Steiner Lukas, Synek Alexander, Pahr Dieter H
Institute for Lightweight Design and Structural Biomechanics, TU-Wien, Vienna, Austria.
Department of Anatomy and Biomechanics, Karl Landsteiner University of Health Sciences, Krems, Austria.
Bone Rep. 2020 May 4;12:100261. doi: 10.1016/j.bonr.2020.100261. eCollection 2020 Jun.
MicroCT-based morphological parameters are often used to quantify the structural properties of trabecular bone. Various software tools are available for calculating these parameters. Studies that examine the comparability of their results are rare. Four different software tools were used to analyse a set of 701 microCT images from human trabecular bone samples. Bone volume to total volume (/), bone surface (), trabecular thickness (. .) and degree of anisotropy () were evaluated. / shows very low difference (-0.18 ± 0.15%). The difference in could be reduced below 5% if artificial cut surfaces are not included. . . and . . show differences of maximal -12% although the same theoretical background is used. is most critical with differences from 4.75 ± 3.70% (medtool vs. Scanco), over -38.61 ± 13.15% (BoneJ vs. Scanco), up to 80.52 ± 50.04% (medtool vs. BoneJ). Quantitative results should be considered with caution, especially when comparing different studies. Introducing standardization procedures and the disclosure of underlying algorithms and their respective implementations could improve this issue.
基于微计算机断层扫描(MicroCT)的形态学参数常用于量化松质骨的结构特性。有多种软件工具可用于计算这些参数。但研究这些工具结果可比性的研究却很少见。使用了四种不同的软件工具来分析一组来自人类松质骨样本的701张微CT图像。评估了骨体积与总体积之比(/)、骨表面积()、小梁厚度(..)和各向异性程度()。/显示出非常低的差异(-0.18±0.15%)。如果不包括人工切割表面,的差异可降低至5%以下。..和..显示出最大-12%的差异,尽管使用的是相同的理论背景。最为关键,差异范围从4.75±3.70%(medtool与Scanco相比)、超过-38.61±13.15%(BoneJ与Scanco相比)到高达80.52±50.04%(medtool与BoneJ相比)。定量结果应谨慎考虑,尤其是在比较不同研究时。引入标准化程序以及公开基础算法及其各自的实现方式可能会改善这一问题。