非線性與復(fù)雜網(wǎng)絡(luò)理論在腦電數(shù)據(jù)分析中的應(yīng)用研究
本文關(guān)鍵詞:非線性與復(fù)雜網(wǎng)絡(luò)理論在腦電數(shù)據(jù)分析中的應(yīng)用研究 出處:《太原理工大學(xué)》2014年博士論文 論文類型:學(xué)位論文
更多相關(guān)文章: 腦電 復(fù)雜網(wǎng)絡(luò) 腦網(wǎng)絡(luò) 酗酒 同步性 非線性
【摘要】:人腦中的復(fù)雜連接在靜態(tài)和動態(tài)活動時(shí)均存在,不同的大腦神經(jīng)元、神經(jīng)元集群、腦區(qū)之間在各種時(shí)間-空間尺度上相互協(xié)調(diào)、相互作用,形成一個(gè)非線性的、高度復(fù)雜的網(wǎng)絡(luò),大腦的一切功能認(rèn)知均基于這個(gè)復(fù)雜網(wǎng)絡(luò)。腦電信號則是來自大腦這一復(fù)雜系統(tǒng)的生物電信號。近年來,將非線性動力學(xué)及復(fù)雜網(wǎng)絡(luò)理論應(yīng)用到腦電信號分析中,,力圖挖掘大腦不同腦區(qū)之間的耦合關(guān)系,以及腦網(wǎng)絡(luò)的拓?fù)浣Y(jié)構(gòu)在不同狀態(tài)下的變化規(guī)律,成為多學(xué)科交叉研究的熱點(diǎn)。非線性動力學(xué)與復(fù)雜網(wǎng)絡(luò)理論為大腦這一復(fù)雜系統(tǒng)的研究,提供了一個(gè)新的視角。 本研究分析并改進(jìn)了時(shí)間序列非線性檢測方法,對腦電數(shù)據(jù)的非線性特性進(jìn)行了檢測;多角度分析了大腦中的同步性;探討腦電功能腦網(wǎng)絡(luò)的構(gòu)建及分析方法。并在酗酒者腦電數(shù)據(jù)上,通過同步性與腦網(wǎng)絡(luò)屬性的分析,探索腦疾病狀態(tài)下大腦動力學(xué)的變化模式;挖掘酗酒這種神經(jīng)退行性疾病狀態(tài)下大腦同步性及網(wǎng)絡(luò)屬性與正常被試之間的差異;利用機(jī)器學(xué)習(xí)算法,尋找酗酒早期診斷的生理學(xué)指標(biāo)。 本文主要?jiǎng)?chuàng)新工作如下: (1)提出兩種新的時(shí)間序列非線性檢測方法 傳統(tǒng)的非線性檢測方法中采用的特征量有一定的不足,本研究提出基于樣本熵和基于模糊熵的兩種新的非線性檢測方法,并在仿真數(shù)據(jù)集上驗(yàn)證了新方法的準(zhǔn)確性與時(shí)間效率。與傳統(tǒng)檢測方法相比,新方法在準(zhǔn)確率上基本相同,但是時(shí)間效率大幅度提升。 (2)結(jié)合腦電的溯源技術(shù),提出一種新的同步性分析方法 目前腦電的同步性研究中,同步性的分析都是基于頭皮EEG數(shù)據(jù)進(jìn)行的。然而,由于腦電的容積導(dǎo)體效應(yīng),使得頭皮電極之間可能出現(xiàn)偽相關(guān)性,進(jìn)一步影響同步性的度量。本研究利用最新的腦電溯源技術(shù),度量了大腦皮層上自定義的ROI之間的同步性。最后在酗酒數(shù)據(jù)上對兩種同步性分析方法進(jìn)行了對比研究,發(fā)現(xiàn)與傳統(tǒng)的頭皮電極同步性相比,本文提出的溯源后EEG同步性分析方法能夠更加精確地度量大腦進(jìn)行認(rèn)知活動時(shí)的動力學(xué)變化。 (3)提出一種新的EEG腦網(wǎng)絡(luò)構(gòu)建的閾值選取方法 構(gòu)建無權(quán)無向網(wǎng)絡(luò)時(shí),需要設(shè)置合適的閾值T將相關(guān)矩陣轉(zhuǎn)換為二值矩陣。目前研究多是在單閾值下分別討論網(wǎng)絡(luò)的拓?fù)涮匦裕狙芯刻岢鲆环N新的EEG腦網(wǎng)絡(luò)構(gòu)建的閾值選擇方法,可利用網(wǎng)絡(luò)的小世界屬性,計(jì)算出一個(gè)稀疏度范圍作為閾值空間,并在該閾值空間內(nèi)分析了網(wǎng)絡(luò)的拓?fù)鋵傩宰兓?(4)首次利用復(fù)雜網(wǎng)絡(luò)理論構(gòu)建并分析了酗酒患者EEG功能腦網(wǎng)絡(luò) 酗酒者EEG的研究中,目前多采用計(jì)算某個(gè)通道或區(qū)域的能量、功率譜、熵等特性。本文首次嘗試?yán)脧?fù)雜網(wǎng)絡(luò)理論構(gòu)建并分析了酗酒者的EEG功能腦網(wǎng)絡(luò),挖掘了酗酒者與正常被試腦網(wǎng)絡(luò)拓?fù)鋵傩陨系牟町悾瑥囊粋(gè)新的角度揭示了酗酒引起的腦功能損傷。 總之,圍繞非線性動力學(xué)與復(fù)雜網(wǎng)絡(luò)相關(guān)理論,本研究重點(diǎn)研究了同步性分析方法和腦網(wǎng)絡(luò)的構(gòu)建分析方法,探討了大腦同步性與腦網(wǎng)絡(luò)拓?fù)鋵傩栽谛锞萍膊顟B(tài)下的變化規(guī)律,為酗酒可能引起的腦損傷提供了新的視角與證據(jù),是多學(xué)科交叉的一個(gè)新的研究成果。
[Abstract]:In recent years , nonlinear dynamics and complex network theory have been applied to EEG analysis . This paper analyzes and improves the time series nonlinear detection method , detects the nonlinear characteristic of the brain electrical data , analyzes the synchronism in the brain by multiple angles , and probes into the structure and the analysis method of the brain electrical function brain network . The main innovations of this paper are as follows : ( 1 ) Two new time series nonlinear detection methods are proposed . In this paper , we propose two new non - linear detection methods based on sample entropy and fuzzy entropy , and verify the accuracy and time efficiency of the new method . Compared with the traditional detection method , the new method is almost the same in accuracy rate , but the time efficiency is