基于小波系数特征融合的小鼠癫痫脑电分类
|
摘要
采集癫痫小鼠模型在常态与致癫状态下的脑电信号以研究其癫痫脑电的自动分类。对经过噪声和伪迹消除预处理的脑电信号进行小波变换,获得不同频率子带的小波系数,对脑电信号及与癫痫特征波相关的小波系数提取相应的线性特征(标准差)和非线性特征(样本熵);基于这些特征及其组合使用支持向量机分类器实现分类。实验发现基于小鼠脑电本身的标准差和样本熵的分类正确率分别为59.10%和58.00%;而融合各相关小波系数的标准差或样本熵,分类正确率分别达到86.60%和88.60%;融合全部相关小波系数的线性和非线性特征后分类正确率为99.80%。这些结果说明基于小波系数特征融合的分类算法性能有显著提升,能有效实现小鼠癫痫脑电的自动分类。
The Electroencephalogram(EEG)of mouse model of epilepsy in normal and epileptic status is collected to study the automatic classification of epileptic EEG. The noise-and artifact-attenuated EEG is wavelet-transformed, and the linear feature(standard deviation)and the nonlinear feature(sample entropy)are then extracted for the EEG signals and those wavelet coefficients related to the characteristic waveforms of epileptic EEG. Classification is implemented using support vector machine with above individual features and their combinations. The classification accuracy based on the standard deviation and sample entropy of EEG signals are 59.1% and 58.00%, respectively.The accuracy increases to86.60% or 88.60%, when the standard deviations or sample entropies of relevant wavelet coefficients are used as input features.After combining both types of features, the classification accuracy is 99.80%. These results show that wavelet coefficient features fusion significantly improves the classification accuracy, achieving effective classification of mouse epileptic EEG.
The Electroencephalogram(EEG)of mouse model of epilepsy in normal and epileptic status is collected to study the automatic classification of epileptic EEG. The noise-and artifact-attenuated EEG is wavelet-transformed, and the linear feature(standard deviation)and the nonlinear feature(sample entropy)are then extracted for the EEG signals and those wavelet coefficients related to the characteristic waveforms of epileptic EEG. Classification is implemented using support vector machine with above individual features and their combinations. The classification accuracy based on the standard deviation and sample entropy of EEG signals are 59.1% and 58.00%, respectively.The accuracy increases to86.60% or 88.60%, when the standard deviations or sample entropies of relevant wavelet coefficients are used as input features.After combining both types of features, the classification accuracy is 99.80%. These results show that wavelet coefficient features fusion significantly improves the classification accuracy, achieving effective classification of mouse epileptic EEG.
引文
[1]Qing W H,Sidra T,Afzal A M,et al.Recording EEGin freely moving neonatal rats using a novel method[J].Journal of Visualized Experiments,2017,123.
[2]Sukhvir W,Kvan H,Lukasz S,et al.Epileptogenic effects of NMDAR antibodies in a passive transfer mouse model[J].Brain,2015,138(11):3159-3167.
[3]Quang M T,Ashwin A,Min C,et al.Mouse epileptic seizure detection with multiple EEG features and simple thresholding technique[J].Journal of Neural Engineering,2017.
[4]陈小千,封洲燕,郭哲彬,等.癫痫模型中痫样棘波自动检测的新算法及其应用[J].生物医学工程学杂志,2017,34(4):485-492.
[5]张永宁,肖素希,肖文,等.癫痫动物模型的研究进展[J].科教导刊,2017,19(7):157-158.
[6]Muhammad I N,Li S,Myeong O K.Maternal epileptic seizure induced by pentylenetetrazol:apoptotic neurodegeneration and decreased GABAB1receptor expression in prenatal rat brain[J].Molecular Brain,2009,2(1):1-11.
[7]张瑞,宋江玲,胡文凤.癫痫脑电的特征提取方法综述[J].西北大学学报(自然科学版),2016,46(6):781-788.
[8]Jahankhani P,Kodogiannis V,Revett K.EEG signal classification using wavelet feature extraction and neural networks[C]//IEEE John Vincent Atanasoff 2006 International Symposium on Modern Computing(JVA’06),2006:52-57.
[9]贺王鹏,杨琳,王芳,等.基于TQWT的癫痫脑电信号的识别[J].生物医学工程研究,2017,36(4):346-350.
[10]李冬梅,张洋,杨日东,等.经验模式分解与代价敏感支持向量机在癫痫脑电信号分类中的应用[J].生物医学工程研究,2017,36(1):33-37.
[11]夏德玲,孟庆芳,牛贺功,等.基于Lempel-Ziv复杂度和经验模态分解的癫痫脑电信号的检测方法[J].计算机物理,2015,32(6):709-714.
[12]张健钊,姜威,元辉,等.基于离散S变换和排列熵的癫痫脑电识别[J].生物医学工程学杂志,2017,34(5):681-686.
[13]蔡冬梅,周卫东,刘凯,等.基于Hurst指数和SVM的癫痫脑电检测方法[J].中国生物医学工程学报,2010,29(6):836-840.
[14]吴晓彬,郭影,邱天爽,等.B样条小波在提取脑电癫痫棘波中的应用[J].生物医学工程学杂志,2006,23(5):947-950.
[15]师梨,陈明静.基于小波包分解和遗传神经网络的癫痫识别[J].计算机工程与应用,2010,46(12):218-220.
[16]崔刚强,夏良斌,梁建峰,等.基于小波多尺度分析和极限学习机的癫痫脑电分类算法[J].生物医学工程学杂志,2016,33(6):1025-1038.
[17]徐长发,李国宽.实用小波方法[M].3版.武汉:华中科技大学出版社,2009:55-163.
