A new method to determine cortical bone thickness in CT images using a hybrid approach of parametric profile representation and local adaptive thresholds: Accuracy results

Abstract

MotivationCortical bone is an important contributor to bone strength and is pivotal to understand the etiology of osteoporotic fractures and the specific mechanisms of antiosteoporotic treatment regimen. 3D computed tomography (CT) can be used to measure cortical thickness, density, and mass in the proximal femur, lumbar vertebrae, and distal forearm. However, the spatial resolution of clinical whole body CT scanners is limited by radiation exposure; partial volume artefacts severely impair the accurate assessment of cortical parameters, in particular in locations where the cortex is thin such as in the lumbar vertebral bodies or in the femoral neck.MethodModel-based deconvolution approaches recover the cortical thickness by numerically deconvolving the image along 1D profiles using an estimated scanner point spread function (PSF) and a hypothesized uniform cortical bone mineral density (reference density). In this work we provide a new essentially analytical unique solution to the model-based cortex recovery problem using few characteristics of the measured profile and thus eliminate the non-linear optimization step for deconvolution. Also, the proposed approach allows to get rid of the PSF in the model and reduces sensitivity to errors in the reference density. Additionally, run-time and memory effective computation of cortical thickness was achieved with the help of a lookup table.ResultsThe method accuracy and robustness was validated and compared to that of a deconvolution approach recently proposed for cortical bone and of the 50% relative threshold technique: in a simulated environment with noise and various error levels in the reference density and using CT acquisitions of the European Forearm Phantom (EFP II), a modification of a widely used anthropomorphic standard of cortical and trabecular bone compartments that was scanned with various scan protocols.ConclusionResults of simulations and of phantom data analysis verified the following properties of the new method: 1) Robustness against errors in the reference density. 2) Excellent accuracy on ground truth data with various noise levels. 3) Very fast computation using a lookup table.

Associations of Muscle Size and Density With Proximal Femur Bone in a Community Dwelling Older Population

“…The Medical Image Analysis Framework for the Femur (MIAF Femur Version 7.1.0MRH) was used to measure three-dimensional bone parameters of the proximal femur ( 14 ). An important advantage of this software is that the three dimensional bone segmentation is combining global and adaptive local thresholds with volume growing and morphological operations ( 16 , 17 ). Both surfaces are displayed in axial, sagittal, and coronal reformation for operator control and manual editing if necessary.…”

Cortical bone mapping improves finite element strain prediction accuracy at the proximal femur

“…Other algorithms that produce validated subject specific estimates of cortical thickness have been recently published and may be candidate to be included in FE models. However, regardless of the implementation (from image deconvolution approaches similar in concept to CBM [62,63], to statistical models that swiftly produce FEfriendly surfaces [64]) they do not provide local cortical density estimates, as they all assume for density a fully mineralised condition. Knowledge of cortical thickness can be enough to improve FE modelling of the overall bone structural behaviour and simulated stress, but local strains are highly influenced by the elastic modulus, in turn related to density estimates.…”

Focal osteoporosis defect is associated with vertebral compression fracture prevalence in a bone mineral density‐independent manner

“…Following intensity profile sampling, the full width at half‐maximum measurement strategy 20 , 21 was adopted to identify the apparent bone void space. Briefly, the ridges and valleys in the intensity profile were located by finding the localized maxima and minima, the localized region with intensity less than 0.5 × (maxima + minima) was defined as the bone void space.…”