Table 1.

Summary of the approaches used for CT measurements of bone density.^{24,26,36–41}

Method	Notes
Traditional phantom-based synchronous calibration	• Patient lies on an ergonomic phantom with materials of known densities (usually 2–5 rods of different human tissue density equivalents)
	• CT attenuation values of the hip or spine are converted to BMD by reference to the known density values (QCT Pro)
	• Hip scans can be adapted to derive areal BMD, suitable for use in FRAX (CTXA)
Phantom-less synchronous internal calibration	• No external calibration phantom scanned
	• CT attenuation of adjacent internal tissues (e.g. blood or fat) used to calibrate attenuation measurements (VirtuOst)
	• Can be adapted to derive areal BMD, suitable for use in FRAX (VirtuOst Hip, T-score)
Asynchronous external calibration	• Phantom scanned regularly.
	• Simple, single-material phantom (Mindways Model 4 phantom, CliniQCT)
	• Hounsfield numbers of bone are then compared with phantom
	• Asynchronous CT of proximal femur can be adapted to derive areal BMD, suitable for use in FRAX (CliniQCT CTXA)
Asynchronous external calibration with the ACRad phantom	• Routine calibration using ACRad phantom
	• Direct CT attenuation values (HUs) are used to determine trabecular radiodensity without a BMD-specific calibration phantom
	• Does not require specialised software – can be performed on PACS workstation or any computer with standard tools used for viewing CT images

ACRad, American College of Radiology; BMD, bone mineral density; CT, computed tomography; CTXA, CT X-ray absorptiometry; FRAX, fracture risk assessment tool; HU, Hounsfield units; PACS, picture archive and communication system; QCT, quantitative CT.