[18]Janjarasjitt S.Classification of the epileptic EEGs using the wavelet-based scale variance feature[J].International Journal of Applied Biomedical Engineering,2010,3(1):19-25.
[19]Richman J S,Moorman J R.Physiological time-series analysis using approximate entropy and sample entropy[J].Am J Phisiol Heart Circ Physiol,2000,278(6).
[20]张健钊,姜威,贲晛烨.基于样本熵与小波包能量特征提取和Real AdaBoot算法的正常期、癫痫间歇与发作期的脑电自动检测[J].生物医学工程学杂志,2016,33(6):1031-1038.
[21]杨松,于凤芹.基于样本熵的语音/音乐识别[J].计算机工程与应用,2012,48(23):125-127.
[22]周志华.机器学习[M].北京:清华大学出版社,2015:121-145.
[23]奉国和.SVM分类核函数及参数选择比较[J].计算机工程与应用,2011,47(3):123-124.
[24]李明阳,陈万忠,张涛.基于DD-DWT和Log-Logistic参数回归的癫痫脑电自动识别[J].仪器仪表学报,2017,38(6):1368-1377.
[25]Martis R J,Tan J H,Chua C K,et al.Epileptic EEGclassification using nonlinear parameters on different frequency bands[J].Journal of Mechanics in Medicine and Biology,2015,15(3):1-20.
[2]Sukhvir W,Kvan H,Lukasz S,et al.Epileptogenic effects of NMDAR antibodies in a passive transfer mouse model[J].Brain,2015,138(11):3159-3167.
[3]Quang M T,Ashwin A,Min C,et al.Mouse epileptic seizure detection with multiple EEG features and simple thresholding technique[J].Journal of Neural Engineering,2017.
[4]陈小千,封洲燕,郭哲彬,等.癫痫模型中痫样棘波自动检测的新算法及其应用[J].生物医学工程学杂志,2017,34(4):485-492.
[5]张永宁,肖素希,肖文,等.癫痫动物模型的研究进展[J].科教导刊,2017,19(7):157-158.
[6]Muhammad I N,Li S,Myeong O K.Maternal epileptic seizure induced by pentylenetetrazol:apoptotic neurodegeneration and decreased GABAB1receptor expression in prenatal rat brain[J].Molecular Brain,2009,2(1):1-11.
[7]张瑞,宋江玲,胡文凤.癫痫脑电的特征提取方法综述[J].西北大学学报(自然科学版),2016,46(6):781-788.
[8]Jahankhani P,Kodogiannis V,Revett K.EEG signal classification using wavelet feature extraction and neural networks[C]//IEEE John Vincent Atanasoff 2006 International Symposium on Modern Computing(JVA’06),2006:52-57.
[9]贺王鹏,杨琳,王芳,等.基于TQWT的癫痫脑电信号的识别[J].生物医学工程研究,2017,36(4):346-350.
[10]李冬梅,张洋,杨日东,等.经验模式分解与代价敏感支持向量机在癫痫脑电信号分类中的应用[J].生物医学工程研究,2017,36(1):33-37.
[11]夏德玲,孟庆芳,牛贺功,等.基于Lempel-Ziv复杂度和经验模态分解的癫痫脑电信号的检测方法[J].计算机物理,2015,32(6):709-714.
[12]张健钊,姜威,元辉,等.基于离散S变换和排列熵的癫痫脑电识别[J].生物医学工程学杂志,2017,34(5):681-686.
[13]蔡冬梅,周卫东,刘凯,等.基于Hurst指数和SVM的癫痫脑电检测方法[J].中国生物医学工程学报,2010,29(6):836-840.
[14]吴晓彬,郭影,邱天爽,等.B样条小波在提取脑电癫痫棘波中的应用[J].生物医学工程学杂志,2006,23(5):947-950.
[15]师梨,陈明静.基于小波包分解和遗传神经网络的癫痫识别[J].计算机工程与应用,2010,46(12):218-220.
[16]崔刚强,夏良斌,梁建峰,等.基于小波多尺度分析和极限学习机的癫痫脑电分类算法[J].生物医学工程学杂志,2016,33(6):1025-1038.
[17]徐长发,李国宽.实用小波方法[M].3版.武汉:华中科技大学出版社,2009:55-163.
[18]Janjarasjitt S.Classification of the epileptic EEGs using the wavelet-based scale variance feature[J].International Journal of Applied Biomedical Engineering,2010,3(1):19-25.
[19]Richman J S,Moorman J R.Physiological time-series analysis using approximate entropy and sample entropy[J].Am J Phisiol Heart Circ Physiol,2000,278(6).
[20]张健钊,姜威,贲晛烨.基于样本熵与小波包能量特征提取和Real AdaBoot算法的正常期、癫痫间歇与发作期的脑电自动检测[J].生物医学工程学杂志,2016,33(6):1031-1038.
[21]杨松,于凤芹.基于样本熵的语音/音乐识别[J].计算机工程与应用,2012,48(23):125-127.
[22]周志华.机器学习[M].北京:清华大学出版社,2015:121-145.
[23]奉国和.SVM分类核函数及参数选择比较[J].计算机工程与应用,2011,47(3):123-124.
[24]李明阳,陈万忠,张涛.基于DD-DWT和Log-Logistic参数回归的癫痫脑电自动识别[J].仪器仪表学报,2017,38(6):1368-1377.
[25]Martis R J,Tan J H,Chua C K,et al.Epileptic EEGclassification using nonlinear parameters on different frequency bands[J].Journal of Mechanics in Medicine and Biology,2015,15(3):1-20.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |