基于多普勒超声信号的脉象分析与分类研究
|
摘要
在传统中医学中，脉诊已经成功沿用几千年，中医大夫利用手指感觉人体手腕脉搏的跳动，脉搏信号中包含了能够反映人体健康程度的重要信息，中医大夫能够利用这些信息判断患者的身体状况。然而，传统中医脉诊方法受到医生临床经验和主观判断等局限。因此，迫切需要发展一种标准化、客观化脉诊方法。近年来，随着计算机和电子技术的发展，已有大量的研究人员利用不同现代化的设备和方法，对脉诊客观化进行了深入研究并取得了初步的研究成果，但仍然还存在诸多问题尚待解决。
针对脉诊客观化研究，本文利用经颅多普勒超声设备获取脉搏信号，即，多普勒超声脉搏信号，分别开展多普勒超声脉搏信号质量实时评价、特征提取和分类方法等方面的脉诊客观化研究。
在脉诊的客观化的研究中，如何准确有效的获取高质量的脉搏信号是最关键的问题。然而，现有脉搏信号采集设备不具备信号质量自动评价功能，导致大量临床采集的脉搏信号不能被及时准确应用于各类辅助诊断。因此，本文提出基于二阶差分样本熵的多普勒超声脉搏信号质量实时评价方法。该方法利用小波变换消除多普勒超声脉搏信号的高频噪声和低频基线漂移后，经过二阶差分和样本熵计算，实现多普勒超声脉搏信号质量实时评价。实际采集数据测试结果说明，该方法能够减少采集因素引起的样本损失，提高获取高质量多普勒超声脉搏信号的效率，具备对多普勒超声脉搏信号的质量评估准确，同时兼顾高效快速的特点。
为了充分利用多普勒超声脉搏信号对疾病进行分析与分类研究，本文提出了基于多尺度样本熵分析方法的多普勒超声脉搏信号分类方法，基于TWED距离的多普勒超声脉搏信号的分类方法和基于多核学习整合多普勒超声脉搏信号多类特征分类方法。
在基于多尺度样本熵分析多普勒超声脉搏信号研究中，根据不同的尺度因子和样本熵方法中的参数m和r，计算多普勒超声脉搏信号多尺度样本熵值，构成一个多维的特征向量。鉴于所构成的多维特征向量中，存在大量特征信息冗余，本文采用多重线性子空间学习的方法，从大量特征中提取有效信息，剔除冗余信息，约减特征向量维数。将降维后的多尺度样本熵特征向量输入到SVM分类器，对健康人和病人的多普勒超声脉搏信号进行分类诊断实验，实验结果表明该分类方法对疾病具有较好的诊断效果。
针对一些相似性度量算法是非弹性的或是弹性非度量的，在计算两信号相似性度量距离时并没有考虑时间轴上所要付出的代价，导致其计算到的两信号相似性度量距离不准确，从而会降低在分类诊断中的精度问题。本文提出基于TWED距离多普勒超声脉搏信号的分类方法，该方法通过计算两时间序列距离时增加一个stiffness参数来控制在时域上的弹性度量，从而在计算多普勒超声脉搏信号相似性度量距离时考虑时间轴上所要付出的代价，提高了计算多普勒超声脉搏信号相似性度量距离的准确性。将该方法与1NN分类器结合，利用多普勒超声脉搏信号和压力脉形信号，对健康人和四类疾病病人进行分类诊断实验。实验结果表明在多普勒超声脉搏信号和压力脉形信号分类中TWED方法具有更优越的分类性能。
基于多核学习整合多普勒超声脉搏信号多类特征分类方法，对多普勒超声脉搏信号每类特征选择恰当的核函数，利用SimpleMKL多核学习方法优化并整合多普勒超声脉搏信号特征，删除多普勒超声脉搏信号冗余的特征，充分利用有价值的特征。实验结果说明该方法能够提高多普勒超声脉搏信号的信息量，从而提高其分类精度和分类速度，同时使得算法具备很好的稳定性。
最后，利用多普勒超声脉搏信号对健康人和一些患有与血流速度和血液粘稠度等因素相关疾病病人进行诊断研究，说明多普勒超声脉搏信号对于疾病诊断分类的意义。
In traditional Chinese medicine (TCM) pulse diagnosis has been successfully usedfor thousands of years. In traditional Chinese pulse diagnosis, doctors put theirfingers on human wrist to feel pulse. In view of pulse signal containing importantinformation of human organ, the doctor could use this information to determine thepatient's physical condition. But the accuracy of pulse diagnosis completely dependson the experience and skills of the doctor. In the diagnosis of the same patientdifferent doctors maybe give different diagnostic results. Therefore it is urgent todevelop a standardized and objective method for pulse diagnosis. In recent years,although a lot of researchers objectively make research for pulse diagnosis usingmodern equipments and methods and get some research, there still are manyproblems to be solved.
In the paper, pulse signals are collected by ultrasound equipment that thecollected signals are called as Doppler ultrasound pulse signals. We will focus onquality evaluation, feature extraction and classification methods of Dopplerultrasound pulse signal for the objective study of pulse diagnosis.
In the objective study of pulse diagnosis, how to accurately and effectivelyacquire pulse signal is the most critical work. However, in current acquisition deviceof pulse signal it is not with the function of quality evaluation, which leads to alarge number of low quality of pulse signals are collected for clinical diseasediagnosis. Thus second-order differential sample entropy-based a real-time qualityevaluation mothed of signal is proposed for Doppler ultrasound pulse signal in thispaper. First, the wavelet transform is adopted for removing the high frequency noiseand low-frequency baseline wander of Doppler ultrasound pulse signals. Then, bycalculating the second-order differential and sample entropy of Doppler ultrasoundpulse signal the real-time evaluation method is carried out. The results show thatthis method could reduce the loss of sample and efficiently acquire high-qualityDoppler ultrasound pulse signals.
In order to make full use of Doppler ultrasound pulse signal for the analysis andclassification of disease, in this paper multiscale sample entropy-based method,TWED distance-based method and multiple kernel learning-based method ofcombination of heterogeneous features for Doppler ultrasound pulse signalclassification are respectively proposed for the analysis of Doppler ultrasound pulsesignal.
In the classification research of multiscale sample entropy-based method first thevalue of sample entropy is calculated with different parameters, i.e., τ, m and r, that the calculated value could constitute a multidimensional vector. In virtue of theconstructed redundant multidimensional vector we proposed multiple linearsubspace learning methods to extract useful information and eliminate redundantinformation from multidimensional vector. The vector with dimensionalityreduction is input into the SVM classifier for the classification of healthy people andpatients. Experimental results show that the proposed method is efficient indiagnosis of disease.
In the classification research of TWED distance-based mothed we proposed aclassification method with non-feature extraction that combined with1NN classifier1NN-TWED classifier is constructed for the classification of Doppler ultrasoundpulse signal. In the classification method it introduces a stiffness parameter tocontrol the elasticity of the metric in the time domain, and thus it is more flexiblefor Doppler ultrasound pulse signal matching and classifying.
In the classification research of multiple kernel learning-based method ofcombination of heterogeneous features for Doppler ultrasound pulse signalclassification, first it respectively selects the appropriate kernel function forheterogeneous features of Doppler ultrasound pulse signal. Then SimpleMKL isadopted to integrate the features of Doppler ultrasound pulse signal that deletesredundancy features and make use of the valuable features for signal classification.The experimental results show that in virue of the enhanced information theproposed method could increase the classification accuracy and classification speed.
