Spatial Fuzzy C-Means Thresholding for Semiautomated Calculation of Percentage Lung Ventilated Volume From Hyperpolarized Gas and 1H MRI
Spatial Fuzzy C-Means Thresholding for Semiautomated Calculation of Percentage Lung Ventilated Volume From Hyperpolarized Gas and 1H MRI
复制标题
DOI：
10.1002/jmri.25804
复制
发表时间：
2018-03-01
影响因子：
4.4
通讯作者：
Wild, Jim M.
中科院分区：
文献类型：
--
作者：
Hughes, Paul J. C.;Horn, Felix C.;Wild, Jim M.
目的：开发一种用于超极化气体肺通气和质子解剖磁共振成像(MRI)扫描对的半自动(SA)和重复性分析的图像处理流水线。将软件计算总肺容量(TLV)、通气量(VV)和肺通气量百分比(%VV)的结果与目前手工“基本”方法和K-均值分割方法进行比较。材料和方法：6例患者接受1.5T超极化He-3和同呼吸H-1磁共振成像，另外6例患者接受超极化Xe-129和分离呼吸H-1 MRI扫描。一名专家观察员和两名有肺部图像分割经验的用户进行了图像分析。结果：当比较%Vv的值时，观察者之间使用SA方法的一致性(平均值；R-0.984，ICC-0.980，LOA-7.5%)比基本方法(平均值；R-0.863，ICC-0.873，LOA-14.2%)无显著差异(p(R)=0.25，p(ICC)=0.25，p(LOA)=0.5)。与基础方法(Mean；DSCVV=0.947，DSCTLV=0.957)相比，SA方法(Mean；DSCVV=0.973，DSCTLV=0.980)显著改善了VV和TLV面罩的DSC(P<0.01)。与基本方法(平均高估=5.0%)和SA方法(平均高估=9.7%)相比，K-Means方法系统地高估了%Vv，而与其他方法(平均ICC；K-Means与Basic=0.685，K-Means与SA=0.740)的一致性较差。结论：开发了一个半自动图像处理软件，与当前使用的基本方法相比，该软件提高了观察者之间的一致性和相关性，并提供了比K-Means方法更一致的分割。
Purpose: To develop an image-processing pipeline for semiautomated (SA) and reproducible analysis of hyperpolarized gas lung ventilation and proton anatomical magnetic resonance imaging (MRI) scan pairs. To compare results from the software for total lung volume (TLV), ventilated volume (VV), and percentage lung ventilated volume (% VV) calculation to the current manual "basic" method and a K-means segmentation method.Materials and Methods: Six patients were imaged with hyperpolarized He-3 and same-breath lung H-1 MRI at 1.5T and six other patients were scanned with hyperpolarized Xe-129 and separate-breath H-1 MRI. One expert observer and two users with experience in lung image segmentation carried out the image analysis. Spearman (R), Intraclass (ICC) correlations, Bland-Altman limits of agreement (LOA), and Dice Similarity Coefficients (DSC) between output lung volumes were calculated.Results: When comparing values of % VV, agreement between observers improved using the SA method (mean; R - 0.984, ICC-0.980, LOA - 7.5%) when compared to the basic method (mean; R - 0.863, ICC - 0.873, LOA - 14.2%) nonsignificantly (p(R) = 0.25, p(ICC) = 0.25, and p(LOA) = 0.50 respectively). DSC of VV and TLV masks significantly improved (P < 0.01) using the SA method (mean; DSCVV = 0.973, DSCTLV = 0.980) when compared to the basic method (mean; DSCVV = 0.947, DSCTLV = 0.957). K-means systematically overestimated % VV when compared to both basic (mean over-estimation = 5.0%) and SA methods (mean overestimation = 9.7%), and had poor agreement with the other methods (mean ICC; K-means vs. basic = 0.685, K-means vs. SA = 0.740).Conclusion: A semiautomated image processing software was developed that improves interobserver agreement and correlation of lung ventilation volume percentage when compared to the currently used basic method and provides more consistent segmentations than the K-means method.