Robust Quantitative Assessment of Trabecular Microarchitecture in Extremity Cone-Beam CT Using Optimized Segmentation...

Area of Science:

• Orthopaedic imaging
• Biomedical engineering
• Radiology

Background:

• Assessing bone microarchitecture in vivo is difficult with current imaging methods.
• Flat-panel detector extremity Cone-Beam CT (CBCT) offers potential for detailed trabecular bone analysis.
• Advanced algorithms are needed to improve CBCT's resolution for fine bone structures.

Purpose of the Study:

• To evaluate the performance of extremity CBCT for quantitative analysis of trabecular bone microarchitecture.
• To develop and test advanced CBCT pre-processing and segmentation algorithms for accurate bone morphometry.

Main Methods:

• 35 transilliac bone biopsy samples were imaged using extremity CBCT (75 μm voxel size) and micro-CT (7.67 μm voxel size).
• Four CBCT segmentation methods were compared: global Otsu's thresholding, Bernsen's local thresholding, local thresholding with pre-thresholding, and local thresholding with pre-thresholding and Laplacian contrast enhancement.
• Correlations were assessed for Bone Volume over Total Volume (BV/TV), Trabecular Thickness (Tb.Th), Trabecular Spacing (Tb.Sp), and Trabecular Number (Tb.N).

Main Results:

• The combination of local thresholding, global pre-thresholding, and Laplacian contrast enhancement yielded the best results.
• This optimal segmentation achieved strong correlations with micro-CT: Pearson coefficients of 0.93 (BV/TV), 0.89 (Tb.Th), 0.91 (Tb.Sp), and 0.88 (Tb.N).
• Advanced segmentation improved correlation coefficients by ~20% for Tb.Th and ~50% for Tb.Sp compared to simple global segmentation.

Conclusions:

• Extremity CBCT with advanced image processing accurately quantifies trabecular bone microstructure.
• This technique shows high correlation with the gold standard micro-CT.
• Clinical applications for in vivo bone health assessment, including early osteoarthritis and osteoporosis, are encouraged.