Finally, by adopting Doppler ultrasound pulse signal for the diagnosis of thedisease related with flow velocity and blood viscosity the classification results showthe performance of Doppler ultrasound pulse signal in disease diagnosis.
针对脉诊客观化研究，本文利用经颅多普勒超声设备获取脉搏信号，即，多普勒超声脉搏信号，分别开展多普勒超声脉搏信号质量实时评价、特征提取和分类方法等方面的脉诊客观化研究。
在脉诊的客观化的研究中，如何准确有效的获取高质量的脉搏信号是最关键的问题。然而，现有脉搏信号采集设备不具备信号质量自动评价功能，导致大量临床采集的脉搏信号不能被及时准确应用于各类辅助诊断。因此，本文提出基于二阶差分样本熵的多普勒超声脉搏信号质量实时评价方法。该方法利用小波变换消除多普勒超声脉搏信号的高频噪声和低频基线漂移后，经过二阶差分和样本熵计算，实现多普勒超声脉搏信号质量实时评价。实际采集数据测试结果说明，该方法能够减少采集因素引起的样本损失，提高获取高质量多普勒超声脉搏信号的效率，具备对多普勒超声脉搏信号的质量评估准确，同时兼顾高效快速的特点。
为了充分利用多普勒超声脉搏信号对疾病进行分析与分类研究，本文提出了基于多尺度样本熵分析方法的多普勒超声脉搏信号分类方法，基于TWED距离的多普勒超声脉搏信号的分类方法和基于多核学习整合多普勒超声脉搏信号多类特征分类方法。
在基于多尺度样本熵分析多普勒超声脉搏信号研究中，根据不同的尺度因子和样本熵方法中的参数m和r，计算多普勒超声脉搏信号多尺度样本熵值，构成一个多维的特征向量。鉴于所构成的多维特征向量中，存在大量特征信息冗余，本文采用多重线性子空间学习的方法，从大量特征中提取有效信息，剔除冗余信息，约减特征向量维数。将降维后的多尺度样本熵特征向量输入到SVM分类器，对健康人和病人的多普勒超声脉搏信号进行分类诊断实验，实验结果表明该分类方法对疾病具有较好的诊断效果。
针对一些相似性度量算法是非弹性的或是弹性非度量的，在计算两信号相似性度量距离时并没有考虑时间轴上所要付出的代价，导致其计算到的两信号相似性度量距离不准确，从而会降低在分类诊断中的精度问题。本文提出基于TWED距离多普勒超声脉搏信号的分类方法，该方法通过计算两时间序列距离时增加一个stiffness参数来控制在时域上的弹性度量，从而在计算多普勒超声脉搏信号相似性度量距离时考虑时间轴上所要付出的代价，提高了计算多普勒超声脉搏信号相似性度量距离的准确性。将该方法与1NN分类器结合，利用多普勒超声脉搏信号和压力脉形信号，对健康人和四类疾病病人进行分类诊断实验。实验结果表明在多普勒超声脉搏信号和压力脉形信号分类中TWED方法具有更优越的分类性能。
基于多核学习整合多普勒超声脉搏信号多类特征分类方法，对多普勒超声脉搏信号每类特征选择恰当的核函数，利用SimpleMKL多核学习方法优化并整合多普勒超声脉搏信号特征，删除多普勒超声脉搏信号冗余的特征，充分利用有价值的特征。实验结果说明该方法能够提高多普勒超声脉搏信号的信息量，从而提高其分类精度和分类速度，同时使得算法具备很好的稳定性。
最后，利用多普勒超声脉搏信号对健康人和一些患有与血流速度和血液粘稠度等因素相关疾病病人进行诊断研究，说明多普勒超声脉搏信号对于疾病诊断分类的意义。
In traditional Chinese medicine (TCM) pulse diagnosis has been successfully usedfor thousands of years. In traditional Chinese pulse diagnosis, doctors put theirfingers on human wrist to feel pulse. In view of pulse signal containing importantinformation of human organ, the doctor could use this information to determine thepatient's physical condition. But the accuracy of pulse diagnosis completely dependson the experience and skills of the doctor. In the diagnosis of the same patientdifferent doctors maybe give different diagnostic results. Therefore it is urgent todevelop a standardized and objective method for pulse diagnosis. In recent years,although a lot of researchers objectively make research for pulse diagnosis usingmodern equipments and methods and get some research, there still are manyproblems to be solved.
In the paper, pulse signals are collected by ultrasound equipment that thecollected signals are called as Doppler ultrasound pulse signals. We will focus onquality evaluation, feature extraction and classification methods of Dopplerultrasound pulse signal for the objective study of pulse diagnosis.
In the objective study of pulse diagnosis, how to accurately and effectivelyacquire pulse signal is the most critical work. However, in current acquisition deviceof pulse signal it is not with the function of quality evaluation, which leads to alarge number of low quality of pulse signals are collected for clinical diseasediagnosis. Thus second-order differential sample entropy-based a real-time qualityevaluation mothed of signal is proposed for Doppler ultrasound pulse signal in thispaper. First, the wavelet transform is adopted for removing the high frequency noiseand low-frequency baseline wander of Doppler ultrasound pulse signals. Then, bycalculating the second-order differential and sample entropy of Doppler ultrasoundpulse signal the real-time evaluation method is carried out. The results show thatthis method could reduce the loss of sample and efficiently acquire high-qualityDoppler ultrasound pulse signals.
In order to make full use of Doppler ultrasound pulse signal for the analysis andclassification of disease, in this paper multiscale sample entropy-based method,TWED distance-based method and multiple kernel learning-based method ofcombination of heterogeneous features for Doppler ultrasound pulse signalclassification are respectively proposed for the analysis of Doppler ultrasound pulsesignal.
In the classification research of multiscale sample entropy-based method first thevalue of sample entropy is calculated with different parameters, i.e., τ, m and r, that the calculated value could constitute a multidimensional vector. In virtue of theconstructed redundant multidimensional vector we proposed multiple linearsubspace learning methods to extract useful information and eliminate redundantinformation from multidimensional vector. The vector with dimensionalityreduction is input into the SVM classifier for the classification of healthy people andpatients. Experimental results show that the proposed method is efficient indiagnosis of disease.
In the classification research of TWED distance-based mothed we proposed aclassification method with non-feature extraction that combined with1NN classifier1NN-TWED classifier is constructed for the classification of Doppler ultrasoundpulse signal. In the classification method it introduces a stiffness parameter tocontrol the elasticity of the metric in the time domain, and thus it is more flexiblefor Doppler ultrasound pulse signal matching and classifying.
In the classification research of multiple kernel learning-based method ofcombination of heterogeneous features for Doppler ultrasound pulse signalclassification, first it respectively selects the appropriate kernel function forheterogeneous features of Doppler ultrasound pulse signal. Then SimpleMKL isadopted to integrate the features of Doppler ultrasound pulse signal that deletesredundancy features and make use of the valuable features for signal classification.The experimental results show that in virue of the enhanced information theproposed method could increase the classification accuracy and classification speed.
Finally, by adopting Doppler ultrasound pulse signal for the diagnosis of thedisease related with flow velocity and blood viscosity the classification results showthe performance of Doppler ultrasound pulse signal in disease diagnosis.
