Modeling the Presence of Myelin and Edema in the Brain Based on Multi-Parametric Quantitative MRI.
Modeling the Presence of Myelin and Edema in the Brain Based on Multi-Parametric Quantitative MRI.
复制标题
DOI：
10.3389/fneur.2016.00016
复制
发表时间：
2016
影响因子：
3.4
通讯作者：
Lundberg P
中科院分区：
文献类型：
--
作者：
Warntjes M;Engström M;Tisell A;Lundberg P
本研究的目的是提出一种模型，使用多参数定量MRI来估计脑中髓鞘和水肿的存在。该模型涉及同时测量R1和R2弛豫速率和质子密度的四个部分体积的隔间，包括髓鞘部分体积，细胞部分体积，自由水部分体积，和过量的实质水部分体积。使用20个健康对照组的空间归一化脑图像获得模型参数。根据髓鞘含量的减少和过量实质水的存在对病理性脑进行建模，这表明水肿的程度。该方法进行了测试，一组20名年龄匹配的多发性硬化症（MS）患者的空间归一化的脑图像。与健康对照组相比，观察到明显差异：MS组的脑体积小79 mL（1069 vs. 1148 mL），髓鞘体积小38 mL（119 vs. 157 mL），脑实质多余水量大21 mL（78 vs. 57 mL）。各种脑结构的感兴趣区域的模板表明，MS组中灰质（GM）结构的髓鞘部分体积低1.6 ± 1.5%，白色（WM）结构的髓鞘部分体积低2.8 ± 1.0%。对于GM和WM，过量的实质水部分体积分别大9 ± 10%和5 ± 2%。手动放置的感兴趣区域表明，使用模板感兴趣区域的结果可能会因空间标准化过程而丢失解剖细节。针对三个个体受试者提供了该方法在高分辨率图像上的应用的示例：45岁的健康受试者、72岁的健康受试者和45岁的MS患者。考虑到年龄和疾病，观察结果与预期的行为一致。总之，基于具有临床可接受扫描时间的单次qMRI采集，所提出的模型可以提供临床重要参数，例如总脑体积、髓鞘形成程度和水肿程度。
The aim of this study was to present a model that uses multi-parametric quantitative MRI to estimate the presence of myelin and edema in the brain. The model relates simultaneous measurement of R1 and R2 relaxation rates and proton density to four partial volume compartments, consisting of myelin partial volume, cellular partial volume, free water partial volume, and excess parenchymal water partial volume. The model parameters were obtained using spatially normalized brain images of a group of 20 healthy controls. The pathological brain was modeled in terms of the reduction of myelin content and presence of excess parenchymal water, which indicates the degree of edema. The method was tested on spatially normalized brain images of a group of 20 age-matched multiple sclerosis (MS) patients. Clear differences were observed with respect to the healthy controls: the MS group had a 79 mL smaller brain volume (1069 vs. 1148 mL), a 38 mL smaller myelin volume (119 vs. 157 mL), and a 21 mL larger excess parenchymal water volume (78 vs. 57 mL). Template regions of interest of various brain structures indicated that the myelin partial volume in the MS group was 1.6 ± 1.5% lower for gray matter (GM) structures and 2.8 ± 1.0% lower for white matter (WM) structures. The excess parenchymal water partial volume was 9 ± 10% larger for GM and 5 ± 2% larger for WM. Manually placed ROIs indicated that the results using the template ROIs may have suffered from loss of anatomical detail due to the spatial normalization process. Examples of the application of the method on high-resolution images are provided for three individual subjects: a 45-year-old healthy subject, a 72-year-old healthy subject, and a 45-year-old MS patient. The observed results agreed with the expected behavior considering both age and disease. In conclusion, the proposed model may provide clinically important parameters, such as the total brain volume, degree of myelination, and degree of edema, based on a single qMRI acquisition with a clinically acceptable scan time.