本文選題：重度抑郁癥 + 稀疏表示��； 參考：《吉林大學(xué)》2017年碩士論文

【摘要】：重度抑郁癥(MDD)是一種常見(jiàn)的精神紊亂疾病。患者呈現(xiàn)長(zhǎng)期過(guò)度傷心、消極情緒及認(rèn)知障礙,甚至自殺,嚴(yán)重影響其身心健康。第二版貝克抑郁量表中,MDD評(píng)分值范圍是29~63。MDD患者語(yǔ)音通常具有單調(diào)、低沉、無(wú)生命力的跡象。同健康者語(yǔ)音相比,抑郁癥語(yǔ)音蘊(yùn)含的音源、系統(tǒng)和韻律信息具有本質(zhì)差別。現(xiàn)有研究表明通過(guò)捕捉聲學(xué)特性變化完全能夠有效檢測(cè)MDD。然而,抑郁癥語(yǔ)料庫(kù)的有限性和分布不平衡等已經(jīng)成為重度抑郁癥檢測(cè)的主要障礙。傳統(tǒng)分類(lèi)模型無(wú)法滿(mǎn)足需要,如支持向量機(jī)(SVM)、高斯混合模型(GMM)和稀疏表示分類(lèi)器(SRC)等。當(dāng)各類(lèi)訓(xùn)練樣本充足時(shí),GMM性能良好。SVM雖然適合小樣本分類(lèi)問(wèn)題,但要求各類(lèi)樣本數(shù)量應(yīng)該平衡。當(dāng)樣本足夠多時(shí),稀疏表示以1l范數(shù)逼近0l范數(shù)的效果要比用2l范數(shù)逼近的效果好,但當(dāng)訓(xùn)練樣本較少且各類(lèi)樣本不平衡時(shí),性能欠佳。此外,1l范數(shù)逼近稀疏表示求解涉及大量迭代,計(jì)算更加復(fù)雜、耗時(shí)較長(zhǎng)。為解決訓(xùn)練數(shù)據(jù)有限性、不平衡和計(jì)算復(fù)雜度高等問(wèn)題,本論文在協(xié)作表示分類(lèi)器(CRC)納入所有類(lèi)樣本字典協(xié)作表示參與分類(lèi)的基礎(chǔ)上,提出修正協(xié)作表示分類(lèi)器(MCRC)。MCRC首先將由原始訓(xùn)練樣本構(gòu)成的字典通過(guò)奇異值分解映射成各類(lèi)平衡的元樣本字典,然后依據(jù)對(duì)測(cè)試樣本的線(xiàn)性表示建立協(xié)作表示模型,同時(shí)施加KL散度(KLD)距離加權(quán)矩陣并約束與特征值相關(guān)的偏差加權(quán)和,實(shí)現(xiàn)適度加大兩類(lèi)殘差距離的目標(biāo),改善分類(lèi)性能�；趪�(guó)內(nèi)外學(xué)術(shù)界認(rèn)可的AVEC2013抑郁癥語(yǔ)料庫(kù)的朗讀語(yǔ)音部分完成性能評(píng)價(jià)實(shí)驗(yàn)。將語(yǔ)料庫(kù)語(yǔ)句按長(zhǎng)度差異分割成三種實(shí)驗(yàn)數(shù)據(jù)集,在留一交叉驗(yàn)證框架下進(jìn)行MDD檢測(cè),對(duì)比SVM、SRC、標(biāo)準(zhǔn)CRC和MCRC的檢測(cè)性能。結(jié)果表明MCRC算法的精確度(Accuracy)最高,且靈敏度(Sensitivity)與特異度(Specificity)更為匹配。此外,因MCRC仍能預(yù)先計(jì)算投影矩陣,大大降低計(jì)算開(kāi)銷(xiāo)。本文主要?jiǎng)?chuàng)新工作如下:(1)將協(xié)作表示思想首次應(yīng)用于語(yǔ)音抑郁癥檢測(cè)。(2)提出修正協(xié)作表示檢測(cè)模型。其優(yōu)點(diǎn):1)適用于小樣本、不平衡樣本的分類(lèi)問(wèn)題;2)分類(lèi)器比較穩(wěn)定,對(duì)正則化參數(shù)變化不敏感;3)受所處理語(yǔ)音段長(zhǎng)度變化影響很小。
[Abstract]:Severe depression (MDD) is a common mental disorder. Patients have long term excessive sadness, negative emotion and cognitive impairment and even suicide, which seriously affect their physical and mental health. In the second edition of Beck depression scale, the range of MDD score is a sign that the phonetics of 29~63.MDD patients are usually monotonous, low and inactive. There are essential differences in the phonetic source, system and prosodic information contained in depression. Existing studies have shown that MDD. can be effectively detected by capturing acoustic characteristics. However, the limited and uneven distribution of depression corpus has become the main obstacle for severe depression detection. Support vector machine (SVM), Gauss mixed model (GMM) and sparse representation classifier (SRC). When all kinds of training samples are sufficient, GMM performance is good.SVM although suitable for small sample classification problem, but the number of samples should be balanced. When the sample is enough, the effect of sparse representation to the 0l norm with the 1L norm is better than the approximation of the 2l norm. The performance is good, but when the training samples are less and the various samples are not balanced, the performance is poor. In addition, the 1L norm approximation sparse representation involves a large number of iterations, and the computation is more complex and time-consuming. In order to solve the problem of limited, unbalanced and computational complexity of training data, this paper introduces the cooperative representation classifier (CRC) into all class sample dictionaries in this paper. On the basis of cooperative representation, a modified cooperative representation classifier (MCRC).MCRC is proposed to map the dictionaries composed of original training samples by singular value decomposition into all kinds of balanced meta sample dictionaries, and then a cooperative representation model is established based on the linear representation of the test samples, and the KL divergence (KLD) distance weighting matrix is applied at the same time. And constrain the deviation weighted sum associated with the eigenvalues, to achieve a moderate increase in the target of two kinds of residual distance and improve the classification performance. Based on the reading speech part of the AVEC2013 depression corpus recognized by the domestic and foreign academics, the performance evaluation experiment is completed. The corpus is divided into three experimental data sets according to the length difference, and a cross test is used. MDD detection under the certificate framework compares the detection performance of SVM, SRC, standard CRC and MCRC. The results show that the accuracy of the MCRC algorithm (Accuracy) is the highest, and the sensitivity (Sensitivity) is more matched with the specificity (Specificity). In addition, the projection matrix can be calculated in advance because MCRC can still be calculated in advance. The main innovations of this paper are as follows: (1) will cooperate with each other. The representation idea is first applied to speech depression detection. (2) a modified cooperative representation detection model is proposed. Its advantages are: 1) suitable for small sample, unbalanced sample classification problem; 2) the classifier is more stable, not sensitive to the regularization parameter change; 3) is affected by the change of the length of the speech segment.

