生物雷达检测技术中心跳与呼吸信号分离技术的研究
|
摘要
生物雷达是近年来国外学者提出的一种新概念雷达,特指探测生命体的雷达,其融合雷达技术、生物医学工程技术于一体,可以不需要任何电极或传感器接触生命体,实现隔着衣服、被褥、纱布等物体非接触、远距离、无约束的检测到人体的呼吸及心脏跳动等信息,在临床应用特别是军事医学上有特殊意义,可以实现一些特殊应用场合,如对大面积创伤、烧伤、恶性传染病患者等不宜采用与皮肤接触的传感器或电极的患者进行监护。
目前生物雷达探测技术中仍然存在以下问题:首先,检测系统可在被检测对象屏住呼吸的情况下较好的检测到人体心跳信号,而当被检测对象正常呼吸时,由于呼吸运动带来的胸廓及身体的微动影响,心跳信号的检测受到了呼吸运动所导致的体表微动的干扰,使检测系统无法检测出规则的心跳信号。因此,必须采取有效的信号处理方法将呼吸与心跳信号分离,提取出比呼吸信号能量小得多的心跳信号。其次,在实际应用中,呼吸与心跳信号究竟是以何种关系形式存在是我们选择信号处理方法的关键所在,由于系统的非线性影响以及呼吸与体动两种信号之间关系的不确定性,导致我们不能够单纯的依靠线性滤波来分离或处理这些信号分量,必须求助于非线性滤波。再者,目前虽然已经明确,生物雷达探测到的人体呼吸、心跳等生命参数信号是一种窄带、准周期、低幅值(数量级约几十μV)的信号,信噪比低、随机性强,而且集中在超低频范围内,但对于其群体性及个体之间存在的差异性并没有进行过比较具体的量化研究。这给我们对人体呼吸、心跳等生命信息的进一步提取以及临床应用带来了一定的困难。
针对以上问题,本研究主要完成了以下工作:
1.心跳信号与心电信号相关性研究
通过建立生物雷达检测的心跳信号与心电图机检测的心电信号的同步监测系统,分别在时域及频域上对同步采集到的心电信号以及心跳信号进行相关性对比分析,验证了生物雷达检测的心跳信号的潜在临床应用价值,为后续的信号分析及临床应用提供了有效依据。
2.生物雷达探测到的人体心跳信号的特点分析
针对生命信号个体差异大,准周期,随机性强的特点,采集大量不同测量条件下的实验数据,用相关系数等指标对生物雷达探测到的人体心跳信号在时域及频域上的特点进行了统计学比较、分析和总结。对心电、心音的生理基础、特性及研究方法的总结以及呼吸影响心功能的机制研究有助于我们更好的了解所检测到的心跳信号特性。
3.小波分析算法提取心跳信号
对小波分析与其它非平稳信号处理方法进行了对比分析,并介绍了基于小波变换的信号分离应用情况,指出了小波应用中存在的问题。选用Sym8小波基函数对采集到的体动路信号进行了5层的小波分解降噪算法,为了验证在小波消噪前是否有必要加入FIR滤波算法,我们还对两种方案下生物雷达心跳信号的分离效果进行了对比分析。
4.基于曲线拟合技术的心跳信号提取算法
尝试采集不经过硬件预处理的混合信号,其中包含了更为丰富的信息内容,直接对这种混合信号进行处理。采用三次样条插值技术,对该信号波形趋势进行曲线拟合,从而得到呼吸信号,实现呼吸与心跳的分离。结果显示,与小波分析算法相比,曲线拟合技术在处理基线漂移性质的呼吸信号时是存在优越性的。
本课题的主要创新点有:
1.将小波分析算法用于心跳信号与呼吸信号的分离,并对FIR滤波方法在小波分离两种信号的应用中的必要性进行了实验验证。结果显示,只要设定好分解的级数,我们在进行小波消噪前可以考虑不进行FIR滤波,这样可以大大减小运算量。
2.将曲线拟合算法应用于心跳信号与呼吸信号的分离。
综合分析表明,研究生命体及其器官的活动与探测到的生理信息间的关系,即关注生命体状况,是生物雷达的研究的一个主要方面。鉴于医学和人类发展的其它需求,如果解决这一阶段的难题,将能为生物雷达技术开辟更为广阔的应用空间,具有重要的意义。
Bioradar is a new concept of radar technology combined with the biomedical engineering technique, which can detect the life parameters (such as respiratory and heartbeat signals) through clothing, bedclothes, pledget without any sensor or electrode touching the human subject. Since the detection is non-contact, long distance and without restriction, it has unusual significance in clinical application especially in military medicine. This method can be applied to some special clinical cases, such as to monitor the patients when sensors or electrodes are unsuitable for touching the skin. Those patients are usually in a condition of seriously burned, severely infected or badly wounded.
