SUPERNORMALS’ OVER-TIME BRAIN MAP LINKS TO ALZHEIMER’S PATHOLOGY: A MULTIVARIATE PATTERN ANALYSIS

超常人脑部随时间变化的图谱与阿尔茨海默病病理学相关:一项多变量模式分析

阅读:2

Abstract

We recently investigated a cross-sectional relationship between the function of a selected set of brain regions and cognition in older adults with excellent memory capacity (a.k.a., Supernormals), and revealed that a unique brain map covering anterior and posterior regions in Supernormals protected their cognition from Alzheimer’s disease (AD) pathology (Lin et al., 2017, Cortex). In the present study, we continued the validation of the longitudinal relationship between the brain map and cognition and AD pathology in Supernormals. We hypothesized that an over-time stable brain map involving anterior and posterior regions would differentiate Supernormals from their cognitively normal or abnormal counterparts, and such a brain map would link to AD pathology. Here, we identified 22 Supernormals, 25 with normal cognition (NC), 70 with amnestic mild cognitive impairment (MCI) and 28 with AD. Multivariate pattern based searchlight analysis of resting-state functional MRI data was applied to identify brain regions with constantly high discriminative powers between Supernormals and NC over 2 years. The regions included right cerebellum, left middle temporal gyrus, right middle frontal gyrus, right precuneus, right precentral gyrus. The brain map composed by these regions predicted multiple cognitive assessments and their longitudinal changes with 51% to 82% accuracy, discriminated NC, MCI and AD at 62% to 78%, and predicted beta-amyloid/pTau at 71%. Our findings suggest that applying searchlight analysis with longitudinal resting-state imaging data from Supernormals construct a brain map that sheds lights on understanding Alzheimer’s disease associated neurodegeneration.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。