引文
[1]赵恩俭.中医脉诊学[M]:天津科学技术出版社.1999
[2] Lee C T,Wei L Y. Spectrum analysis of human pulse[J]. IEEE Transactions onBiomedical Engineering,1983, BME-30(6):348-352.
[3]李时珍[明].濒湖脉学白话解[M]:人民卫生出版社.1978
[4] Chen Y, Zhang L, Zhang D. Wrist pulse signal diagnosis using modifiedGaussian models and Fuzzy C-Means classification[J]. Medical engineering&physics,2009,31(10):1283-1289.
[5] Chen Y, Zhang L, Zhang D. Computerized wrist pulse signal diagnosis usingmodified auto-regressive models[J]. Journal of Medical Systems,2011,35(3):321-328.
[6]傅骢远,牛欣.中医脉象今释[M]:华夏出版社.1993
[7]费兆馥.现代中医脉诊学[M]:人民卫生出版社.2003
[8]李凌,王志中.脉搏波变异型及其心率变异性关系研究[D].上海交通大学博士论文,2000.
[9]黄世林,孙明异.中医脉象研究[M]:人民卫生出版社.1995
[10] Lu W A, Wang L Y Y, Wang W K. Pulse analysis of patients with severe liverproblems. Studying pulse spectrums to determine the effects on otherorgans[J]. Engineering in Medicine and Biology Magazine, IEEE,1999,18(1):73-75.
[11] Yoon S H, Koga Y, Matsumoto I, et al. An objective method of pulsediagnosis[J]. The American journal of Chinese medicine,1987,15(3-4):147-153.
[12] Broffman M,McCulloch M. Instrument-assisted pulse evaluation in theacupuncture practice[J]. American Journal of Acupuncture,1986,14(3):255-259.
[13] Yoon Y-Z, Lee M-H, Soh K-S. Pulse type classification by varying contactpressure[J]. Engineering in Medicine and Biology Magazine, IEEE,2000,19(6):106-110.
[14] Seng H. Objectifying of pulse-taking[J]. Japanese Journal of OrientalMedicine,1977,27(4):7.
[15] Wang W K, Chen H L, Hsu T L, et al. Alteration of pulse in human subjects bythree Chinese herbs[J]. The American journal of Chinese medicine,1994,22(2):197-203.
[16] Mahomed F. The physiology and clinical use of the sphygmograph[J]. MedTimes Gazette,1872,1:62-64.
[17]汤伟昌,孙汉钧. ZMH-Ⅰ型中医脉象传感器的研究设计[J].传感器技术,1990,1:22-26.
[18]陈冬志,牛欣.中医脉诊实验教学模型的建立与评价[D].北京中医药大学博士研究生论文,2006.
[19]袁玄义,王永昌,张延曹.多普勒效应及其现代应用[J].现代物理知识,2003,12(2):123-126.
[20]柳文仪,赖仲平.彩色脉冲多普勒超声仪对高血压病弦脉与非弦脉的研究[J].中医杂志,1993,34(11):684-685.
[21]牛欣,傅骢远.脉诊位、数、形、势变化的心血管生理学探讨[J].北京中医学院学报,1992,115(1):30-33.
[22] Aritomo M, Yonezawa Y, Caldwell W M. A wrist-mounted activity and pulserecording system: Proceedings of the First Joint BMES/EMBS ConferenceServing Humanity[C]. IEEE,1999:693
[23]徐礼胜.中医脉象数字特征研究[D].哈尔滨工业大学博士论文,2006.
[24]郭庆丽.胆囊炎和肾病综合症脉象信号的特征提取与分类研究[D].哈尔滨工业大学硕士论文,2008.
[25]上海中医学院生理教研室.对几种脉象的初步探讨[J].上海中医药杂志,1979,10(3):43-45.
[26]北京中医学院生理教研室.弦、滑、细等脉象的脉搏图特征初探[J].上海中医药杂志,1980,5(2):8-11.
[27]张卫建.涩脉的脉搏图特征与心血管功能状态[J].中国医药学报,1988,3(2):19-21.
[28]杨天权,郑秀春,孙谨等.典型相关分析在脉图研究中的应用[J].浙江中西医结合杂志,1997,7(1):5-6.
[29]傅骢远.脉诊浮、沉、虚、实的客观指标探讨[J].中西医结合杂志,1990,10(10):603.
[30]李冰星,颜文明.气虚、阳虚患者脉图参数多元逐步回归分析[J].湖南中医学院学报,1990,4:213-216.
[31] Zheng G, Huang Q, Yan G, et al. Pulse waveform key point recognition bywavelet transform for central aortic blood pressure estimation[J]. Journal ofInformation and Computational Science,2012,9(1):25-33.
[32] Nawsupe G,Joshi R R. Modified wavelet-based technique for baseline driftremoval and diagnostic scope of spectral energy of radial pulse signal[J].International Journal of Biomedical Engineering and Technology,2011,6(1):1-13.
[33]刘广斌,樊小力,黄洛秀.两种滑脉患者与健康妇女指端容积脉搏波的功率谱分析[J].西安医科大学学报,1990,11(3):199-201.
[34]王炳和,杨颙.脉搏声信号检测系统实验设计及功率谱特征[J].中华物理医学杂志,1998,39(8):117-120.
[35] Guo Q-L, Wang K-Q, Zhang D-Y, et al. A wavelet packet based pulsewaveform analysis for cholecystitis and nephrotic syndrome diagnosis:Wavelet Analysis and Pattern Recognition,2008. ICWAPR'08. InternationalConference on[C]. IEEE,2008:513-517.
[36] Zhang D Y, Zuo W M, Zhang D, et al. Wrist blood flow signal-basedcomputerized pulse diagnosis using spatial and spectrum features[J]. Journalof Biomedical Science and Engineering,2010,3(4):361-366.
[37] Thakker B, Vyas A L, Tripathi D M. Complexity analysis of wrist pulsesignals[J]. International Journal of Biomedical Engineering and Technology,2012,9(2):188-199.
[38] Rauchberger I, Cohen M, Dart A. Analysis of the diurnal variation of arterialpressure waves: Bioelectromagnetism,1998. Proceedings of the2ndInternational Conference on[C]. IEEE,1998:185-186.
[39] Wang L, Wang K, Xu L. Lempel-ziv decomposition based arrhythmic pulsesrecognition: Engineering in Medicine and Biology Society,2005. IEEE-EMBS2005.27th Annual International Conference of the[C]. IEEE,2006:4606-4609.
[40] Wang K, Xu L, Li Z, et al. Approximate entropy based pulse variabilityanalysis: Computer-Based Medical Systems,2003. Proceedings.16th IEEESymposium[C]. IEEE,2003:236-241.
[41] Maniwa Y, Iokibe T, Koyama M, et al. The application of pulse wave chaos inclinical medicine: Proc.17th FUZY System Symposium, Chiba[C].2001:787-790.
[42] Naschitz J, Itzhak R, Shaviv N, et al. Assessment of cardiovascular reactivityby fractal and recurrence quantification analysis of heart rate and pulse transittime[J]. Journal of human hypertension,2003,17(2):111-118.
[43] Andreadis I I, Giannakakis G A, Papageorgiou C, et al. Detecting complexityabnormalities in dyslexia measuring approximate entropy ofelectroencephalographic signals: Engineering in Medicine and BiologySociety,2009. EMBC2009. Annual International Conference of the IEEE[C].IEEE,2009:6292-6295.