However, there are still many crucial techniques to be solved to meet the requirements for practical use. First, the heartbeat signal can be well detected when the human subject held his or her breath. While the subject freely breathed, the heartbeat signal can’t be detected. The reason is that the minute chest movement is caused not only by the heartbeat, but also by the interferences introduced by the respiration and the body movement. Therefore, an effective signal processing method should be applied to separate the respiratory and heartbeat signal in order to extract the heartbeat signal which has less energy than respiratory signal. Secondly, it’s of great importance to find out the relationship between the two kinds of signals, which is the key point for us to choose a signal processing method. Taking into consideration of the nonlinear influence from the system and the uncertainty of the two signals, it’s difficult for us to totally separate them by linear filtering. The nonlinear filtering method should also be considered. Thirdly, it has been proved that the life parameters detected by the bioradar are a kind of narrowband, semi periodicity and low amplitudes. And also the signal is characterized by low Signal Noise Ratio (SNR), strong randomness and usually converges towards low frequency portion. But by now the quantitative differences between groups and individuals haven’t been studied in detail, which is necessary for the separation, when is practically applied in clinical.
Taking into account of the problems discussed above, the study mainly involves the followings:
1) Study on the correlation between the Ballistocardiogram (Heartbeat) signal in life detecting system and the Electrocardiograph signal The system is constructed which synchronously detects the electrocardiogram signals by the electrocardiograph and the ballistocardiogram signals by the non-contact life parameter detecting technology. Also, the detected signals are analyzed respectively in the time and frequency domain. The underlying clinical medicine usefulness of ballistocardiogram detected by the non-contact technology is approved, and the credible evidence for the succeeding signal analysis and the clinical application is provided.
2) Analysis on the characteristics of the heartbeat signal detected by bioradar
Due to the fact that the life signals are semi periodicity, strong randomness and dissimilarity, large quantities of experimental data are collected in different testing conditions. The data are also statistically compared, analyzed and summarized by the use of statistical parameters, such as correlation coefficients. The physiological basics, characteristics and study methods for ECG and cardiac sound are summed up. And the mechanisms study on the respiratory influence to heart is also introduced. Those are helpful for us to better understand the characteristics of the heartbeat signals.
3) Extraction of the heartbeat signal based on the method of wavelet analysis Wavelet analysis and other non-steady signal processing method are contrasted in this part. The application of wavelet analysis to signal separation and the existent problems are also introduced. By the use of Sym8 wavelet basis function, the signal is decomposed to 5 levels and denoised. Finally, the separation results of wavelet method with or without FIR filtering method are contrasted by the parameter of SNR.
4) Extraction of heartbeat signal based on the method of curve fitting It’s taken on trial to collect the signal without preprocessing procedure of the hardware. This kind of mixed signal contains more information about heartbeat and respiration. By the use of the technique of cubic spline interpolation, the baseline wander is removed, which is mostly caused by respiration. The results have shown that compared with wavelet analysis, Curve Fitting method has advantages in dealing with the respiration signal which is in the form of baseline wander.
The thesis obtained some results as follows:
1) The application of wavelet analysis to the separation of heartbeat and respiration signals. And the existent problems are also pointed out. The necessity of FIR filtering method are approved. The results show that if only the level number of decomposition is fixed, FIR filtering can be left out. As a result of that, the operation will be simple.
2) Curve fitting method is used to the separation.
It has been shown that one of the mainly study aspect for bioradar is to concern the life status, which is to study the relationship between the detected life parameters and the activities of the living objects and its organs. As a result of the needs for medicine and other development for human, it’s of great importance to solve the existent problems, which can make the technique of bioradar more widely used than before.
