北京生物醫(yī)學(xué)工程

基于DSMD和局域特征變換的ECG特征波檢測(cè)_________

Characteristic wave detection in ECG based on DSMD and local feature conversion

作者：張超邱天爽

單位：           大連理工大學(xué)電子信息與電氣工程學(xué)部(遼寧大連116024)    

關(guān)鍵詞：心電信號(hào)；數(shù)學(xué)形態(tài)學(xué)；導(dǎo)數(shù)；局域變換法；特征波提取

分類號(hào)：

出版年·卷·期（頁碼）：2013·32·1（22-27）

摘要：

目的提高心電自動(dòng)診斷中，心電(electrocardiogram，ECG)信號(hào)特征波定位的準(zhǔn)確度。方法在多尺度數(shù)學(xué)形態(tài)學(xué)導(dǎo)數(shù)(multi-scale morphological derivative，MMD)的基礎(chǔ)上提出雙層尺度數(shù)學(xué)形態(tài)學(xué)導(dǎo)數(shù)(double-layer scales morphological derivative，DSMD)的概念，并結(jié)合局域變換法對(duì)心電信號(hào)特征波進(jìn)行檢測(cè)。首先，在對(duì)心電信號(hào)進(jìn)行大尺度DSMD變換基礎(chǔ)上定位R、P、T波波峰位置，然后，在R、P、T波波峰位置確定的前提下，重新選擇尺度，在ECG信號(hào)的小尺度DSMD變換上準(zhǔn)確定位R波、P波和T波邊界點(diǎn)，最后，在R波邊界點(diǎn)確定的前提下運(yùn)用局域變換法檢測(cè)Q、S波，并最終確定QRS波邊界點(diǎn)。結(jié)果 DSMD算法結(jié)合局域變換法對(duì)ECG特征波的提取效果良好。結(jié)論本文提出的DSMD在尺度選擇上更具靈活性，相比于MMD算法顯著提高了檢測(cè)精度,并且達(dá)到量化心電圖通用標(biāo)準(zhǔn)(common standards for quantitative electrocardiography，CSE)。

Objective To improve the detection accuracy of characteristic waves in electrocardiogram (ECG).Methods Based on multi-scale morphological derivative (MMD)，a new modified method，double-layer scales morphological derivative (DSMD) combined with local feature conversion，is presented to detect ECG characteristic wave in this paper.First，DSMD is applied to ECG signal to position R，P，T peaks.Then，on the premise that R，P，T peaks are positioned，the boundaries of R，P，T peaks are accurately positioned on DSMD in a smaller scales than those used in the first step.Finally，the local feature conversion is used to detect whether Q and S waves exist，and eventually determine the boundaries of QRS complex.Results The detection effect of ECG characteristic wave is good by using DSMD algorithm combined with local feature conversion. Conclusions Compared with MMD algorithm，the proposed algorithm is more flexible in the choice of scales，greatly declines the detection error and meets the common standards for quantitative electrocardiography(CSE).

參考文獻(xiàn)：

［1］王大雄, 顧永跟，徐耕，等.T波檢測(cè)和QT間期提取策略［J］.中國生物醫(yī)學(xué)工程學(xué)報(bào)，2011，30(3)：352-356.
Wang Daxiong，Gu Yonggen，Xu Geng，et al.The strategy of T wave detection and QT interval extraction［J］.Chinese Journal of Biomedical Engineering，2011，30(3)：352-356.
［2］王春光，劉金江，孫即祥.基于稀疏分解和模糊理論的心電信號(hào)波形檢測(cè)及識(shí)別［J］.信號(hào)處理,2009,25(7)：1057-1061.
Wang Chunguang，Liu Jinjiang，Su Jixiang.The detection and recognition of electrocardiogram’s waveform based on sparse decomposition and fuzzy theory［J］.Signal Processing,2009,25(7)：1057-1061.
［3］Arafat AK，Hasan.Automatic detection of ECG wave boundaries using empirical mode decomposition ［C］.IEEE ICASSP 2009 Conference,2009：461-464.
［4］王大雄, 蔣云良，徐耕，等.基于LS估計(jì)檢測(cè)QRS波群寬度［J］.中國生物醫(yī)學(xué)工程學(xué)報(bào),2010，29(3)：321-325.
Wang Daxiong，Jiang Yunliang，Xu Geng，et al.The LS estimation-based detection of width for QRS complexes［J］.Chinese Journal of Biomedical Engineering,2010，29(3)：321-325.
［5］Sun Yan, Chan YK，Shankar Muthu Krishnan.Characteristic wave detection in ECG signal using morphological transform ［J］.BMC Cardiovascular Disorders，2005,5(1)：28.
［6］高艷, 胡陽.基于數(shù)學(xué)形態(tài)學(xué)方法的心電圖波形分離技術(shù) ［J］.生物醫(yī)學(xué)工程學(xué)雜志,2001,18(1)：55-59.
Gao Yan，Hu Yang.An ECG waves separation technique based on mathematical morphology［J］.Journal of Biomedical Engineering，2001,18(1)：55-59.
［7］張玉霞.基于MATLAB與WFDB的PhysioBank數(shù)據(jù)庫讀取［J］.北京生物醫(yī)學(xué)工程,2011,30(3)：318-320.
Zhang Yuxia.PhysioBank database reading based on MATLAB and WFDB［J］.Beijing Biomedical Engineering,2011,30(3)：318-320.
［8］Martínez JP，Almeida R，Olmos S，et al.A wavelet-based ECG delineator：evaluation on standard databases［J］.IEEE Trans Biomed Eng，2004，51(4)：570-581.

服務(wù)與反饋：

【文章下載】【加入收藏】

提示：您還未登錄，請(qǐng)登錄！點(diǎn)此登錄

51黑料吃瓜在线观看,51黑料官网|51黑料捷克街头搭讪_51黑料入口最新视频