[44] álvarez D, Hornero R, García M, et al. Cross approximate entropy analysis ofnocturnal oximetry signals in the diagnosis of the obstructive sleep apneasyndrome: Engineering in Medicine and Biology Society,2006. EMBS'06.28th Annual International Conference of the IEEE[C]. IEEE,2006:6149-6152.
[45]陈荣山,陈东汉.模糊数学在中医脉象模式识别中应用[J].医学信息,1998,3:20-21.
[46] Lee H-L, Suzuki S, Adachi Y, et al. Fuzzy theory in traditional Chinese pulsediagnosis: Neural Networks,1993. IJCNN'93-Nagoya. Proceedings of1993International Joint Conference on[C]. IEEE,1993:774-777.
[47] Bratteli C, Finkelstein S, Alinder C, et al. Analysis of aging effects on thearterial pulse contour using an artificial neural network: Engineering inMedicine and Biology Society,1998. Proceedings of the20th AnnualInternational Conference of the IEEE[C]. IEEE,1998:1333-1336.
[48] Murthy V S, Ramamoorthy S, Srinivasan N, et al. Analysis ofphotoplethysmographic signals of cardiovascular patterns: Proceedings of the23rd Annual EMBS International Conference[C]. Istanbul, Turkey:2001:2204-2207.
[49]胡家宁,阎述池.脉象人工神经网络分析系统模型[J].中国医科大学学报,1997,26(2):134-137.
[50] Allen J,Murray A. Comparison of three arterial pulse waveform classificationtechniques[J]. Journal of medical engineering&technology,1996,20(3):109-114.
[51] Wang L, Wang K-Q, Xu L-S. Recognizing wrist pulse waveforms withimproved dynamic time warping algorithm: Machine Learning andCybernetics,2004. Proceedings of2004International Conference on[C]. IEEE,2004:3644-3649.
[52] Zhang D, Zuo W, Zhang D, et al. Time series classification using supportvector machine with Gaussian elastic metric kernel:201020th InternationalConference on Pattern Recognition, ICPR2010[C]. Istanbul, Turkey:Instituteof Electrical and Electronics Engineers Inc.,2010:29-32.
[53] Zhang A,Yang F. Study on recognition of sub-health from pulse signal: NeuralNetworks and Brain,2005. ICNN&B'05. International Conference on[C].IEEE,2005:1516-1518.
[54] Richard J P,Kenneth W H. Pulse wave changes with aging[J]. Journal of theAmerican Geriatrics Society,1967,15(2):153-165.
[55] Lawrence M, Resnick, Militianu D, Cunnings A J, et al. Pulse waveformanalysis of arterial compliance: relation to other techniques, age, andmetabolic variables[J]. American journal of hypertension,2000,13(12):1243-1249.
[56] Macdougall J D, Brittain M, MacDonald J, et al. Validity of predicting meanarterial blood pressure during exercise[J]. Medicine and science in sports andexercise,1999,31(12):1876.
[57]盖宇清,田广庆.歼击机与运输机飞行员某些心功能参数的脉图分析[J].航天医学与医学工程,1997,10(5):349-352.
[58]吕萍,闵一建.运动前后脉搏波谱变化的分析研究[J].体育学刊,1997,1:22-23.
[59] Eckert S,Horstkotte D. Comparison of Portapres non-invasive blood pressuremeasurement in the finger with intra-aortic pressure measurement duringincremental bicycle exercise[J]. Blood pressure monitoring,2002,7(3):179-183.
[60] Barreto A, Li C, Zhai J. Computer evaluation of exercised based on bloodvolume pulse waveform changes[J]. WSEAS Transactions on Biology andBiomedicine,2004,1(1):120-125.
[61] Ifuku H, Taniguchi K, Matsumoto H. Continuous record of carotid artery pulseduring exercise[J]. The Japanese Journal of Physiology,1993,43(1):111-116.
[62] Michael F, O'Rourke. Pulse wave mechanics revisited: relevance to therapy ofcardiovascular disease with calcium antagonists[J]. Heart and vessels,1992,7(3):113-122.
[63] Wei L Y,Chow P. Frequency distribution of human pulse spectra[J].Biomedical Engineering, IEEE Transactions on,1985, BME-32(3):245-246.
[64] Chlebus H,Raczyński J. Comparative examination of the pulse of both rightand left carotid arteries[J]. Bibliotheca cardiologica,1979(37):159-163.
[65]赵莺.病证与脉图相关性研究概述:第七次全国中西医结合四诊研究学术会议论文汇编[C].2004:93-95.
[66] Izzo Jr J L,Shykoff B E. Arterial stiffness: clinical relevance, measurement,and treatment[J]. Cardiovascular Medicine,2001,2(1):29-40.
[67] Asmar R. Arterial stiffness and pulse wave velocity: Clinical Applications[C].Paris:1999.
[68] Yu Y H. Experimental validation of pulse contour methods for estimatingstroke volume at pacing inset: Proceedings of the20th annual internationalconference of the IEEE Engineering in Medicine and Biology society[C].1998:401-404.
[69] Zhang D, Zhang L, Zhang D, et al. Wavelet based analysis of dopplerultrasonic wrist-pulse signals: BioMedical Engineering and Informatics: NewDevelopment and the Future-1st International Conference on BioMedicalEngineering and Informatics, BMEI2008[C]. Sanya, Hainan, China:Inst. ofElec. and Elec. Eng. Computer Society,2008:539-543.
[70] Zhang D, Zuo W, Li Y, et al. Pulse waveform classification using ERP-baseddifference-weighted KNN classifier[J]. Medical Biometrics,2010:191-200.
[71]张冬雨.面向脉诊的脉搏信号与血流信号分类研究[D].哈尔滨工业大学博士论文,2010.
[72] Thomas T L,Antony C S. Detection of transients in noise with theundecimated discrete wavelet transform[J]. IEEE Transactions on SignalProcessing,2000,48(5):1458-1462.
[73] Zhang D, Wang K, Wu X, et al. Hilbert-Huang transform based doppler bloodflow signals analysis:20092nd International Conference on BiomedicalEngineering and Informatics, BMEI2009, October17,2009-October19,2009[C]. Tianjin, China:IEEE Computer Society,2009: IEEE Engineering inMedicine and Biology Society.
[74] Al-Angari H M,Sahakian A V. Use of sample entropy approach to study heartrate variability in obstructive sleep apnea syndrome[J]. IEEE Transactions onBiomedical Engineering,2007,54(10):1900-1904.
[75] Lake D E, Richman J S, Griffin M P, et al. Sample entropy analysis ofneonatal heart rate variability[J]. American Journal of Physiology-Regulatory,Integrative and Comparative Physiology,2002,283(3): R789-R797.
[76] Masry M,Sermadevi Y. A scalable wavelet-based video distortion metric andapplications[J]. IEEE Transactions on Circuits and Systems for VideoTechnology,2006,16(2):260-272.
[77] Yao S, Lin W, Ong E, et al. Contrast signal-to-noise ratio for image qualityassessment: IEEE International Conference on Image Processing2005, ICIP2005, September11,2005-September14,2005[C]. Genova, Italy:Institute ofElectrical and Electronics Engineers Computer Society,2005:397-400.
