Bligh Michael, Bidaut Luc, White R Allen, Murphy William A, Stevens Donna M, Cody Dianna D
Radiation Oncology Centers of Southwest Florida, Bradenton, FL, USA.
Acad Radiol. 2009 Feb;16(2):150-9. doi: 10.1016/j.acra.2008.08.007.
The impact of varying image acquisition parameters on the precision of measurements using quantitative computed tomography is currently based on studies performed before the advent of helical image acquisition and multidetector-row scanners. The aim of this study was to evaluate helical multidetector-row quantitative computed tomography to determine the factors contributing to the overall precision of measurements on quantitative computed tomography conducted using current vintage computed tomographic (CT) scanners.
The effects of CT protocol parameters (x-ray tube voltage and current, pitch, gantry rotation speed, detector configuration, table height, and reconstruction algorithm) and short-term scanner variation were examined on two commercially available quantitative CT (QCT) systems (ie, a combination of reference phantoms and analysis software) using seven multidetector-row CT scanners (available from a single vendor) operated in helical mode. Combined with simulated patient repositioning using three ex vivo spine specimens, precision (coefficient of variation) estimates were made on the basis of three scenarios: "best case," "routine case," and "worst case."
The overall best-case QCT precision was 1.4%, provided that no changes were permitted to the bone mineral density (BMD) scan protocol. Routine-case examination (with a BMD reference phantom in place) that permitted some variation in the x-ray tube current and table speed produced a precision of 1.8%. Without any constraints on the clinical QCT examinations, the worst-case precision was estimated at 3.6%.
Although small in appearance, these errors are for single time points and may increase substantially when monitoring changes through QCT measurements over several time points. This calls for increased caution and attention to detail whenever using helical multidetector-row quantitative computed tomography for the assessment of BMD change.
目前，关于使用定量计算机断层扫描时不同图像采集参数对测量精度的影响，是基于螺旋图像采集和多排探测器扫描仪出现之前所进行的研究。本研究的目的是评估螺旋多排探测器定量计算机断层扫描，以确定影响使用当前老式计算机断层扫描（CT）扫描仪进行的定量计算机断层扫描测量总体精度的因素。
使用七台以螺旋模式运行的多排探测器CT扫描仪（来自同一供应商），在两个商用定量CT（QCT）系统（即参考体模和分析软件的组合）上，研究了CT协议参数（X射线管电压和电流、螺距、机架旋转速度、探测器配置、检查床高度和重建算法）以及短期扫描仪变化的影响。结合使用三个离体脊柱标本模拟患者重新定位，基于三种情况进行精度（变异系数）估计：“最佳情况”、“常规情况”和“最差情况”。
如果不允许改变骨密度（BMD）扫描协议，总体最佳情况的QCT精度为1.4%。允许X射线管电流和检查床速度有一定变化的常规情况检查（放置BMD参考体模）产生的精度为1.8%。在临床QCT检查没有任何限制的情况下，最差情况的精度估计为3.6%。
尽管这些误差在单次测量时看起来较小，但在通过QCT测量监测多个时间点的变化时可能会大幅增加。这就要求在使用螺旋多排探测器定量计算机断层扫描评估BMD变化时，要格外谨慎并注重细节。