目前生物雷达探测技术中仍然存在以下问题:首先,检测系统可在被检测对象屏住呼吸的情况下较好的检测到人体心跳信号,而当被检测对象正常呼吸时,由于呼吸运动带来的胸廓及身体的微动影响,心跳信号的检测受到了呼吸运动所导致的体表微动的干扰,使检测系统无法检测出规则的心跳信号。因此,必须采取有效的信号处理方法将呼吸与心跳信号分离,提取出比呼吸信号能量小得多的心跳信号。其次,在实际应用中,呼吸与心跳信号究竟是以何种关系形式存在是我们选择信号处理方法的关键所在,由于系统的非线性影响以及呼吸与体动两种信号之间关系的不确定性,导致我们不能够单纯的依靠线性滤波来分离或处理这些信号分量,必须求助于非线性滤波。再者,目前虽然已经明确,生物雷达探测到的人体呼吸、心跳等生命参数信号是一种窄带、准周期、低幅值(数量级约几十μV)的信号,信噪比低、随机性强,而且集中在超低频范围内,但对于其群体性及个体之间存在的差异性并没有进行过比较具体的量化研究。这给我们对人体呼吸、心跳等生命信息的进一步提取以及临床应用带来了一定的困难。
针对以上问题,本研究主要完成了以下工作:
1.心跳信号与心电信号相关性研究
通过建立生物雷达检测的心跳信号与心电图机检测的心电信号的同步监测系统,分别在时域及频域上对同步采集到的心电信号以及心跳信号进行相关性对比分析,验证了生物雷达检测的心跳信号的潜在临床应用价值,为后续的信号分析及临床应用提供了有效依据。
2.生物雷达探测到的人体心跳信号的特点分析
针对生命信号个体差异大,准周期,随机性强的特点,采集大量不同测量条件下的实验数据,用相关系数等指标对生物雷达探测到的人体心跳信号在时域及频域上的特点进行了统计学比较、分析和总结。对心电、心音的生理基础、特性及研究方法的总结以及呼吸影响心功能的机制研究有助于我们更好的了解所检测到的心跳信号特性。
3.小波分析算法提取心跳信号
对小波分析与其它非平稳信号处理方法进行了对比分析,并介绍了基于小波变换的信号分离应用情况,指出了小波应用中存在的问题。选用Sym8小波基函数对采集到的体动路信号进行了5层的小波分解降噪算法,为了验证在小波消噪前是否有必要加入FIR滤波算法,我们还对两种方案下生物雷达心跳信号的分离效果进行了对比分析。
4.基于曲线拟合技术的心跳信号提取算法
尝试采集不经过硬件预处理的混合信号,其中包含了更为丰富的信息内容,直接对这种混合信号进行处理。采用三次样条插值技术,对该信号波形趋势进行曲线拟合,从而得到呼吸信号,实现呼吸与心跳的分离。结果显示,与小波分析算法相比,曲线拟合技术在处理基线漂移性质的呼吸信号时是存在优越性的。
本课题的主要创新点有:
1.将小波分析算法用于心跳信号与呼吸信号的分离,并对FIR滤波方法在小波分离两种信号的应用中的必要性进行了实验验证。结果显示,只要设定好分解的级数,我们在进行小波消噪前可以考虑不进行FIR滤波,这样可以大大减小运算量。
2.将曲线拟合算法应用于心跳信号与呼吸信号的分离。
综合分析表明,研究生命体及其器官的活动与探测到的生理信息间的关系,即关注生命体状况,是生物雷达的研究的一个主要方面。鉴于医学和人类发展的其它需求,如果解决这一阶段的难题,将能为生物雷达技术开辟更为广阔的应用空间,具有重要的意义。
Bioradar is a new concept of radar technology combined with the biomedical engineering technique, which can detect the life parameters (such as respiratory and heartbeat signals) through clothing, bedclothes, pledget without any sensor or electrode touching the human subject. Since the detection is non-contact, long distance and without restriction, it has unusual significance in clinical application especially in military medicine. This method can be applied to some special clinical cases, such as to monitor the patients when sensors or electrodes are unsuitable for touching the skin. Those patients are usually in a condition of seriously burned, severely infected or badly wounded.
However, there are still many crucial techniques to be solved to meet the requirements for practical use. First, the heartbeat signal can be well detected when the human subject held his or her breath. While the subject freely breathed, the heartbeat signal can’t be detected. The reason is that the minute chest movement is caused not only by the heartbeat, but also by the interferences introduced by the respiration and the body movement. Therefore, an effective signal processing method should be applied to separate the respiratory and heartbeat signal in order to extract the heartbeat signal which has less energy than respiratory signal. Secondly, it’s of great importance to find out the relationship between the two kinds of signals, which is the key point for us to choose a signal processing method. Taking into consideration of the nonlinear influence from the system and the uncertainty of the two signals, it’s difficult for us to totally separate them by linear filtering. The nonlinear filtering method should also be considered. Thirdly, it has been proved that the life parameters detected by the bioradar are a kind of narrowband, semi periodicity and low amplitudes. And also the signal is characterized by low Signal Noise Ratio (SNR), strong randomness and usually converges towards low frequency portion. But by now the quantitative differences between groups and individuals haven’t been studied in detail, which is necessary for the separation, when is practically applied in clinical.
