脑电非线性动力学特征指标自动定量方法在阿尔茨海默病早期诊断中的应用研究

阿尔茨海默病是一种最常见的发生在老年人中的脑部疾病，也是最常见的痴呆类型，但目前临床上还没有特效的治疗方法。轻度认知功能障碍被认为是预防和控制阿尔茨海默病发生的关键阶段。脑电图是目前临床上研究大脑功能活动和脑疾病诊断的一种经济、无创的检查手段。近年来，大量学者将非线性动力学分析方法应用于阿尔茨海默病的脑电研究，发现脑电信号的非线性动力学特征指标能够反映患者的认知功能损害程度。随着我国人口老龄化进程的加速，阿尔茨海默病的发病率逐年增加。因此，设计和开发脑电的非线性动力学特征指标自动定量算法，实现阿尔茨海默病的早期诊断和干预具有重要的科学意义和临床应用价值。本项目拟综合运用多学科的理论知识和方法，开展脑电非线性动力学特征指标自动定量方法研究，基于支持向量机方法实现轻度认知功能障碍和阿尔茨海默病分类识别的目的。研究成果将推动脑电非线性动力学研究向临床应用的转化，为阿尔茨海默病的早期诊断提供参考。

Alzheimer's disease(AD) is the most common form of dementia and the most common brain disease for the elderly. However, currently, there is no effective therapy method for it yet. Mild Congnitive Impair(MCI)is regarded as the key period to prevent and control AD. EEG is an inexpensive and non-invasive tool for brain functional research and diagnosis of brain diseases. In recent years, Nonlinear dynamical analysis of EEG signals has received much attention, and many researches find that nonlinear dynamical features of EEG are helpful for the diagnosis of AD. With the rapid increase of elderly population, the number of AD will increase year by year. Therefore,it is very important to design and develop nonlinear dynamical analysis algorithm for diagnosis of AD from EEG signals. This research try to achieve quantitative analysis of nonlinear dynamical features of EEG signals using multidisciplinary theoretical knowledge and methods, and aims to classify MCI and AD from normal aging using Support Vector Machine(SVM). The results will not only promote the application of nonlinear dynamical analysis from EEG signals to clinical diagnosis of AD, but also provide reference for early diagnosis of AD.