Maximov Ivan I, Vellmer Sebastian
Department of Psychology, University of Oslo, Oslo, Norway; Norwegian Centre for Mental Disorders Research (NORMENT), KG Jebsen Centre for Psychosis Research, Oslo University Hospital, Oslo, Norway and Institute of Clinical Medicine, University of Oslo, Oslo, Norway.
Bernstein Center for Computational Neuroscience, Berlin, Germany; Department of Physics, Humboldt University, Berlin, Germany.
Magn Reson Imaging. 2019 Apr;57:124-132. doi: 10.1016/j.mri.2018.11.007. Epub 2018 Nov 23.
Perfusion magnetic resonance imaging (MRI) is a promising non-invasive technique providing insights regarding the brain's microvascular architecture in vivo. The scalar perfusion metrics can be used for quantitative diagnostics of various brain abnormalities, in particular, in the stroke cases and tumours. However, conventional MRI-based perfusion approaches such as dynamic contrast-enhanced perfusion imaging or arterial spin labelling have a few weaknesses, for instance, contrast agent deposition, low signal-to-noise ratio, limited temporal and spatial resolution, and specific absorption rate constraints. As an alternative, the intravoxel incoherent motion (IVIM) approach exploits an extension of diffusion MRI in order to estimate perfusion parameters in the human brain. Application of IVIM imaging at ultra-high field MRI might employ the advantage of a higher signal-to-noise ratio, and thereby the use of higher spatial and temporal resolutions. In the present work, we demonstrate an application of recently developed isotropic diffusion weighted sequences to the evaluation of IVIM parameters at an ultra-high 7T field. The used sequence exhibits high immunity to image degrading factors and allows one to acquire the data in a fast and efficient way. Utilising the bi-exponential fitting model of the signal attenuation, we performed an extensive analysis of the IVIM scalar metrics obtained by a isotropic diffusion weighted sequence in vivo and compared results with a conventional pulsed gradient sequence at 7T. In order to evaluate a possible metric bias originating from blood flows, we additionally used a truncated b-value protocol (b-values from 100 to 200 s/mm with the step 20 s/mm) accompanied to the full range (b-values from 0 to 200 s/mm). The IVIM scalar metrics have been assessed and analysed together with a large and middle vessel density atlas of the human brain. We found that the diffusion coefficients and perfusion fractions of the voxels consisting of large and middle vessels have higher values in contrast to other tissues. Additionally, we did not find a strong dependence of the IVIM metrics on the density values of the vessel atlas. Perspectives and limitations of the developed isotropic diffusion weighted perfusion are presented and discussed.
灌注磁共振成像（MRI）是一种很有前景的非侵入性技术，可在体内深入了解大脑的微血管结构。标量灌注指标可用于各种脑部异常的定量诊断，特别是在中风病例和肿瘤中。然而，传统的基于MRI的灌注方法，如动态对比增强灌注成像或动脉自旋标记，存在一些弱点，例如造影剂沉积、低信噪比、有限的时间和空间分辨率以及比吸收率限制。作为一种替代方法，体素内不相干运动（IVIM）方法利用扩散MRI的扩展来估计人脑的灌注参数。在超高场MRI中应用IVIM成像可能利用更高信噪比的优势，从而使用更高的空间和时间分辨率。在本研究中，我们展示了最近开发的各向同性扩散加权序列在超高7T场评估IVIM参数中的应用。所使用的序列对图像退化因素具有高度抗性，并允许以快速有效的方式获取数据。利用信号衰减的双指数拟合模型，我们对通过各向同性扩散加权序列在体内获得的IVIM标量指标进行了广泛分析，并将结果与7T时的传统脉冲梯度序列进行了比较。为了评估可能源于血流的指标偏差，我们还使用了截断的b值协议（b值从100到200 s/mm，步长为20 s/mm）以及全范围（b值从0到200 s/mm）。IVIM标量指标已与人类大脑的大、中血管密度图谱一起进行了评估和分析。我们发现，与其他组织相比，由大、中血管组成的体素的扩散系数和灌注分数具有更高的值。此外，我们没有发现IVIM指标对血管图谱密度值有强烈依赖性。本文介绍并讨论了所开发的各向同性扩散加权灌注的前景和局限性。
