Linearity and accuracy errors in bone densitometry

Area of Science:

• Medical Imaging
• Bone Densitometry
• Biomedical Engineering

Background:

• Dual energy X-ray absorptiometry (DXA) is a standard method for assessing bone mineral content (BMC) and bone mineral density (BMD).
• Variations in DXA device performance can affect the accuracy and reliability of osteoporosis diagnosis and monitoring.
• Understanding instrument-specific errors is crucial for clinical interpretation and inter-system comparisons.

Purpose of the Study:

• To quantify accuracy and linearity errors in bone densitometry using dual energy X-ray absorptiometry (DXA).
• To identify specific sources of error in commercially available DXA instruments for planar bone measurements.

Main Methods:

• Conducted in vitro phantom studies and in vivo investigations with human volunteers.
• Evaluated commercially available bone densitometers used for planar densitometry.
• Assessed measurements of vertebral size, bone contours, bone mineral content (BMC), bone mineral density (BMD), and projected bone area.

Main Results:

• Observed significant differences in measurements of vertebral size and bone contours between instruments.
• Identified falsely low BMC in low bone mass cases due to software omission of edge regions.
• Noted increases in projected bone area with increasing nominal bone mass in some devices.
• Found clinically significant accuracy errors in BMC and BMD, and substantial linearity errors in BMC measurements.
• Demonstrated that linearity errors reduce the utility of BMC as a parameter.

Conclusions:

• DXA devices exhibit a combination of accuracy and linearity errors, varying by instrument.
• Linearity errors significantly impair the ability to monitor changes in BMC and, to a lesser extent, BMD.
• The presence of these errors complicates intercomparison between different DXA systems, affecting clinical utility.