基于神经网络振荡的痴呆信号分析

发布时间：2018-05-19 05:13

本文选题：老年痴呆 + 轻度认知障碍　；参考：《燕山大学》2013年硕士论文

【摘要】：老年性痴呆是一种常见的功能性精神疾病，，随着社会老龄化的到来，老年性痴呆疾病已经给很多家庭带来沉重的负担，也越来越受到社会的重视。然而，当前对于这种疾病确切的病因和发病机制尚不清楚，且老年性痴呆前期的症状不明显，目前只有通过在脑皮质区域中发现广泛存在的老年斑及神经纤维缠结才能确诊，而确诊之后，病人已经处于较严重的疾病状态。大脑是老年痴呆的发病部位，由于功能性的变化要早于器质性变化，所以前期通过功能性诊断对于及早的发现和治疗老年痴呆具有重要意义。脑电分析是老年性痴呆诊断的重要途径，脑电中包含了认知和其它脑功能方面的丰富信息，我们可以通过研究脑电信号来揭示老年性痴呆疾病发作的生理机制，并采用脑电信号分析做为老年痴呆的诊断依据。本文采用全局同步指数方法和平行因子分析方法首次分别对轻度认知障碍患者和正常对照组进行分析。全局同步指数侧重于大脑的同步性，它可以综合整个脑区计算同时记录的多变量时间序列，结合所有特征值得到一个同步指数，根据其大小来判断同步性大小。与其他的同步算法相比，这种方法整合信息的能力更强，同时还据有灵敏度高、计算误差相对小和时间稳定性强等特性。因此可以将全局同步指数应用到复杂的、多维的、不稳定的系统中。平行因子分析是将连续小波变换得到的三维张量数据降维分解成时间域、频率域和空间域上的二维信息，然后利用统计方法分别揭示时域、频域和空间域上的脑电信号特征。这种方法对于多通道脑电信号的时间、空间和频率分解具有唯一性，能够准确提取出脑电信号中包含的信息。分析结果显示，经全局同步指数方法分析后，轻度认知障碍患者在δ，α和β3频带处的同步指数与对照组的同步指数有显著性差异，而且在这三个频带处的同步指数与症状的严重程度显著相关。经平行因子分析后，与对照组相比轻度认知障碍患者的优势节律显著降低，对应的能量大小也有所下降。
[Abstract]:Alzheimer's disease (AD) is a common functional mental disease. With the coming of aging society, Alzheimer's disease has brought heavy burden to many families and has been paid more and more attention by the society. However, the exact etiology and pathogenesis of the disease are not clear, and the symptoms of Alzheimer's disease are not obvious. It is only through the discovery of widespread senile plaques and neurofibrillary tangles in the cortical area of the brain that the diagnosis can be confirmed. After the diagnosis, the patient has been in a more serious state of disease. The brain is the site of Alzheimer's disease, because the functional changes are earlier than the organic changes, so the early functional diagnosis is of great significance for the early detection and treatment of Alzheimer's disease. EEG analysis is an important approach to the diagnosis of Alzheimer's disease. EEG contains rich information in cognitive and other brain functions. We can reveal the physiological mechanism of Alzheimer's disease by studying EEG. EEG analysis was used as the basis for the diagnosis of senile dementia. The global synchronous index method and parallel factor analysis method were used to analyze the patients with mild cognitive impairment and the normal control group for the first time. The global synchronization index focuses on the synchronicity of the brain. It can synthesize the whole brain region to calculate the multivariable time series recorded simultaneously, and combine all the eigenvalues to obtain a synchronization index, which can be used to judge the synchronicity. Compared with other synchronization algorithms, this method is more capable of integrating information, and has the characteristics of high sensitivity, relatively small calculation error and strong time stability. Therefore, the global synchronization index can be applied to complex, multidimensional, unstable systems. Parallel factor analysis (PFA) decomposes 3D Zhang Liang data obtained by continuous wavelet transform into two dimensional information in time domain, frequency domain and spatial domain, and then uses statistical methods to reveal EEG characteristics in time domain, frequency domain and spatial domain, respectively. This method is unique for the decomposition of time, space and frequency of multichannel EEG signals, and can extract the information contained in EEG signals accurately. The results showed that there was a significant difference between the synchro index at 未, 伪 and 尾 3 bands in mild cognitive impairment patients and the control group after global synchronization index analysis. And the synchronization index at these three bands was significantly correlated with the severity of the symptoms. After parallel factor analysis, compared with the control group, the dominant rhythm of patients with mild cognitive impairment decreased significantly, and the corresponding energy decreased.
【学位授予单位】：燕山大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TN911.6;R749.16

【参考文献】