[78] W.A. Lu C H C, Y.Y. Lin Wang, W.K. A pulse spectrum analysis of thehospital patients with possible liver problems[J]. American Journal of ChineseMedicine,1996,24:311-320.
[79] W.A. Lu Y Y L W, W. K. Wang. Pulse analysis of patients with severe liverproblems-studying pulse spectrums to determine the dffects on other organs[J].IEEE Engineering in Medicine and Biology,1999,18:73-75.
[80] O.Y. Elizabeth Lau A T C. Wrist pulse contour analysis[J]. The University ofHong Kong, PhD Thesis,2002.
[81] Zhang D, Zhang L, Zhang D, et al. Wavelet based analysis of dopplerultrasonic wrist-pulse signals: BioMedical Engineering and Informatics: NewDevelopment and the Future-1st International Conference on BioMedicalEngineering and Informatics, BMEI2008, May27,2008-May30,2008[C].Sanya, Hainan, China:Inst. of Elec. and Elec. Eng. Computer Society,2008:539-543.
[82] Xu L, Meng M Q H, Qi X, et al. Morphology variability analysis of wristpulse waveform for assessment of arteriosclerosis status[J]. Journal ofMedical Systems,2010,34(3):331-339.
[83] Costa M, Goldberger A L, Peng C K. Multiscale entropy analysis of complexphysiologic time series[J]. Physical Review Letters,2002,89(6):068102/1-068102/4.
[84] Richman J S,Moorman J R. Physiological time-series analysis usingapproximate entropy and sample entropy[J]. American Journal ofPhysiology-Heart and Circulatory Physiology,2000,278(6): H2039-H2049.
[85] Zhang Y C. Complexity and1/f noise. A phase space approach[J]. Journal dePhysique I,1991,1(7):971-977.
[86] Valencia J F, Porta A, Vallverdu M, et al. Refined multiscale entropy:application to24-h holter recordings of heart period variability in healthy andaortic stenosis subjects[J]. Biomedical Engineering, IEEE Transactions on,2009,56(9):2202-2213.
[87] Costa M, Peng C K, L Goldberger A, et al. Multiscale entropy analysis ofhuman gait dynamics[J]. Physica A: Statistical Mechanics and its Applications,2003,330(1):53-60.
[88] Lu H, Plataniotis K N, Venetsanopoulos A N. A survey of multilinear subspacelearning for tensor data[J]. Pattern Recognition,2011,44(7):1540-1551.
[89] Lu H, Plataniotis K N, Venetsanopoulos A N. MPCA: Multilinear principalcomponent analysis of tensor objects[J]. IEEE Transactions on NeuralNetworks,2008,19(1):18-39.
[90] Lu H, Plataniotis K N, Venetsanopoulos A N. Uncorrelated multilineardiscriminant analysis with regularization and aggregation for tensor objectrecognition[J]. IEEE Transactions on Neural Networks,2009,20(1):103-123.
[91] Belhumeur P N, Hespanha J P, Kriegman D J. Eigenfaces vs. fisherfaces:recognition using class specific linear projection[J]. IEEE Transactions onPattern Analysis and Machine Intelligence,1997,19(7):711-720.
[92] Jin Z, Yang J-Y, Hu Z-S, et al. Face recognition based on the uncorrelateddiscriminant transformation[J]. Pattern Recognition,2001,34(7):1405-1416.
[93] Veropoulos K, Campbell C, Cristianini N. Controlling the sensitivity ofsupport vector machines: Proceedings of the International Joint Conference onArtificial Intelligence[C]. Citeseer,1999:55-60.
[94] Keogh E,Ratanamahatana C A. Exact indexing of dynamic time warping[J].Knowledge and information systems,2005,7(3):358-386.
[95] Chen L,Ng R, On the marriage of LP-norms and edit distance, in Proceedingsof the Thirtieth international conference on Very large data bases-Volume30.2004, VLDB Endowment: Toronto, Canada. p.792-803.
[96] Marteau P F. Time warp edit distance with stiffness adjustment for time seriesmatching[J]. Pattern Analysis and Machine Intelligence, IEEE Transactions on,2009,31(2):306-318.
[97] Chen Y, Zhang L, Zhang D, et al. The Public Pulse Signal Dataset.2010;Available from: www4.comp.polyu.edu.hk/~cslzhang/Pulse_datasets.htm.
[98] Zhang D, Zuo W, Zhang D, et al. Gaussian ERP kernel classifier for pulsewaveforms classification:201020th International Conference on PatternRecognition, ICPR2010, August23,2010-August26,2010[C]. Istanbul,Turkey:Institute of Electrical and Electronics Engineers Inc.,2010:2736-2739.
[99] Burges C J C. A tutorial on support vector machines for pattern recognition[J].Data mining and knowledge discovery,1998,2(2):121-167.
[100]Wang H,Cheng Y. A quantitative system for pulse diagnosis in traditionalChinese medicine:200527th Annual International Conference of theEngineering in Medicine and Biology Society, EMBS2005[C]. Shanghai,China:Institute of Electrical and Electronics Engineers Inc.,2005:5676-5679.
[101]Chiu C, Yeh S, Yu Y. Classification of the pulse signals based onself-organizing neural network for the analysis of the autonomic nervoussystem[J]. Chin. J. Med. Biol. Eng.,1996,16:461-476.
[102]Xu L, Meng M Q H, Liu R, et al. Robust peak detection of pulse waveformusing height ratio:30th Annual International Conference of the IEEEEngineering in Medicine and Biology Society, EMBS'08[C]. Vancouver, BC,Canada:Inst. of Elec. and Elec. Eng. Computer Society,2008:3856-3859.
[103]Xu L, Meng M Q H, Wang K, et al. Pulse images recognition using fuzzyneural network[J]. Expert systems with applications,2009,36(2):3805-3811.
[104]Wang Y, Wu X, Liu B, et al. Definition and application of indices in Dopplerultrasound sonogram[J]. Shanghai Journal of Biomedical Engineering,1997,18:26-29.
[105]Pincus S M. Approximate entropy as a measure of system complexity[J].Proceedings of the National Academy of Sciences,1991,88(6):2297-2301.
[106]Lanckriet G R G, De Bie T, Cristianini N, et al. A statistical framework forgenomic data fusion[J]. Bioinformatics,2004,20(16):2626.
[107]Sonnenburg S, Ratsch G, Schafer C, et al. Large scale multiple kernellearning[J]. Journal of Machine Learning Research,2006,7:1531-1565.
[108]Rakotomamonjy A, Bach F R, Canu S, et al. SimpleMKL[J]. Journal ofMachine Learning Research,2008,9:2491-2521.
[109]Wei L Y,Chow P. Frequency distribution of human pulse spectra[J]. IEEETransactions on Biomedical Engineering,1985, BME-32(3):245-246.
[110] Thompson L, Jelinek H, Cornforth D. Establishing normative data forperipheral arterial disease using pulse wave analysis:2008InternationalConference on Intelligent Sensors, Sensor Networks and InformationProcessing, ISSNIP2008, December15,2008-December18,2008[C].Sydney, NSW, Australia:Inst. of Elec. and Elec. Eng. Computer Society,2008:351-356.
[111] Lu W-A, Lin Wang Y Y, Wang W K. Pulse analysis of patients with severeliver problems: Studying pulse spectrums to determine the effects on otherorgans[J]. IEEE Engineering in Medicine and Biology Magazine,1999,18(1):73-75.
