Title: Preclinical validation of automated DXA- and CT-based body composition measurements
Authors: Devriese, Joke ×
Beels, Laurence
Van de Wiele, C
Maes, A
Gheysens, Olivier
Pottel, Hans #
Issue Date: Mar-2016
Publisher: Minerva Medica
Series Title: The Quarterly Journal of Nuclear Medicine and Molecular Imaging vol:60 issue:1 pages:40-7
Abstract: BACKGROUNDː To determine and validate a set of Hounsfield unit (HU) ranges to segment computed tomography (CT) images into tissue types and to test the validity of dual-energy X-ray absorptiometry (DXA) tissue segmentation on pure, unmixed porcine tissues. METHODSː This preclinical prospective study was approved by the local ethical committee. Different quantities of porcine bone tissue (BT), lean tissue (LT) and adipose tissue (AT) were scanned using DXA and CT. Tissue type segmentation in DXA was performed via the standard clinical protocol and in CT through different sets of HU ranges. Percent coefficients of variation (%CV) were used to assess precision while % differences of observed masses were tested against zero using the Wilcoxon signed-rank test. RESULTSː Total mass DXA measurements differ little but significantly (p = 0.016) from true mass, while total mass CT measurements based on literature values show non-significant (p = 0.69) differences of 1.7% and 2.0%. BT mass estimates with DXA differed more from true mass (median -78.2 to -75.8%) than other tissue types (median -11.3 to -8.1%). Tissue mass estimates with CT and literature HU ranges showed small differences from true mass for every tissue type (median -10.4 to 8.8%). CONCLUSIONː The most suited method for automated tissue segmentation is CT and can become a valuable tool in quantitative nuclear medicine.
ISSN: 1824-4785
Publication status: published
KU Leuven publication type: IT
Appears in Collections:Public Health and Primary Care, Campus Kulak Kortrijk
Nuclear Medicine & Molecular Imaging
Department of Public Health miscellaneous
× corresponding author
# (joint) last author

Files in This Item:
File Description Status SizeFormat
PreclinicalValidationJoke_2015.pdf Accepted 557KbAdobe PDFView/Open Request a copy

These files are only available to some KU Leuven Association staff members


All items in Lirias are protected by copyright, with all rights reserved.