Kovalev Vassili, Kruggel Frithjof
United Institute of Informatics Problems, Belarus National Academy of Sciences, 220012 Minsk, Belarus.
IEEE Trans Med Imaging. 2007 May;26(5):678-85. doi: 10.1109/TMI.2007.895481.
The purpose of this work was to study specific texture properties of the brain's white matter (WM) based on conventional high-resolution T1-weighted magnetic resonance imaging (MRI) datasets. Quantitative parameters anisotropy and laminarity were derived from 3-D texture analysis. Differences in WM texture associated with gender were evaluated on an age-matched sample of 210 young healthy subjects (mean age 24.8, SD 3.97 years, 103 males and 107 females). Changes of WM texture with age were studied using 112 MRI-T1 datasets of healthy subjects aged 16 to 70 years (57 males and 55 females). Both texture measures indicated a "more regular" WM structure in females (p < 10(-6)). An age-related deterioration of WM structure manifests itself as a remarkable decline of both parameters (p < 10(-6)) that is more prominent in females (p < 10(-6)) than in males (p = 0.02). Texture analysis of anatomical MRI-T1 brain datasets provides quantitative information about macroscopic WM characteristics and helps discriminating between normal and pathological aging.
这项工作的目的是基于传统高分辨率T1加权磁共振成像(MRI)数据集研究大脑白质(WM)的特定纹理特性。通过三维纹理分析得出定量参数各向异性和分层性。在210名年龄匹配的年轻健康受试者(平均年龄24.8岁,标准差3.97岁,103名男性和107名女性)样本中评估与性别相关的WM纹理差异。使用112个年龄在16至70岁的健康受试者(57名男性和55名女性)的MRI-T1数据集研究WM纹理随年龄的变化。两种纹理测量均表明女性的WM结构“更规则”(p < 10^(-6))。WM结构与年龄相关的恶化表现为两个参数均显著下降(p < 10^(-6)),且在女性中(p < 10^(-6))比在男性中(p = 0.02)更明显。解剖学MRI-T1脑数据集的纹理分析提供了有关宏观WM特征的定量信息,并有助于区分正常衰老和病理性衰老。