Alzheimer’s disease (AD) is a neurodegenerative disease and now seen as a major and increasing public health concern. The early intervention will be able to prevent Alzheimer’s symptoms and/or slow down the process of deterioration. However, there is currently no effective clinical methods for the early diagnosis of AD. Further, it is still a big challenge to find accurate, non-invasive and robust biomarkers for the early diagnosis of AD. Recent evidences suggest the sulcal morphology is associated with cognitive functioning in the elderly, and is abnormal in mild cognitive impairment (MCI) patients. Those findings suggest sulcal patterns have potential to be valuable biomarkers for the early diagnosis of AD. The main aim of present project is to identify robust neuroimaging biomarkers predicting transition from cognitive normal to AD, by measuring cortical folding (e.g. global sulcal indies, sulcal width and depth) and traditional indices of brain structure (e.g. cortical thickness, and cortical gray matter, white matter and subcortical volumes) based on large-scale longitudinal multi-site magnetic resonance imaging (MRI) data. We will investigate longitudinal changes in brain structure with the three categories of MCI: MCI -converter (MCI convert to AD), MCI –stable (persisting MCI) and MCI reverter (MCI revert to cognitive normal), and thus help in determining the effective biomarkers. We will combine different features (including cognitive performances) using a multi-view learning strategy to improve the accuracy of the early detection. The project could help to understand the changes in brain structure, especially in cortical sulci, during the course of disease progression from normal cognition to MCI to AD, and thus help in predicting the prognosis of MCI and detecting AD early.
阿尔茨海默病(AD)是一种严重威胁老年人健康的神经退行性疾病。虽然对AD的早期干预将有效延缓病情发展,但因缺乏准确、非侵入性的生物标志,目前并没有非常有效的AD早期临床诊断的方法。近期的研究发现脑沟的形态学特征与认知功能存在显著的相关性,而且轻度认知障碍(MCI)患者的脑沟呈现异常表征。这表明脑沟结构的评定,有望成为AD早期诊断的辅助方法之一。本项目拟基于纵向型、多中心、大样本的核磁共振成像(MRI)数据,测量并构建多条脑沟的系列指标(包括沟回复杂度、宽度和深度等),和传统MRI指标(如灰质、白质体积、皮层厚度等)。针对MCI的三种转归状态,追踪脑结构的动态变化,确定有效的生物标志。并结合认知功能指标,构建基于多视角学习的分类与预测框架,以提高AD早期识别的准确率。本项目可以在一定程度上提供新的科学依据,以理解AD早期脑结构的变化异常,并有望在此基础上,为AD的早期诊断提供新的辅助方法。
阿尔茨海默病(AD)是一种严重威胁老年人健康的神经退行性疾病。虽然对AD的早期干预将有效延缓病情发展,但因缺乏准确、非侵入性的生物标志,目前并没有非常有效的AD早期临床诊断的方法。本项目基于纵向型、多中心、大样本的核磁共振成像(MRI)数据,针对老年人群与AD早期阶段人群,研究与分析了一系列脑沟的特征指标(包括沟回复杂度、宽度和深度等),和传统MRI指标(如灰质、白质体积、 皮层厚度等),并基于这些指标运用机器学习方法构建疾病的分类模型。代表性结果如下:1)针对老年人群,构建并分析大脑主要的16条脑沟结构指标,发现了脑沟结构,特别是额上沟,随脑老化的非线性变化。非线性轨迹的拐点被发现在75到80岁之间,这表明大脑皮层从75岁之后开始加速脑萎缩;2)基于多种结构与认知指标,对AD的早期MCI阶段进行了亚型分类。对记忆损伤型MCI和正常认知被试的分类,我们取得了81%的正确率,并发现对MCI亚型分类有效的特征中,海马和杏仁核等特征的判别效果最好;3)构建了3种分类模型,基于脑沟,皮层厚度,皮层体积,皮层下体积等影像特征,识别早期阶段的AD患者。结果表明脑沟复杂度与外侧裂宽度是2个最敏感的特征指标,有望作为AD早期诊断的有效影像标志物。综上,本项目可以在一定程度上提供新的科学依据,以理解AD早期脑结构的变化异常,并有望在此基础上,为AD的早期诊断提供新的辅助方法。
{{i.achievement_title}}
数据更新时间:2023-05-31
基于SSVEP 直接脑控机器人方向和速度研究
伴有轻度认知障碍的帕金森病~(18)F-FDG PET的统计参数图分析
内点最大化与冗余点控制的小型无人机遥感图像配准
针灸治疗胃食管反流病的研究进展
天津市农民工职业性肌肉骨骼疾患的患病及影响因素分析
面向阿尔茨海默病的脑功能网络及分类研究
早期阿尔茨海默病胼胝体与海马的时空特征研究
基于脑网络组的阿尔茨海默病早期脑功能代偿机制的研究
基于纵向多模态数据的阿尔茨海默症早期诊断方法研究