greatly improved . ( 2 ) Based on the tracing technique of EEG , a new synchronization analysis method is proposed . At present , the synchronization analysis of EEG is based on scalp EEG data . However , due to the volume conductor effect of EEG , there may be a false correlation between scalp electrodes , which further affects the measurement of synchronicity . In this study , the latest EEG tracing technique is used to measure the synchronicity between the customized ROI on the cerebral cortex . Finally , compared with the traditional scalp electrode synchronization , the EEG synchronization analysis method proposed in this paper can more accurately measure the dynamic change of the brain in cognitive activity . ( 3 ) A new threshold selection method for EEG network construction is proposed . In this paper , it is necessary to set appropriate threshold T to transform the correlation matrix into two - valued matrices when constructing the right - to - no network . At present , the topological properties of the network are discussed separately under the single threshold . A new threshold selection method for EEG network construction is proposed . A new EEG network constructed threshold selection method is proposed . A sparse range is calculated as the threshold space by using the small world attribute of the network , and the topological property change of the network is analyzed in the threshold space . ( 4 ) For the first time , the complex network theory is used to construct and analyze the EEG function brain network of the alcoholic patient . In the study of the EEG of alcoholics , the characteristics of energy , power spectrum and entropy of a certain channel or region have been calculated . First , this paper attempts to construct and analyze the EEG function brain network of alcoholics by using complex network theory . In a word , around the theory of nonlinear dynamics and complex network , this study focuses on the analysis of synchronicity analysis method and building analysis method of brain network , discusses the changing law of brain synchronization and brain network topology attribute in the state of alcoholism , provides a new visual angle and evidence for the brain injury which may be caused by alcohol abuse , and is a new research result of multi - disciplinary cross .
【學(xué)位授予單位】:太原理工大學(xué)
【學(xué)位級別】:博士
【學(xué)位授予年份】:2014
【分類號】:TN911.7;O157.5
【參考文獻(xiàn)】
相關(guān)期刊論文 前7條
1 孫俊峰;洪祥飛;童善保;;復(fù)雜腦網(wǎng)絡(luò)研究進(jìn)展——結(jié)構(gòu)、功能、計(jì)算與應(yīng)用[J];復(fù)雜系統(tǒng)與復(fù)雜性科學(xué);2010年04期
2 梁夏;王金輝;賀永;;人腦連接組研究:腦結(jié)構(gòu)網(wǎng)絡(luò)和腦功能網(wǎng)絡(luò)[J];科學(xué)通報(bào);2010年16期
3 毛文琪;李穎潔;;酗酒者認(rèn)知腦電能量的頭皮分布[J];上海大學(xué)學(xué)報(bào)(自然科學(xué)版);2009年05期
4 張文超;譚思超;高璞珍;張虹;張紅巖;;搖擺條件下自然循環(huán)流動不穩(wěn)定性的混沌特性研究[J];原子能科學(xué)技術(shù);2012年06期
5 方錦清;迅速發(fā)展的復(fù)雜網(wǎng)絡(luò)研究與面臨的挑戰(zhàn)[J];自然雜志;2005年05期
6 范金鋒;邵晨曦;王劍;楊明;王子才;;醉酒者腦電和正常腦電非線性特性的比較評估[J];中國生物醫(yī)學(xué)工程學(xué)報(bào);2008年01期
7 楊福生,廖旺才;近似熵:一種適用于短數(shù)據(jù)的復(fù)雜性度量[J];中國醫(yī)療器械雜志;1997年05期
本文編號:1410145
本文鏈接:http://www.sikaile.net/kejilunwen/wltx/1410145.html