【學(xué)位授予單位】：吉林大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：R749.4;TN912.3

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 文安;;重度抑郁癥大學(xué)生經(jīng)短周期有氧運(yùn)動(dòng)處方干預(yù)后抑郁癥狀和心肺功能的變化[J];中國(guó)組織工程研究與臨床康復(fù);2007年17期

2 沈勤峰,和立;綜合醫(yī)院重度抑郁癥5例漏誤診分析[J];云南醫(yī)藥;2001年05期

3 樂(lè)洪波;馬樹(shù)華;程曉玲;王兆新;胡治國(guó);張慧紅;丘春柳;李惠;張龔巍;廖玲敏;;重度抑郁癥對(duì)負(fù)性情緒面孔的注意偏向[J];中國(guó)心理衛(wèi)生雜志;2009年11期

4 王麗;姚志劍;盧青;劉海燕;陳景旭;卞清濤;;重度抑郁癥癥狀緩解前后海馬局部一致性研究[J];中國(guó)神經(jīng)精神疾病雜志;2010年05期

5 夏訓(xùn)明;;褪黑素類(lèi)似物阿戈美拉汀可有效治療重度抑郁癥[J];廣東藥學(xué)院學(xué)報(bào);2011年03期

6 李洪;卓金春;黃中華;王慕歆;;關(guān)注白領(lǐng)的“心靈感冒”[J];中國(guó)保健營(yíng)養(yǎng);2006年05期

7 張建忠;;2013年9月FDA批準(zhǔn)新藥概況[J];上海醫(yī)藥;2013年21期

8 駱文;;今夜 你失眠了嗎[J];山西青年(新晉商);2007年11期

9 羅紅玲;李雪;;圍絕經(jīng)期婦女重度抑郁癥綜合康復(fù)治療的臨床療效分析[J];中國(guó)臨床新醫(yī)學(xué);2011年09期

10 楊冬林;申連城;李遵清;柳群方;欒清明;錢(qián)麗菊;;重度抑郁癥藥物合并認(rèn)知行為治療的療效及其跟蹤隨訪(fǎng)研究[J];中國(guó)民康醫(yī)學(xué);2014年13期

相關(guān)重要報(bào)紙文章 前3條

1 朱夢(mèng)龍 北京中醫(yī)藥大學(xué);馮世綸經(jīng)方治重度抑郁癥一例[N];中國(guó)中醫(yī)藥報(bào);2014年

2 余志平;2005年美國(guó)臨床藥學(xué)會(huì)年會(huì)熱點(diǎn)追蹤[N];中國(guó)醫(yī)藥報(bào);2006年

3 雷諾島;抗抑郁領(lǐng)域仍待深挖[N];醫(yī)藥經(jīng)濟(jì)報(bào);2012年

相關(guān)博士學(xué)位論文 前2條

1 王慶中;SNAP25、CMYA5、ZNF804A、CDH7基因與精神分裂癥和重度抑郁癥的關(guān)聯(lián)分析研究[D];上海交通大學(xué);2014年

2 賀寬軍;漢族人群中CACNA1C、ITIHs家族基因與精神分裂癥、重度抑郁癥的關(guān)聯(lián)研究[D];上海交通大學(xué);2014年

相關(guān)碩士學(xué)位論文 前10條

1 王靜;重度抑郁癥患者認(rèn)知障礙與磁共振腦結(jié)構(gòu)的相關(guān)性研究[D];鄭州大學(xué);2015年

2 譚春燕;DTI聯(lián)合VBM對(duì)首發(fā)重度抑郁癥治療前后腦結(jié)構(gòu)改變的初步研究[D];昆明醫(yī)科大學(xué);2015年

3 高梅云;青中年重度抑郁癥患者認(rèn)知功能及睡眠特征研究[D];天津醫(yī)科大學(xué);2015年

4 江雪艷;重度抑郁癥功能結(jié)構(gòu)腦半球網(wǎng)絡(luò)研究[D];杭州師范大學(xué);2016年

5 劉澤s，

本文編號(hào)：1786941