Effect of micro-computed tomography voxel size and segmentation method on trabecular bone microstructure measures in...

Area of Science:

• Biomedical Engineering
• Orthopedics
• Materials Science

Background:

• Micro-computed tomography (μCT) is the standard for assessing trabecular bone microstructure in small animal models.
• Scanning and evaluation parameters in μCT can influence morphologic results, but their specific effects on bone outcomes are not fully understood.

Purpose of the Study:

• To investigate how μCT scanning voxel size (6-30 μm) affects trabecular bone outcomes in mouse vertebral bodies.
• To compare two global segmentation thresholding methods: qualitative 2D image assessment versus quantitative image histogram analysis.

Main Methods:

• Mouse vertebral body trabecular bone was scanned using μCT with varying nominal voxel sizes (6-30 μm).
• Trabecular bone parameters were quantified.
• Two segmentation thresholding methods (Qualitative vs. Histogram) were applied and compared.

Main Results:

• Nominal voxel size significantly affected connectivity density, trabecular thickness, and bone tissue mineral density.
• Segmentation methods yielded similar results at small voxel sizes but diverged at larger sizes.
• Qualitative segmentation better estimated bone volume fraction (BV/TV) and trabecular thickness; Histogram segmentation better estimated trabecular number, separation, and structure model index.

Conclusions:

• High-resolution μCT scans (smaller voxel sizes) are recommended for accurate trabecular bone microstructure estimation.
• The choice of segmentation method impacts specific trabecular bone outcomes, particularly with larger voxel sizes.
• Consideration of scan parameters and segmentation limitations is vital when interpreting results and drawing conclusions in small animal bone studies.