[2] Lee C T,Wei L Y. Spectrum analysis of human pulse[J]. IEEE Transactions onBiomedical Engineering,1983, BME-30(6):348-352.
[3]李时珍[明].濒湖脉学白话解[M]:人民卫生出版社.1978
[4] Chen Y, Zhang L, Zhang D. Wrist pulse signal diagnosis using modifiedGaussian models and Fuzzy C-Means classification[J]. Medical engineering&physics,2009,31(10):1283-1289.
[5] Chen Y, Zhang L, Zhang D. Computerized wrist pulse signal diagnosis usingmodified auto-regressive models[J]. Journal of Medical Systems,2011,35(3):321-328.
[6]傅骢远,牛欣.中医脉象今释[M]:华夏出版社.1993
[7]费兆馥.现代中医脉诊学[M]:人民卫生出版社.2003
[8]李凌,王志中.脉搏波变异型及其心率变异性关系研究[D].上海交通大学博士论文,2000.
[9]黄世林,孙明异.中医脉象研究[M]:人民卫生出版社.1995
[10] Lu W A, Wang L Y Y, Wang W K. Pulse analysis of patients with severe liverproblems. Studying pulse spectrums to determine the effects on otherorgans[J]. Engineering in Medicine and Biology Magazine, IEEE,1999,18(1):73-75.
[11] Yoon S H, Koga Y, Matsumoto I, et al. An objective method of pulsediagnosis[J]. The American journal of Chinese medicine,1987,15(3-4):147-153.
[12] Broffman M,McCulloch M. Instrument-assisted pulse evaluation in theacupuncture practice[J]. American Journal of Acupuncture,1986,14(3):255-259.
[13] Yoon Y-Z, Lee M-H, Soh K-S. Pulse type classification by varying contactpressure[J]. Engineering in Medicine and Biology Magazine, IEEE,2000,19(6):106-110.
[14] Seng H. Objectifying of pulse-taking[J]. Japanese Journal of OrientalMedicine,1977,27(4):7.
[15] Wang W K, Chen H L, Hsu T L, et al. Alteration of pulse in human subjects bythree Chinese herbs[J]. The American journal of Chinese medicine,1994,22(2):197-203.
[16] Mahomed F. The physiology and clinical use of the sphygmograph[J]. MedTimes Gazette,1872,1:62-64.
[17]汤伟昌,孙汉钧. ZMH-Ⅰ型中医脉象传感器的研究设计[J].传感器技术,1990,1:22-26.
[18]陈冬志,牛欣.中医脉诊实验教学模型的建立与评价[D].北京中医药大学博士研究生论文,2006.
[19]袁玄义,王永昌,张延曹.多普勒效应及其现代应用[J].现代物理知识,2003,12(2):123-126.
[20]柳文仪,赖仲平.彩色脉冲多普勒超声仪对高血压病弦脉与非弦脉的研究[J].中医杂志,1993,34(11):684-685.
[21]牛欣,傅骢远.脉诊位、数、形、势变化的心血管生理学探讨[J].北京中医学院学报,1992,115(1):30-33.
[22] Aritomo M, Yonezawa Y, Caldwell W M. A wrist-mounted activity and pulserecording system: Proceedings of the First Joint BMES/EMBS ConferenceServing Humanity[C]. IEEE,1999:693
[23]徐礼胜.中医脉象数字特征研究[D].哈尔滨工业大学博士论文,2006.
[24]郭庆丽.胆囊炎和肾病综合症脉象信号的特征提取与分类研究[D].哈尔滨工业大学硕士论文,2008.
[25]上海中医学院生理教研室.对几种脉象的初步探讨[J].上海中医药杂志,1979,10(3):43-45.
[26]北京中医学院生理教研室.弦、滑、细等脉象的脉搏图特征初探[J].上海中医药杂志,1980,5(2):8-11.
[27]张卫建.涩脉的脉搏图特征与心血管功能状态[J].中国医药学报,1988,3(2):19-21.
[28]杨天权,郑秀春,孙谨等.典型相关分析在脉图研究中的应用[J].浙江中西医结合杂志,1997,7(1):5-6.
[29]傅骢远.脉诊浮、沉、虚、实的客观指标探讨[J].中西医结合杂志,1990,10(10):603.
[30]李冰星,颜文明.气虚、阳虚患者脉图参数多元逐步回归分析[J].湖南中医学院学报,1990,4:213-216.
[31] Zheng G, Huang Q, Yan G, et al. Pulse waveform key point recognition bywavelet transform for central aortic blood pressure estimation[J]. Journal ofInformation and Computational Science,2012,9(1):25-33.
[32] Nawsupe G,Joshi R R. Modified wavelet-based technique for baseline driftremoval and diagnostic scope of spectral energy of radial pulse signal[J].International Journal of Biomedical Engineering and Technology,2011,6(1):1-13.
[33]刘广斌,樊小力,黄洛秀.两种滑脉患者与健康妇女指端容积脉搏波的功率谱分析[J].西安医科大学学报,1990,11(3):199-201.
[34]王炳和,杨颙.脉搏声信号检测系统实验设计及功率谱特征[J].中华物理医学杂志,1998,39(8):117-120.
[35] Guo Q-L, Wang K-Q, Zhang D-Y, et al. A wavelet packet based pulsewaveform analysis for cholecystitis and nephrotic syndrome diagnosis:Wavelet Analysis and Pattern Recognition,2008. ICWAPR'08. InternationalConference on[C]. IEEE,2008:513-517.
[36] Zhang D Y, Zuo W M, Zhang D, et al. Wrist blood flow signal-basedcomputerized pulse diagnosis using spatial and spectrum features[J]. Journalof Biomedical Science and Engineering,2010,3(4):361-366.
[37] Thakker B, Vyas A L, Tripathi D M. Complexity analysis of wrist pulsesignals[J]. International Journal of Biomedical Engineering and Technology,2012,9(2):188-199.
[38] Rauchberger I, Cohen M, Dart A. Analysis of the diurnal variation of arterialpressure waves: Bioelectromagnetism,1998. Proceedings of the2ndInternational Conference on[C]. IEEE,1998:185-186.
[39] Wang L, Wang K, Xu L. Lempel-ziv decomposition based arrhythmic pulsesrecognition: Engineering in Medicine and Biology Society,2005. IEEE-EMBS2005.27th Annual International Conference of the[C]. IEEE,2006:4606-4609.
[40] Wang K, Xu L, Li Z, et al. Approximate entropy based pulse variabilityanalysis: Computer-Based Medical Systems,2003. Proceedings.16th IEEESymposium[C]. IEEE,2003:236-241.
[41] Maniwa Y, Iokibe T, Koyama M, et al. The application of pulse wave chaos inclinical medicine: Proc.17th FUZY System Symposium, Chiba[C].2001:787-790.
[42] Naschitz J, Itzhak R, Shaviv N, et al. Assessment of cardiovascular reactivityby fractal and recurrence quantification analysis of heart rate and pulse transittime[J]. Journal of human hypertension,2003,17(2):111-118.
[43] Andreadis I I, Giannakakis G A, Papageorgiou C, et al. Detecting complexityabnormalities in dyslexia measuring approximate entropy ofelectroencephalographic signals: Engineering in Medicine and BiologySociety,2009. EMBC2009. Annual International Conference of the IEEE[C].IEEE,2009:6292-6295.
