癲癇腦電棘波檢測(cè).doc
約40頁(yè)DOC格式手機(jī)打開(kāi)展開(kāi)
癲癇腦電棘波檢測(cè),1.6萬(wàn)字自己原創(chuàng)的畢業(yè)論文,僅在本站獨(dú)家出售,重復(fù)率低,推薦下載使用摘要 癲癇,是一種生活中篅@<哪圓考膊?,熏F(xiàn)氐匚:α宋頤塹納硤寰竦慕】?。它发醉d氖焙蟣硐治竽云げ慵捌げ閬祿抑釋磐蝗還鵲鬧馗蔥苑諾紓賈虜∪松硤宓某櫬ぃ饈兜母謀湟約熬竦囊斐5鵲?。所疫\(yùn)擔(dān)頤竊躚擇拆錟緣縲藕漚蟹淺W既返姆...
內(nèi)容介紹
此文檔由會(huì)員 淘寶大夢(mèng) 發(fā)布
癲癇腦電棘波檢測(cè)
1.6萬(wàn)字
自己原創(chuàng)的畢業(yè)論文,僅在本站獨(dú)家出售,重復(fù)率低,推薦下載使用
摘要 癲癇,是一種生活中篅@<哪圓考膊?,熏F(xiàn)氐匚:α宋頤塹納硤寰竦慕】?。它发醉d氖焙蟣硐治竽云げ慵捌げ閬祿抑釋磐蝗還鵲鬧馗蔥苑諾紓賈虜∪松硤宓某櫬?,茵}兜母謀湟約熬竦囊斐5鵲?。所疫\(yùn)擔(dān)頤竊躚擇拆錟緣縲藕漚蟹淺W既返姆治讎卸暇拖緣煤苤匾簿哂兄匾庖?。木壚_�(Electroencephalogram,EEG)產(chǎn)生于大腦神經(jīng)細(xì)胞群在生理過(guò)程中自發(fā)或誘發(fā)的電活動(dòng),表現(xiàn)為神經(jīng)細(xì)胞群在頭皮表面或者顱內(nèi)引起的電位信號(hào)的變化,它反映了大腦生物電的節(jié)律活動(dòng)規(guī)律。它包含了大量的生理信息,疾病信息,是研究癲癇發(fā)作特征的重要工具。
本文是基于形態(tài)成分分析腦電棘波,腦電棘波檢測(cè)選用離散余弦變換作為提取EEG的背景信號(hào)的字典,選用Curvelet變換作為提取癲癇腦電棘波信號(hào)的字典。首先我們要對(duì)癲癇腦電信號(hào)數(shù)據(jù)樣本進(jìn)行特征提取,其提取出的特征向量送入支持向量機(jī),進(jìn)行訓(xùn)練,得到一個(gè)訓(xùn)練模型,利用此模型,進(jìn)行混合腦電信號(hào)的癲癇分析判斷。對(duì)于一組含有癲癇腦電信號(hào)和正常腦電信號(hào)的數(shù)據(jù),用訓(xùn)練出的模型進(jìn)行分析判斷,分出其中癲癇腦電信號(hào)和正常腦電信號(hào),與已知結(jié)果進(jìn)行比較,由此得到模型判斷的準(zhǔn)確率,由此得到離散余弦變換,Curvelet變換分析的準(zhǔn)確率。
關(guān)鍵詞 癲癇腦電 棘波特征提取 Curvelet變換 離散余弦變換 支持向量機(jī)分類(lèi)
Epileptic EEG Spike Detection
Abstract Epilepsy is a common chronic brain diseases,which is serious harm to human health.It is time to attack the performance of the cerebral cortex and subcortical gray matter sudden excessive repetitive discharge,resulting in the patient's body convulsions,changes and abnormal mental awareness,etc.So how to analysis epileptic EEG accurately has important practical significance. EEG (Electroencephalogram,EEG) is a potential signal specific parts of the brain nerve cell populations in spontaneous or induced electrical activity in physiological processes or intracranial specific parts of the surface of the scalp caused,it reflects the rhythm of bioelectric brain activity patterns. It contains a large number of physiological and disease information which is an important tool to study the characteristics of epileptic seizures.
This article is based on morphological analysis of EEG spikes ingredients.EEG Spike Detection chooses discrete cosine transform as the extraction of the background EEG signal dictionary,and choose Curvelet transform component analysis as a form of a dictionary to extract the signal spikes.Samples for feature extraction,which extracts the feature vector into the support vector machines,training,get a training model,the use of this model for the epileptic EEG analysis and judgment.That is,a set of data containing epileptic EEG and normal EEG,with training models to analyze the judgment,in which the separation of epileptic EEG and normal EEG,the results were compared with the known, whereby the model to determine accuracy,thereby obtaining a discrete cosine transform,Curvelet transform accuracy of the analysis.
