imaging Study in the Distal Radius

Clinical evaluation of fracture healing is often limited to an assessment of fracture bridging from radiographic images, without consideration for other aspects of bone quality. However, recent advances in HRpQCT offer methods to accurately monitor microstructural bone remodeling throughout the healing process. In this study, local bone formation and resorption were investigated during the first year post fracture in both the fractured (n = 22) and contralateral (n = 19) radii of 34 conservatively treated patients (24 female, 10 male) who presented with a unilateral radius fracture at the Innsbruck University Hospital, Austria. HRpQCT images and clinical metrics were acquired at six time points for each patient. The standard HRpQCT image acquisition was captured for all radii, with additional distal and proximal image acquisitions for the fractured radii. Measured radial bone densities were isolated with a voxel‐based mask and images were rigidly registered to images from the previous imaging session using a pyramid‐based approach. From the registered images, bone formation and resorption volume fractions were quantified for multiple density‐based thresholds and compared between the fractured and contralateral radius and relative to demographics, bone morphometrics, and fracture metrics using regression. Compared with the contralateral radius, both bone formation and resorption were significantly increased in the fractured radius throughout the study for nearly all evaluated thresholds. Higher density cortical bone formation continually increased throughout the duration of the study and was significantly greater than resorption during late‐stage healing in both the fractured and intact regions of the radius. With the small and diverse study population, only weak relationships between fracture remodeling and patient‐specific parameters were unveiled. However this study provides methods for the analysis of local bone remodeling during fracture healing and highlights relevant considerations for future studies, specifically that remodeling postfracture is likely to continue beyond 12‐months postfracture. © 2021 The Authors. JBMR Plus published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research.

Precision of 3D Registration Analysis for Longitudinal Study of Second-Generation HR-pQCT

OBJECTIVE

The objective of this research was to develop 3D registration analysis method in longitudinal studies of high-resolution peripheral quantitative computed tomography (HR-pQCT), to analyze ranges of bone microstructure parameters in addition to standard parameters, and to test the precision of these measurements.

METHODS

Scans of HR-pQCT and analysis of bone microstructure were performed at 3 times in 15 subjects. The 3 images were matched 3-dimensionally, and bone microstructures were analyzed in the common region. In addition to standard measurement parameters of geometry, bone mineral density (BMD), trabecular bone, and cortical bone, parameters showing plate to rod-like structure, connectivity, cavity formation of trabecular bone, and bending stability of cortical bone were also measured. Precision was evaluated with the root mean square percent coefficient variance (RMS%CV).

RESULTS

RMS%CV was 0.1%-1.3% for geometry, 0.6%-1.9% for BMD, 0.8%-3.3% for trabecular bone, 2.1%-9.8% for additionally measured trabecular bone, 1.0%-3.4% for cortical bone excluding Ct.Po, 6.0%-6.1% for Ct.Po, and 0.8%-1.5% for additionally measured cortical bone. Precision was higher for 3D registration than for 2D registration in geometry, BV/TV, and Ct.Po.

CONCLUSIONS

3D registration analysis of a range of bone microstructural parameters in longitudinal analysis of HR-pQCT showed good precision, offering potential for contributing to future research on osteoporosis and bone metabolic diseases.

Pre-flight exercise and bone metabolism predict unloading-induced bone loss due to spaceflight

Objectives

Bone loss remains a primary health concern for astronauts, despite in-flight exercise. We examined changes in bone microarchitecture, density and strength before and after long-duration spaceflight in relation to biochemical markers of bone turnover and exercise.

Methods

Seventeen astronauts had their distal tibiae and radii imaged before and after space missions to the International Space Station using high-resolution peripheral quantitative CT. We estimated bone strength using finite element analysis and acquired blood and urine biochemical markers of bone turnover before, during and after spaceflight. Pre-flight exercise history and in-flight exercise logs were obtained. Mixed effects models examined changes in bone and biochemical variables and their relationship with mission duration and exercise.

Results

At the distal tibia, median cumulative losses after spaceflight were −2.9% to −4.3% for bone strength and total volumetric bone mineral density (vBMD) and −0.8% to −2.6% for trabecular vBMD, bone volume fraction, thickness and cortical vBMD. Mission duration (range 3.5–7 months) significantly predicted bone loss and crewmembers with higher concentrations of biomarkers of bone turnover before spaceflight experienced greater losses in tibia bone strength and density. Lower body resistance training volume (repetitions per week) increased 3–6 times in-flight compared with pre-spaceflight. Increases in training volume predicted preservation of tibia bone strength and trabecular vBMD and thickness.

Conclusions

Findings highlight the fundamental relationship between mission duration and bone loss. Pre-flight markers of bone turnover and exercise history may identify crewmembers at greatest risk of bone loss due to unloading and may focus preventative measures.