[44] álvarez D, Hornero R, García M, et al. Cross approximate entropy analysis ofnocturnal oximetry signals in the diagnosis of the obstructive sleep apneasyndrome: Engineering in Medicine and Biology Society,2006. EMBS'06.28th Annual International Conference of the IEEE[C]. IEEE,2006:6149-6152.
[45]陈荣山,陈东汉.模糊数学在中医脉象模式识别中应用[J].医学信息,1998,3:20-21.
[46] Lee H-L, Suzuki S, Adachi Y, et al. Fuzzy theory in traditional Chinese pulsediagnosis: Neural Networks,1993. IJCNN'93-Nagoya. Proceedings of1993International Joint Conference on[C]. IEEE,1993:774-777.
[47] Bratteli C, Finkelstein S, Alinder C, et al. Analysis of aging effects on thearterial pulse contour using an artificial neural network: Engineering inMedicine and Biology Society,1998. Proceedings of the20th AnnualInternational Conference of the IEEE[C]. IEEE,1998:1333-1336.
[48] Murthy V S, Ramamoorthy S, Srinivasan N, et al. Analysis ofphotoplethysmographic signals of cardiovascular patterns: Proceedings of the23rd Annual EMBS International Conference[C]. Istanbul, Turkey:2001:2204-2207.
[49]胡家宁,阎述池.脉象人工神经网络分析系统模型[J].中国医科大学学报,1997,26(2):134-137.
[50] Allen J,Murray A. Comparison of three arterial pulse waveform classificationtechniques[J]. Journal of medical engineering&technology,1996,20(3):109-114.
[51] Wang L, Wang K-Q, Xu L-S. Recognizing wrist pulse waveforms withimproved dynamic time warping algorithm: Machine Learning andCybernetics,2004. Proceedings of2004International Conference on[C]. IEEE,2004:3644-3649.
[52] Zhang D, Zuo W, Zhang D, et al. Time series classification using supportvector machine with Gaussian elastic metric kernel:201020th InternationalConference on Pattern Recognition, ICPR2010[C]. Istanbul, Turkey:Instituteof Electrical and Electronics Engineers Inc.,2010:29-32.
[53] Zhang A,Yang F. Study on recognition of sub-health from pulse signal: NeuralNetworks and Brain,2005. ICNN&B'05. International Conference on[C].IEEE,2005:1516-1518.
[54] Richard J P,Kenneth W H. Pulse wave changes with aging[J]. Journal of theAmerican Geriatrics Society,1967,15(2):153-165.
[55] Lawrence M, Resnick, Militianu D, Cunnings A J, et al. Pulse waveformanalysis of arterial compliance: relation to other techniques, age, andmetabolic variables[J]. American journal of hypertension,2000,13(12):1243-1249.
[56] Macdougall J D, Brittain M, MacDonald J, et al. Validity of predicting meanarterial blood pressure during exercise[J]. Medicine and science in sports andexercise,1999,31(12):1876.
[57]盖宇清,田广庆.歼击机与运输机飞行员某些心功能参数的脉图分析[J].航天医学与医学工程,1997,10(5):349-352.
[58]吕萍,闵一建.运动前后脉搏波谱变化的分析研究[J].体育学刊,1997,1:22-23.
[59] Eckert S,Horstkotte D. Comparison of Portapres non-invasive blood pressuremeasurement in the finger with intra-aortic pressure measurement duringincremental bicycle exercise[J]. Blood pressure monitoring,2002,7(3):179-183.
[60] Barreto A, Li C, Zhai J. Computer evaluation of exercised based on bloodvolume pulse waveform changes[J]. WSEAS Transactions on Biology andBiomedicine,2004,1(1):120-125.
[61] Ifuku H, Taniguchi K, Matsumoto H. Continuous record of carotid artery pulseduring exercise[J]. The Japanese Journal of Physiology,1993,43(1):111-116.
[62] Michael F, O'Rourke. Pulse wave mechanics revisited: relevance to therapy ofcardiovascular disease with calcium antagonists[J]. Heart and vessels,1992,7(3):113-122.
[63] Wei L Y,Chow P. Frequency distribution of human pulse spectra[J].Biomedical Engineering, IEEE Transactions on,1985, BME-32(3):245-246.
[64] Chlebus H,Raczyński J. Comparative examination of the pulse of both rightand left carotid arteries[J]. Bibliotheca cardiologica,1979(37):159-163.
[65]赵莺.病证与脉图相关性研究概述:第七次全国中西医结合四诊研究学术会议论文汇编[C].2004:93-95.
[66] Izzo Jr J L,Shykoff B E. Arterial stiffness: clinical relevance, measurement,and treatment[J]. Cardiovascular Medicine,2001,2(1):29-40.
[67] Asmar R. Arterial stiffness and pulse wave velocity: Clinical Applications[C].Paris:1999.
[68] Yu Y H. Experimental validation of pulse contour methods for estimatingstroke volume at pacing inset: Proceedings of the20th annual internationalconference of the IEEE Engineering in Medicine and Biology society[C].1998:401-404.
[69] Zhang D, Zhang L, Zhang D, et al. Wavelet based analysis of dopplerultrasonic wrist-pulse signals: BioMedical Engineering and Informatics: NewDevelopment and the Future-1st International Conference on BioMedicalEngineering and Informatics, BMEI2008[C]. Sanya, Hainan, China:Inst. ofElec. and Elec. Eng. Computer Society,2008:539-543.
[70] Zhang D, Zuo W, Li Y, et al. Pulse waveform classification using ERP-baseddifference-weighted KNN classifier[J]. Medical Biometrics,2010:191-200.
[71]张冬雨.面向脉诊的脉搏信号与血流信号分类研究[D].哈尔滨工业大学博士论文,2010.
[72] Thomas T L,Antony C S. Detection of transients in noise with theundecimated discrete wavelet transform[J]. IEEE Transactions on SignalProcessing,2000,48(5):1458-1462.
[73] Zhang D, Wang K, Wu X, et al. Hilbert-Huang transform based doppler bloodflow signals analysis:20092nd International Conference on BiomedicalEngineering and Informatics, BMEI2009, October17,2009-October19,2009[C]. Tianjin, China:IEEE Computer Society,2009: IEEE Engineering inMedicine and Biology Society.
[74] Al-Angari H M,Sahakian A V. Use of sample entropy approach to study heartrate variability in obstructive sleep apnea syndrome[J]. IEEE Transactions onBiomedical Engineering,2007,54(10):1900-1904.
[75] Lake D E, Richman J S, Griffin M P, et al. Sample entropy analysis ofneonatal heart rate variability[J]. American Journal of Physiology-Regulatory,Integrative and Comparative Physiology,2002,283(3): R789-R797.
[76] Masry M,Sermadevi Y. A scalable wavelet-based video distortion metric andapplications[J]. IEEE Transactions on Circuits and Systems for VideoTechnology,2006,16(2):260-272.
[77] Yao S, Lin W, Ong E, et al. Contrast signal-to-noise ratio for image qualityassessment: IEEE International Conference on Image Processing2005, ICIP2005, September11,2005-September14,2005[C]. Genova, Italy:Institute ofElectrical and Electronics Engineers Computer Society,2005:397-400.