Keywords Epileptic EEG Spikes feature extraction Curvelet transform discrete cosine transform Support vector machine classification
目 錄
第一章 緒論 1
1.1 癲癇腦電研究背景 1
1.2 癲癇腦電研究目的及意義 2
1.3 本文研究?jī)?nèi)容: 3
第二章 癲癇腦電研究相關(guān)方法概述 4
2.1 癲癇腦電棘波檢測(cè)方法 4
2.2 形態(tài)成分分析的研究現(xiàn)狀 7
第三章 癲癇腦電形態(tài)成分分析原理 9
3.1 形態(tài)成分分析 9
3.1.1信號(hào)的稀疏表示 9
3.1.2形態(tài)成分分析的基本概念以及算法 10
3.1.3快速隱式變換 10
3.1.4形態(tài)成分分解 11
3.2 字典的選擇 12
3.3 離散余弦變換 13
3.3.1 離散余弦變換(DCT)簡(jiǎn)介 13
3.3.2 離散余弦變換(DCT)變換式 13
3.4 Curvelet變換 14
3.4.1 Curvelet變換簡(jiǎn)介 14
3.4.2 Curvelet變換的基本實(shí)驗(yàn)步驟 15
3.5 支持向量機(jī) 16
3.5.1 支持向量機(jī)概述 16
3.5.2 核函數(shù) 17
3.5.3 線性分類(lèi)器 18
3.6 癲癇腦電分析過(guò)程 21
3.6.1 實(shí)驗(yàn)數(shù)據(jù) 21
3.6.2 癲癇腦電分析實(shí)驗(yàn)流程圖 21
3.6.3 癲癇腦電信號(hào)特征提取 22
3.6.4 支持向量機(jī)分類(lèi)及算法 24
第四章 結(jié)果 27
4.1 基于Curvelet變換和離散余弦變換的癲癇腦電棘波檢測(cè) 27
4.1.1仿真結(jié)果與比較 27
第五章 總結(jié) 33
致 謝 34
參考文獻(xiàn) 35
1.6萬(wàn)字
自己原創(chuàng)的畢業(yè)論文,僅在本站獨(dú)家出售,重復(fù)率低,推薦下載使用
摘要 癲癇,是一種生活中篅@<哪圓考膊?,熏F(xiàn)氐匚:α宋頤塹納硤寰竦慕】?。它发醉d氖焙蟣硐治竽云げ慵捌げ閬祿抑釋磐蝗還鵲鬧馗蔥苑諾紓賈虜∪松硤宓某櫬?,茵}兜母謀湟約熬竦囊斐5鵲?。所疫\(yùn)擔(dān)頤竊躚擇拆錟緣縲藕漚蟹淺W既返姆治讎卸暇拖緣煤苤匾簿哂兄匾庖?。木壚_�(Electroencephalogram,EEG)產(chǎn)生于大腦神經(jīng)細(xì)胞群在生理過(guò)程中自發(fā)或誘發(fā)的電活動(dòng),表現(xiàn)為神經(jīng)細(xì)胞群在頭皮表面或者顱內(nèi)引起的電位信號(hào)的變化,它反映了大腦生物電的節(jié)律活動(dòng)規(guī)律。它包含了大量的生理信息,疾病信息,是研究癲癇發(fā)作特征的重要工具。
本文是基于形態(tài)成分分析腦電棘波,腦電棘波檢測(cè)選用離散余弦變換作為提取EEG的背景信號(hào)的字典,選用Curvelet變換作為提取癲癇腦電棘波信號(hào)的字典。首先我們要對(duì)癲癇腦電信號(hào)數(shù)據(jù)樣本進(jìn)行特征提取,其提取出的特征向量送入支持向量機(jī),進(jìn)行訓(xùn)練,得到一個(gè)訓(xùn)練模型,利用此模型,進(jìn)行混合腦電信號(hào)的癲癇分析判斷。對(duì)于一組含有癲癇腦電信號(hào)和正常腦電信號(hào)的數(shù)據(jù),用訓(xùn)練出的模型進(jìn)行分析判斷,分出其中癲癇腦電信號(hào)和正常腦電信號(hào),與已知結(jié)果進(jìn)行比較,由此得到模型判斷的準(zhǔn)確率,由此得到離散余弦變換,Curvelet變換分析的準(zhǔn)確率。
關(guān)鍵詞 癲癇腦電 棘波特征提取 Curvelet變換 離散余弦變換 支持向量機(jī)分類(lèi)
Epileptic EEG Spike Detection
Abstract Epilepsy is a common chronic brain diseases,which is serious harm to human health.It is time to attack the performance of the cerebral cortex and subcortical gray matter sudden excessive repetitive discharge,resulting in the patient's body convulsions,changes and abnormal mental awareness,etc.So how to analysis epileptic EEG accurately has important practical significance. EEG (Electroencephalogram,EEG) is a potential signal specific parts of the brain nerve cell populations in spontaneous or induced electrical activity in physiological processes or intracranial specific parts of the surface of the scalp caused,it reflects the rhythm of bioelectric brain activity patterns. It contains a large number of physiological and disease information which is an important tool to study the characteristics of epileptic seizures.
