Robust quantitative assessment of trabecular microarchitecture in extremity cone-beam CT using optimized segmentation algorithms

Abstract

Purpose: In-vivo evaluation of bone microarchitecture remains challenging because of limited resolution of conventional orthopaedic imaging modalities. We investigate the performance of flat-panel detector extremity Cone-Beam CT (CBCT) in quantitative analysis of trabecular bone. To enable accurate morphometry of fine trabecular bone architecture, advanced CBCT pre-processing and segmentation algorithms are developed. Methods: The study involved 35 transilliac bone biopsy samples imaged on extremity CBCT (voxel size 75 μm, imaging dose ~13 mGy) and gold standard μCT (voxel size 7.67 μm). CBCT image segmentation was performed using (i) global Otsu’s thresholding, (ii) Bernsen’s local thresholding, (iii) Bernsen’s local thresholding with additional histogram-based global pre-thresholding, and (iv) the same as (iii) but combined with contrast enhancement using a Laplacian Pyramid. Correlations between extremity CBCT with the different segmentation algorithms and gold standard μCT were investigated for measurements of Bone Volume over Total Volume (BV/TV), Trabecular Thickness (Tb.Th), Trabecular Spacing (Tb.Sp), and Trabecular Number (Tb.N). Results: The combination of local thresholding with global pre-thresholding and Laplacian contrast enhancement outperformed other CBCT segmentation methods. Using this optimal segmentation scheme, strong correlation between extremity CBCT and μCT was achieved, with Pearson coefficients of 0.93 for BV/TV, 0.89 for Tb.Th, 0.91 for Tb.Sp, and 0.88 for Tb.N (all results statistically significant). Compared to a simple global CBCT segmentation using Otsu’s algorithm, the advanced segmentation method achieved ~20% improvement in the correlation coefficient for Tb.Th and ~50% improvement for Tb.Sp. Conclusions: Extremity CBCT combined with advanced image pre-processing and segmentation achieves high correlation with gold standard μCT in measurements of trabecular microstructure. This motivates ongoing development of clinical applications of extremity CBCT in in-vivo evaluation of bone health e.g. in early osteoarthritis and osteoporosis.

Computed Tomography

“…Moreover, with advanced image postprocessing, CBCT has demonstrated strong correlations with gold standard micro-CT in measurements of trabecular microstructure. 80 In its ability to simultaneously obtain bone vBMD and microstructural quality measures, CBCT might provide a useful adjunct to preclinical imaging using high-resolution peripheral quantitative CT, with the benefit of being readily applicable to larger joints. However, in comparison to high-resolution peripheral quantitative CT, CBCT tends to overestimate trabecular parameters and underestimate cortical parameters despite demonstrating overall bone architecture assessment feasibility.…”

“…Compared with traditional MDCT, the high isotropic spatial resolutions and lower effective radiation doses offered by CBCT for soft tissue imaging have a high potential to be sensitive to subtle changes in joint space morphology encountered in osteoarthritis and related conditions. Moreover, with advanced image postprocessing, CBCT has demonstrated strong correlations with gold standard micro-CT in measurements of trabecular microstructure 80 . In its ability to simultaneously obtain bone vBMD and microstructural quality measures, CBCT might provide a useful adjunct to preclinical imaging using high-resolution peripheral quantitative CT, with the benefit of being readily applicable to larger joints.…”

Quantitative evaluation of bone microstructure using high-resolution extremity cone-beam CT with a CMOS detector

“…We followed our previously developed framework for robust trabecular measurements in CBCT. 2 First, regions of interests (ROIs) were identified in the μCT of each core. Binary ROI masks were mapped to the CBCT volumes using transformations obtained from rigid registration of μCT and CBCT reconstructions.…”

“…5 In the pre-thresholding step, a Gaussian curve was fit to the global image histogram and all voxels with values below that corresponding to 50% of maximum of the fit were set to zero. 2 Bernsen's segmentation assigned each voxel to bone or background depending on mid-grey value of a sliding window centered on that voxel. 5 Bernsen's segmentation algorithm has two free parameters, the radius of the sliding window and the local contrast threshold.…”

“…1 We have shown that extremity CBCT can achieve high correlation with reference micro-CT (μCT) in metrics of trabecular microarchitecture (Pearson coefficient of ~0.8-0.9). 2 To enhance the performance of extremity CBCT in imaging of bone microstructure, we have recently developed a new system by replacing the aSi FPD with a custom CMOS detector. 3,4 The CMOS technology offers higher frame rates, smaller detector pixels, and lower electronic noise than aSi FPD.…”