Monash Biomedical Imaging, Monash University, Clayton, VIC, Australia; Faculty of Information Technology, Monash University, Clayton, VIC, Australia; ARC Centre of Excellence for Integrative Brain Function, Melbourne, VIC, Australia; Monash Institute of Cognitive and Clinical Neurosciences, Monash University, Clayton, VIC, Australia.
Monash Biomedical Imaging, Monash University, Clayton, VIC, Australia; Monash Imaging, Monash Health, Clayton, VIC, Australia.
Neuroimage. 2018 Jan 15;165:294-305. doi: 10.1016/j.neuroimage.2017.10.049. Epub 2017 Oct 25.
To improve the accuracy of automated vein segmentation by combining susceptibility-weighted images (SWI), quantitative susceptibility maps (QSM), and a vein atlas to produce a resultant image called a composite vein image (CV image).
An atlas was constructed in common space from manually traced MRI images from ten volunteers. The composite vein image was derived for each subject as a weighted sum of three inputs; an SWI image, a QSM image and the vein atlas. The weights for each input and each anatomical location, called template priors, were derived by assessing the accuracy of each input over an independent data set. The accuracy of vein segmentations derived automatically from each of the CV image, SWI, and QSM image sets was assessed by comparison with manual tracings. Three different automated vein segmentation techniques were used, and ten performance metrics evaluated.
Vein segmentations using the CV image were comprehensively better than those derived from SWI or QSM images (mean Cohen's d = 1.1). Sixty permutations of performance metric, benchmark image, and automated segmentation technique were evaluated. Vein identification improvements that were both large and significant (Cohen's d > 0.80, p < 0.05) were found in 77% of the permutations, compared to no improvement in 5%.
The accuracy of automated vein segmentations derived from the composite vein image was overwhelmingly superior to segmentations derived from SWI or QSM alone.
通过结合磁敏感加权成像（SWI）、定量磁化率图（QSM）和静脉图谱，以产生一种称为复合静脉图像（CV 图像）的结果图像，来提高自动静脉分割的准确性。
从十个志愿者的手动追踪 MRI 图像中构建图谱。为每个受试者生成复合静脉图像，作为三个输入的加权和；SWI 图像、QSM 图像和静脉图谱。每个输入和每个解剖位置的权重，称为模板先验，是通过评估每个输入在独立数据集上的准确性来得出的。通过与手动追踪进行比较，评估自动从 CV 图像、SWI 和 QSM 图像集中得出的静脉分割的准确性。使用了三种不同的自动静脉分割技术，并评估了十个性能指标。
使用 CV 图像的静脉分割总体上优于从 SWI 或 QSM 图像得出的分割（平均 Cohen's d=1.1）。评估了性能指标、基准图像和自动分割技术的 60 次排列。在 77%的排列中发现了具有较大和显著（Cohen's d>0.80，p<0.05）的静脉识别改进，而 5%的排列没有改进。
从复合静脉图像得出的自动静脉分割的准确性远远优于单独从 SWI 或 QSM 得出的分割。
