Division of Neuroradiology, Department of Medical Imaging, University of Toronto, Toronto, Ontario, Canada.
Can J Neurol Sci. 2010 Jul;37(4):498-503. doi: 10.1017/s0317167100010532.
To compare the reproducibility of semi-automated vessel analysis software to manual measurement of carotid artery stenosis on computed tomography angiography (CTA).
Two observers separately analyzed 81 carotid artery CTAs using semi-automated vessel analysis software according to a blinded protocol. The software measured the narrowest stenosis in millimeters (mm), distal internal carotid artery (ICA) in mm, and calculated percent stenosis based on NASCET criteria. One observer performed this task twice on each carotid, the second analysis delayed two months in order to mitigate recall bias. Two other observers manually measured the narrowest stenosis in mm, distal ICA in mm, and calculated NASCET percent stenosis in a blinded fashion. Correlation coefficients were calculated for each group comparing the narrowest stenosis in mm, distal ICA in mm, and NASCET percent stenosis.
The semi-automated vessel analysis software provided excellent intraobserver correlation for narrowest stenosis in mm, distal ICA in mm, and NACSET percent stenosis (Pearson correlation coefficients of 0.985, 0.954, and 0.977 respectively). The semi-automated vessel analysis software provided excellent interobserver correlation (0.925, 0.881, and 0.892 respectively). The interobserver correlation for manual measurement was good (0.595, 0.625, and 0.555 respectively). There was a statistically significant difference in the interobserver correlation between the semi-automated vessel analysis software observers and the manual measurement observers (P < 0.001).
Semi-automated vessel analysis software is a highly reproducible method of quantifying carotid artery stenosis on CTA. In this study, semi-automated vessel analysis software determination of carotid stenosis was shown to be more reproducible than manual measurement.
比较半自动血管分析软件与计算机断层血管造影（CTA）上手动测量颈动脉狭窄的可重复性。
两名观察者根据盲法方案分别使用半自动血管分析软件对 81 例颈动脉 CTA 进行分析。该软件以毫米（mm）测量狭窄最窄处，以毫米（mm）测量颈内动脉（ICA）远端，并根据 NASCET 标准计算狭窄百分比。一名观察者在每侧颈动脉上进行两次该任务，第二次分析延迟两个月，以减轻回忆偏倚。另外两名观察者以盲法手动测量狭窄最窄处的狭窄处（mm）、颈内动脉（ICA）远端（mm），并计算 NASCET 狭窄百分比。为每组计算狭窄最窄处的狭窄处（mm）、颈内动脉（ICA）远端（mm）和 NASCET 狭窄百分比的相关系数。
半自动血管分析软件在狭窄最窄处的狭窄处（mm）、颈内动脉（ICA）远端（mm）和 NACSET 狭窄百分比（Pearson 相关系数分别为 0.985、0.954 和 0.977）方面提供了出色的观察者内相关性。半自动血管分析软件提供了出色的观察者间相关性（分别为 0.925、0.881 和 0.892）。手动测量的观察者间相关性良好（分别为 0.595、0.625 和 0.555）。半自动血管分析软件观察者与手动测量观察者之间的观察者间相关性存在统计学显著差异（P <0.001）。
半自动血管分析软件是一种高度可重复的 CTA 颈动脉狭窄量化方法。在这项研究中，半自动血管分析软件确定的颈动脉狭窄比手动测量更具可重复性。
