Volumetric measurements of bone mineral density of the lumbar spine: Comparison of three geometrical approximations using dual-energy X-ray absorptiometry (DXA)
Measurement of bone mineral density using dual-energy X-ray absorptiometry (DXA) gives area values (g cm) rather than true volumetric values (g cm). To calculate the vertebral volume using planar postero-anterior and lateral DXA values, several different geometrical approximations were usedcubic, cy...
Gespeichert in:
Veröffentlicht in: | Nuclear medicine communications 1998-08, Vol.19 (8), p.727-734 |
---|---|
Hauptverfasser: | , , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Measurement of bone mineral density using dual-energy X-ray absorptiometry (DXA) gives area values (g cm) rather than true volumetric values (g cm). To calculate the vertebral volume using planar postero-anterior and lateral DXA values, several different geometrical approximations were usedcubic, cylindrical with a circular cross-section and cylindrical with an elliptical cross-section. The aim of this study was to compare these geometrical approximations with each other and with a reference standard, defined as the volume found on a computed tomographic (CT) scan. L2 and L3 were evaluated in a phantom study. Volume approximations by the cube or cylinder with circular cross-section geometry showed more than a 50% overestimation (range 54–74%). However, the elliptical cylinder approach showed very good argreement2.1%, and 1.2% for L2 and L3, respectively, when compared to the CT volumes. In addition, we performed four patient studies with both CT and DXA to evaluate the elliptical cylinder estimate in a clinical setting. For L2 and L3, the mean relative difference was less than 2%. We conclude that the elliptical cylinder approach results in the most accurate bone volume estimates in both the phantom and patients. (© 1998 Lippincott Williams & Wilkins) |
---|---|
ISSN: | 0143-3636 1473-5628 |
DOI: | 10.1097/00006231-199808000-00002 |