This article is based on morphological analysis of EEG spikes ingredients.EEG Spike Detection chooses discrete cosine transform as the extraction of the background EEG signal dictionary,and choose Curvelet transform component analysis as a form of a dictionary to extract the signal spikes.Samples for feature extraction,which extracts the feature vector into the support vector machines,training,get a training model,the use of this model for the epileptic EEG analysis and judgment.That is,a set of data containing epileptic EEG and normal EEG,with training models to analyze the judgment,in which the separation of epileptic EEG and normal EEG,the results were compared with the known, whereby the model to determine accuracy,thereby obtaining a discrete cosine transform,Curvelet transform accuracy of the analysis.
Keywords Epileptic EEG Spikes feature extraction Curvelet transform discrete cosine transform Support vector machine classification
目 錄
第一章 緒論 1
1.1 癲癇腦電研究背景 1
1.2 癲癇腦電研究目的及意義 2
1.3 本文研究?jī)?nèi)容: 3
第二章 癲癇腦電研究相關(guān)方法概述 4
2.1 癲癇腦電棘波檢測(cè)方法 4
2.2 形態(tài)成分分析的研究現(xiàn)狀 7
第三章 癲癇腦電形態(tài)成分分析原理 9
3.1 形態(tài)成分分析 9
3.1.1信號(hào)的稀疏表示 9
3.1.2形態(tài)成分分析的基本概念以及算法 10
3.1.3快速隱式變換 10
3.1.4形態(tài)成分分解 11
3.2 字典的選擇 12
3.3 離散余弦變換 13
3.3.1 離散余弦變換(DCT)簡(jiǎn)介 13
3.3.2 離散余弦變換(DCT)變換式 13
3.4 Curvelet變換 14
3.4.1 Curvelet變換簡(jiǎn)介 14
3.4.2 Curvelet變換的基本實(shí)驗(yàn)步驟 15
3.5 支持向量機(jī) 16
3.5.1 支持向量機(jī)概述 16
3.5.2 核函數(shù) 17
3.5.3 線性分類(lèi)器 18
3.6 癲癇腦電分析過(guò)程 21
3.6.1 實(shí)驗(yàn)數(shù)據(jù) 21
3.6.2 癲癇腦電分析實(shí)驗(yàn)流程圖 21
3.6.3 癲癇腦電信號(hào)特征提取 22
3.6.4 支持向量機(jī)分類(lèi)及算法 24
第四章 結(jié)果 27
4.1 基于Curvelet變換和離散余弦變換的癲癇腦電棘波檢測(cè) 27
4.1.1仿真結(jié)果與比較 27
第五章 總結(jié) 33
致 謝 34
參考文獻(xiàn) 35
TA們正在看...
- 新課程背景下高中數(shù)學(xué)學(xué)習(xí)方法.doc
- 家長(zhǎng)會(huì)班主任發(fā)言稿_2.doc
- 新課程視野下在計(jì)算教學(xué)中培養(yǎng)學(xué)生推理能力的實(shí)踐...doc
- 家長(zhǎng)會(huì)班主任發(fā)言稿_3.doc
- 新課程課堂教學(xué)反思的內(nèi)容及方法.doc
- 家長(zhǎng)會(huì)班主任發(fā)言稿優(yōu)秀范文.doc
- 新穎的數(shù)學(xué)課堂才能煥發(fā)生機(jī)與活力.doc
- 家長(zhǎng)會(huì)班主任發(fā)言稿范文.doc
- 新高一學(xué)生數(shù)學(xué)學(xué)習(xí)狀況的調(diào)查報(bào)告.doc
- 家長(zhǎng)會(huì)班主任發(fā)言稿:家長(zhǎng),我拿什么奉獻(xiàn)給您.doc