[78] W.A. Lu C H C, Y.Y. Lin Wang, W.K. A pulse spectrum analysis of thehospital patients with possible liver problems[J]. American Journal of ChineseMedicine,1996,24:311-320.
[79] W.A. Lu Y Y L W, W. K. Wang. Pulse analysis of patients with severe liverproblems-studying pulse spectrums to determine the dffects on other organs[J].IEEE Engineering in Medicine and Biology,1999,18:73-75.
[80] O.Y. Elizabeth Lau A T C. Wrist pulse contour analysis[J]. The University ofHong Kong, PhD Thesis,2002.
[81] Zhang D, Zhang L, Zhang D, et al. Wavelet based analysis of dopplerultrasonic wrist-pulse signals: BioMedical Engineering and Informatics: NewDevelopment and the Future-1st International Conference on BioMedicalEngineering and Informatics, BMEI2008, May27,2008-May30,2008[C].Sanya, Hainan, China:Inst. of Elec. and Elec. Eng. Computer Society,2008:539-543.
[82] Xu L, Meng M Q H, Qi X, et al. Morphology variability analysis of wristpulse waveform for assessment of arteriosclerosis status[J]. Journal ofMedical Systems,2010,34(3):331-339.
[83] Costa M, Goldberger A L, Peng C K. Multiscale entropy analysis of complexphysiologic time series[J]. Physical Review Letters,2002,89(6):068102/1-068102/4.
[84] Richman J S,Moorman J R. Physiological time-series analysis usingapproximate entropy and sample entropy[J]. American Journal ofPhysiology-Heart and Circulatory Physiology,2000,278(6): H2039-H2049.
[85] Zhang Y C. Complexity and1/f noise. A phase space approach[J]. Journal dePhysique I,1991,1(7):971-977.
[86] Valencia J F, Porta A, Vallverdu M, et al. Refined multiscale entropy:application to24-h holter recordings of heart period variability in healthy andaortic stenosis subjects[J]. Biomedical Engineering, IEEE Transactions on,2009,56(9):2202-2213.
[87] Costa M, Peng C K, L Goldberger A, et al. Multiscale entropy analysis ofhuman gait dynamics[J]. Physica A: Statistical Mechanics and its Applications,2003,330(1):53-60.
[88] Lu H, Plataniotis K N, Venetsanopoulos A N. A survey of multilinear subspacelearning for tensor data[J]. Pattern Recognition,2011,44(7):1540-1551.
[89] Lu H, Plataniotis K N, Venetsanopoulos A N. MPCA: Multilinear principalcomponent analysis of tensor objects[J]. IEEE Transactions on NeuralNetworks,2008,19(1):18-39.
[90] Lu H, Plataniotis K N, Venetsanopoulos A N. Uncorrelated multilineardiscriminant analysis with regularization and aggregation for tensor objectrecognition[J]. IEEE Transactions on Neural Networks,2009,20(1):103-123.
[91] Belhumeur P N, Hespanha J P, Kriegman D J. Eigenfaces vs. fisherfaces:recognition using class specific linear projection[J]. IEEE Transactions onPattern Analysis and Machine Intelligence,1997,19(7):711-720.
[92] Jin Z, Yang J-Y, Hu Z-S, et al. Face recognition based on the uncorrelateddiscriminant transformation[J]. Pattern Recognition,2001,34(7):1405-1416.
[93] Veropoulos K, Campbell C, Cristianini N. Controlling the sensitivity ofsupport vector machines: Proceedings of the International Joint Conference onArtificial Intelligence[C]. Citeseer,1999:55-60.
[94] Keogh E,Ratanamahatana C A. Exact indexing of dynamic time warping[J].Knowledge and information systems,2005,7(3):358-386.
[95] Chen L,Ng R, On the marriage of LP-norms and edit distance, in Proceedingsof the Thirtieth international conference on Very large data bases-Volume30.2004, VLDB Endowment: Toronto, Canada. p.792-803.
[96] Marteau P F. Time warp edit distance with stiffness adjustment for time seriesmatching[J]. Pattern Analysis and Machine Intelligence, IEEE Transactions on,2009,31(2):306-318.
[97] Chen Y, Zhang L, Zhang D, et al. The Public Pulse Signal Dataset.2010;Available from: www4.comp.polyu.edu.hk/~cslzhang/Pulse_datasets.htm.
[98] Zhang D, Zuo W, Zhang D, et al. Gaussian ERP kernel classifier for pulsewaveforms classification:201020th International Conference on PatternRecognition, ICPR2010, August23,2010-August26,2010[C]. Istanbul,Turkey:Institute of Electrical and Electronics Engineers Inc.,2010:2736-2739.
[99] Burges C J C. A tutorial on support vector machines for pattern recognition[J].Data mining and knowledge discovery,1998,2(2):121-167.
[100]Wang H,Cheng Y. A quantitative system for pulse diagnosis in traditionalChinese medicine:200527th Annual International Conference of theEngineering in Medicine and Biology Society, EMBS2005[C]. Shanghai,China:Institute of Electrical and Electronics Engineers Inc.,2005:5676-5679.
[101]Chiu C, Yeh S, Yu Y. Classification of the pulse signals based onself-organizing neural network for the analysis of the autonomic nervoussystem[J]. Chin. J. Med. Biol. Eng.,1996,16:461-476.
[102]Xu L, Meng M Q H, Liu R, et al. Robust peak detection of pulse waveformusing height ratio:30th Annual International Conference of the IEEEEngineering in Medicine and Biology Society, EMBS'08[C]. Vancouver, BC,Canada:Inst. of Elec. and Elec. Eng. Computer Society,2008:3856-3859.
[103]Xu L, Meng M Q H, Wang K, et al. Pulse images recognition using fuzzyneural network[J]. Expert systems with applications,2009,36(2):3805-3811.
[104]Wang Y, Wu X, Liu B, et al. Definition and application of indices in Dopplerultrasound sonogram[J]. Shanghai Journal of Biomedical Engineering,1997,18:26-29.
[105]Pincus S M. Approximate entropy as a measure of system complexity[J].Proceedings of the National Academy of Sciences,1991,88(6):2297-2301.
[106]Lanckriet G R G, De Bie T, Cristianini N, et al. A statistical framework forgenomic data fusion[J]. Bioinformatics,2004,20(16):2626.
[107]Sonnenburg S, Ratsch G, Schafer C, et al. Large scale multiple kernellearning[J]. Journal of Machine Learning Research,2006,7:1531-1565.
[108]Rakotomamonjy A, Bach F R, Canu S, et al. SimpleMKL[J]. Journal ofMachine Learning Research,2008,9:2491-2521.
[109]Wei L Y,Chow P. Frequency distribution of human pulse spectra[J]. IEEETransactions on Biomedical Engineering,1985, BME-32(3):245-246.
[110] Thompson L, Jelinek H, Cornforth D. Establishing normative data forperipheral arterial disease using pulse wave analysis:2008InternationalConference on Intelligent Sensors, Sensor Networks and InformationProcessing, ISSNIP2008, December15,2008-December18,2008[C].Sydney, NSW, Australia:Inst. of Elec. and Elec. Eng. Computer Society,2008:351-356.
[111] Lu W-A, Lin Wang Y Y, Wang W K. Pulse analysis of patients with severeliver problems: Studying pulse spectrums to determine the effects on otherorgans[J]. IEEE Engineering in Medicine and Biology Magazine,1999,18(1):73-75.