Section Biomedical Imaging, Department of Radiology and Neuroradiology, University Medical Center Schleswig-Holstein, MOIN CC, Am Botansichen Garten 14, 24118 Kiel, Germany.
Helmholtz Zentrum Geesthacht, Institute for Materials Research, Max-Planck Straße 1, 21502 Geesthacht, Germany.
Bone. 2019 Mar;120:194-203. doi: 10.1016/j.bone.2018.08.024. Epub 2018 Sep 8.
Vertebral whole bone strength is substantially affected by cortical bone properties. Disease and therapy may affect cancellous and cortical bone differently. Unlike Dual X-ray Absorptiometry (DXA), Quantitative Computed Tomography (QCT) permits selective assessment of cortical and cancellous bone, but image quality limits the accuracy. We present an image processing method specifically adopted to thin cortices that substantially improves accuracy. Ten human vertebrae embedded in epoxy resin were imaged using clinical QCT and High-Resolution QCT (HR-QCT) protocols, both acquired on a clinical whole body CT scanner, whereas high resolution peripheral QCT (HR-pQCT) was used as gold standard. Microstructural variables and BMD were calculated using in-house software StructuralInsight for QCT and HR-QCT and the manufacturer's μCT evaluation software for HR-pQCT. An adjusted measure, a deconvolved cortical thickness (dcCt.Th), corrected for partial volume effects, was derived applying the new Iterative Convolution OptimizatioN (ICON) method. Direct measurements of cortical thickness (Ct.Th) showed substantial overestimation with mean ± standard deviation of 1.8 ± 0.5 mm for QCT and 1.5 ± 0.3 mm for HR-QCT compared to 0.37 ± 0.07 mm using HR-pQCT. Correlations of both QCT (r = 0.05, p > 0.5.) and HR-QCT (r = 0.38, p = 0.060) with the gold standard HR-pQCT were not significant. Also QCT-based BMD and BMC as well as HR-QCT-based BMD did not show a significant correlation with the gold standard approach. Only HR-QCT-based BMC showed a modest correlation (r = 0.59, p = 0.01) After applying ICON corrections, dcCt.Th resulted in 0.52 ± 0.09 mm for QCT and 0.43 ± 0.07 mm for HR-QCT, both significantly correlated to HR-pQCT (r = 0.75, p = 0.0012 and r = 0.93, p < 0.0001, respectively). The average overestimation bias of Ct.Th was reduced from (402 ± 157)% to (45 ± 17)% for QCT and from (330 ± 69)% to (19 ± 8)% for HR-QCT. Due to inaccurate segmentation uncorrected QCT-based Ct.Th measures as well as BMD and BMC showed no correlation to HR-pQCT and thus such bias cortical data can be misleading. The application of ICON reduced random overestimation bias to about 50 μm and 20 μm for QCT and HR-QCT, respectively, leading to a recovery of a significant correlation with the reference data of HR-pQCT. This reveals the potential for fairly accurate assessment of cortical thickness, allowing to better characterize cortical mechanical competence. These results warrant testing of the performance in vivo.
骨整体强度受皮质骨特性的显著影响。疾病和治疗可能会对松质骨和皮质骨产生不同的影响。与双能 X 线吸收法（DXA）不同，定量计算机断层扫描（QCT）允许选择性评估皮质骨和松质骨，但图像质量限制了准确性。我们提出了一种专门用于薄皮质骨的图像处理方法，该方法显著提高了准确性。将 10 个人类椎体嵌入环氧树脂中，使用临床 QCT 和高分辨率 QCT（HR-QCT）协议进行成像，这两种协议均在临床全身 CT 扫描仪上采集，而高分辨率外周 QCT（HR-pQCT）则作为金标准。使用内部软件 StructuralInsight for QCT 和 HR-QCT 以及制造商的 μCT 评估软件计算微观结构变量和 BMD 用于 HR-pQCT。应用新的迭代卷积优化（ICON）方法得出校正部分容积效应的去卷积皮质厚度（dcCt.Th）的校正测量值。直接测量的皮质厚度（Ct.Th）显示出明显的高估，QCT 的平均值±标准偏差为 1.8±0.5mm，HR-QCT 为 1.5±0.3mm，而 HR-pQCT 为 0.37±0.07mm。QCT（r=0.05，p>0.5）和 HR-QCT（r=0.38，p=0.060）与金标准 HR-pQCT 的相关性均不显著。基于 QCT 的 BMD 和 BMC 以及基于 HR-QCT 的 BMD 与金标准方法也没有显示出显著相关性。只有基于 HR-QCT 的 BMC 显示出适度的相关性（r=0.59，p=0.01）。应用 ICON 校正后，QCT 的 dcCt.Th 结果为 0.52±0.09mm，HR-QCT 的结果为 0.43±0.07mm，均与 HR-pQCT 显著相关（r=0.75，p=0.0012 和 r=0.93，p<0.0001）。Ct.Th 的平均高估偏差从 QCT 的（402±157）%降低到（45±17）%，从 HR-QCT 的（330±69）%降低到（19±8）%。由于分割不准确，未经校正的基于 QCT 的 Ct.Th 测量值以及 BMD 和 BMC 与 HR-pQCT 均无相关性，因此这种皮质数据的偏差可能会产生误导。ICON 的应用将随机高估偏差降低到 QCT 和 HR-QCT 分别约为 50μm 和 20μm，从而恢复与 HR-pQCT 参考数据的显著相关性。这表明评估皮质厚度的准确性有很大的潜力，可以更好地描述皮质机械能力。这些结果值得进一步在体内进行性能测试。
J Bone Miner Metab. 2016-11
AJR Am J Roentgenol. 1996-7
J Bone Miner Res. 2022-2
Osteoporos Int. 2020-2
Zotero 插件