Quantitative dynamic contrast-enhanced MRI for mouse models using automatic detection of the arterial input function
Quantitative dynamic contrast-enhanced MRI for mouse models using automatic detection of the arterial input function
复制标题
DOI：
10.1002/nbm.1784
复制
发表时间：
2012-04-01
影响因子：
2.9
通讯作者：
Lee, Jung Hee
中科院分区：
文献类型：
--
作者：
Kim, Jae-Hun;Im, Geun Ho;Lee, Jung Hee
动态增强磁共振成像(DCE-MRI)是一种被广泛接受的肿瘤评估方法。DCE-MRI是一种无创性的微血管通透性、血容量和血流量的测量方法，在临床前研究中对于了解疾病机制和监测治疗反应非常有用。为了使用DCE-MRI准确量化药代动力学参数，需要测定肿瘤附近大动脉的动脉输入功能(AIF)。然而，由于肿瘤的空间尺寸或位置较小，手动确定小鼠MR图像中的AIF通常是困难的。在这项研究中，我们提出了一种利用Kendall协调系数从小鼠DCE-MR图像中自动检测AIF的算法。所提出的方法通过计算机模拟进行了测试，然后应用于荷瘤小鼠(n=8)。计算机仿真结果表明，该算法能够根据噪声水平对模拟AIF信号进行分类。我们发现用我们的方法计算的药代动力学参数与人工测定的AIF的药代动力学参数是相似的，KTrans(5.14+/-3.60%)、Ve(6.02+/-3.22%)、Vp(5.10+/-7.05%)和kep(5.38+/-4.72%)的差异是可以接受的。目前的研究结果表明，使用肯德尔协调系数自动定义的AIF对于定量DCE-MRI在小鼠癌症模型中的评估是有用的。版权所有(C)2011 John Wiley&Sons，Ltd.
Dynamic contrast-enhanced MRI (DCE-MRI) is widely accepted for the evaluation of cancer. DCE-MRI, a noninvasive measurement of microvessel permeability, blood volume and blood flow, is extremely useful for understanding disease mechanisms and monitoring therapeutic responses in preclinical research. For the accurate quantification of pharmacokinetic parameters using DCE-MRI, determination of the arterial input function (AIF) from a large arterial vessel near the tumor is required. However, a manual determination of AIF in mouse MR images is often difficult because of the small spatial dimensions or the location of the tumor. In this study, we propose an algorithm for the automatic detection of AIF from mouse DCE-MR images using Kendall's coefficient of concordance. The proposed method was tested with computer simulations and then applied to tumor-bearing mice (n=8). Results from computer simulations showed that the proposed algorithm is capable of categorizing simulated AIF signals according to their noise levels. We found that the resulting pharmacokinetic parameters computed from our method were comparable with those from the manual determination of AIF, with acceptable differences in Ktrans (5.14 +/- 3.60%), ve (6.02 +/- 3.22%), vp (5.10 +/- 7.05%) and kep (5.38 +/- 4.72%). The results of the current study suggest the usefulness of an automatically defined AIF using Kendall's coefficient of concordance for quantitative DCE-MRI in mouse models for cancer evaluation. Copyright (C) 2011 John Wiley & Sons, Ltd.