Young Karl, Du An-Tao, Kramer Joel, Rosen Howard, Miller Bruce, Weiner Michael, Schuff Norbert
Department of Radiology, University of California-San Francisco, and VA Medical Center, 4150 Clement Street, San Francisco, CA 94121, USA.
Hum Brain Mapp. 2009 May;30(5):1667-77. doi: 10.1002/hbm.20632.
The goal of this project was to utilize an information theoretic formalism for medical image analysis initially proposed in [Young et al. (2005): Phys Rev Lett 94:098701-1] to detect and quantify subtle global and regional differences in spatial patterns in patients suffering from Alzheimer's disease (AD) and frontotemporal dementia (FTD) by estimating the structural complexity of anatomical brain MRI. The sensitivity and specificity of the results are compared with those of a recent analysis, currently considered state of the art for MR studies of neurodegeneration. The previous study used regional estimates of cortical thinning and/or volume loss to differentiate between normal aging, AD, and FTD. The analysis illustrates that the structural complexity estimation method, a general multivariate approach to the study of variation in brain structure which does not depend on highly specialized volumetric and thickness estimates, is capable of providing sensitive and interpretable diagnostic information.
本项目的目标是利用最初在[杨等人（2005年）：《物理评论快报》94:098701-1]中提出的一种信息论形式主义，通过估计解剖学脑磁共振成像（MRI）的结构复杂性，来检测和量化阿尔茨海默病（AD）和额颞叶痴呆（FTD）患者在空间模式上细微的整体和区域差异。将结果的敏感性和特异性与最近一项分析的结果进行比较，该分析目前被认为是神经退行性疾病磁共振研究的最新技术水平。先前的研究使用皮质变薄和/或体积损失的区域估计来区分正常衰老、AD和FTD。分析表明，结构复杂性估计方法是一种研究脑结构变化的通用多变量方法，不依赖于高度专业化的体积和厚度估计，能够提供敏感且可解释的诊断信息。
