Repositioning errors in measurement of vertebral attenuation values by computed tomography

Area of Science:

• Medical Imaging
• Radiology
• Biomedical Engineering

Background:

• Reproducibility errors in bone mineral density (BMD) measurements by computed tomography (CT) are a significant concern.
• Patient repositioning during scans is a primary source of these measurement variations.

Purpose of the Study:

• To quantify the impact of repositioning errors on bone mineral measurements using CT.
• To assess the magnitude of measurement variations due to displacement and rotation errors.

Main Methods:

• Two cadaver spine sections, each with three vertebral bodies, were used.
• Incremental movements in three displacement directions and two rotation directions were simulated.
• Measured attenuation changes were analyzed within the range of current scan plane selection techniques.

Main Results:

• Changes in measured attenuation were minimal within the achievable range of current scan plane selection.
• Compounded repositioning errors averaged 8.3 Hounsfield units (HU) for a single vertebra and 4.9 HU for three adjacent vertebrae.
• The largest estimated change attributable to repositioning error was approximately 15 HU for a single vertebra and 8 HU for three adjacent vertebrae.

Conclusions:

• Repositioning errors contribute to measurement variability in CT-based bone mineral assessment.
• Current scan plane selection techniques minimize variations within their specified limits.
• Understanding and mitigating repositioning errors are crucial for improving the reproducibility of bone mineral measurements.