Improved accuracy in the assessment of vertebral cortical thickness by quantitative computed tomography using the...

Area of Science:

• Bone biology and biomechanics
• Medical imaging and image processing
• Osteoporosis research

Background:

• Vertebral bone strength is significantly influenced by cortical bone properties.
• Diseases and therapies can differentially affect cancellous and cortical bone.
• Quantitative Computed Tomography (QCT) allows selective assessment of bone compartments, but accuracy is limited by image quality, especially for thin cortices.

Purpose of the Study:

• To present and validate an image processing method, Iterative Convolution OptimizatioN (ICON), designed to improve the accuracy of cortical bone thickness measurements from QCT and HR-QCT.
• To assess the impact of ICON on the accuracy of cortical thickness (Ct.Th) and bone mineral density (BMD) measurements compared to a high-resolution peripheral QCT (HR-pQCT) gold standard.

Main Methods:

• Ten human vertebrae were scanned using clinical QCT, HR-QCT, and HR-pQCT (gold standard).
• An in-house software (StructuralInsight) and manufacturer's software were used for microstructural variable and BMD calculations.
• The ICON method was applied to derive deconvolved cortical thickness (dcCt.Th), correcting for partial volume effects.

Main Results:

• Uncorrected QCT and HR-QCT measurements significantly overestimated cortical thickness compared to HR-pQCT.
• Initial QCT and HR-QCT based Ct.Th, BMD, and BMC showed no significant correlation with HR-pQCT.
• After ICON correction, dcCt.Th measurements from both QCT and HR-QCT showed significant correlations with HR-pQCT (r²=0.75 and r²=0.93, respectively).
• ICON reduced the overestimation bias of Ct.Th from >300% to <50% for QCT and <20% for HR-QCT.

Conclusions:

• Uncorrected QCT-based cortical bone data can be misleading due to segmentation inaccuracies.
• The ICON method substantially reduces random overestimation bias in cortical thickness measurements.
• ICON enables more accurate assessment of cortical thickness, improving the characterization of bone mechanical competence and warrants in vivo testing.