Taking into account of the problems discussed above, the study mainly involves the followings:
1) Study on the correlation between the Ballistocardiogram (Heartbeat) signal in life detecting system and the Electrocardiograph signal The system is constructed which synchronously detects the electrocardiogram signals by the electrocardiograph and the ballistocardiogram signals by the non-contact life parameter detecting technology. Also, the detected signals are analyzed respectively in the time and frequency domain. The underlying clinical medicine usefulness of ballistocardiogram detected by the non-contact technology is approved, and the credible evidence for the succeeding signal analysis and the clinical application is provided.
2) Analysis on the characteristics of the heartbeat signal detected by bioradar
Due to the fact that the life signals are semi periodicity, strong randomness and dissimilarity, large quantities of experimental data are collected in different testing conditions. The data are also statistically compared, analyzed and summarized by the use of statistical parameters, such as correlation coefficients. The physiological basics, characteristics and study methods for ECG and cardiac sound are summed up. And the mechanisms study on the respiratory influence to heart is also introduced. Those are helpful for us to better understand the characteristics of the heartbeat signals.
3) Extraction of the heartbeat signal based on the method of wavelet analysis Wavelet analysis and other non-steady signal processing method are contrasted in this part. The application of wavelet analysis to signal separation and the existent problems are also introduced. By the use of Sym8 wavelet basis function, the signal is decomposed to 5 levels and denoised. Finally, the separation results of wavelet method with or without FIR filtering method are contrasted by the parameter of SNR.
4) Extraction of heartbeat signal based on the method of curve fitting It’s taken on trial to collect the signal without preprocessing procedure of the hardware. This kind of mixed signal contains more information about heartbeat and respiration. By the use of the technique of cubic spline interpolation, the baseline wander is removed, which is mostly caused by respiration. The results have shown that compared with wavelet analysis, Curve Fitting method has advantages in dealing with the respiration signal which is in the form of baseline wander.
The thesis obtained some results as follows:
1) The application of wavelet analysis to the separation of heartbeat and respiration signals. And the existent problems are also pointed out. The necessity of FIR filtering method are approved. The results show that if only the level number of decomposition is fixed, FIR filtering can be left out. As a result of that, the operation will be simple.
2) Curve fitting method is used to the separation.
It has been shown that one of the mainly study aspect for bioradar is to concern the life status, which is to study the relationship between the detected life parameters and the activities of the living objects and its organs. As a result of the needs for medicine and other development for human, it’s of great importance to solve the existent problems, which can make the technique of bioradar more widely used than before.
引文
[1] Chuang HR, Devedra M, Wang H, Postow E.An X-band microwave life-detection systems[J]. IEEE Trans Biomed Eng, 1986,33(7):697-701.
[2] 本村和磨,荒井郁男.24GHz マィクロ波心拍モニタの開発[J]. 医用電子と生体工学,1997,35(3):17-23.
[3] D.K.Misra. Scattering of electromagnetic waves by human body and its applications[D]. Ph.D. dissertation, Michigan State Univ, East Lansing, 1984
[4] Alihanka J, Vaahtoranta K. A new method for long-term monitoring of the ballistocardiogram,heart rate, and respiration[J]. AMJ Physio,1981,240(5):384-392.
[5] K.M.Chen, D.K.Misra. An X-band microwave life detection system[C]. Presented at the 6th Annu.Meet. Bioelectromagn.Soc,Atlanta,GA,July 1984,15-19.
[6] Yamaguchi,Mitsumoto M,Sengoku M,et al.Synthetic Aperture FM-CW Radar Applied to the Detection of Objects Buried in Snowpack [J]. IEEE Trans Geoscience and Remote Sensing,1994,32(1):11-18.
[7] Chen KM,Huang Y,Zhang JP,et a1.Microwave life-detection system for searching human subjects under earthquake rubble or behind barrier[J]. IEEE Trans BME,2000,27(1):105-114.
[8] Caro CG,Bloice JA.Contactless apnea detector based on radar[J]. Lancet,1971,2(7731):959-961.
[9] Franks CI,Watson JB,Foster EF.Respiration Patterns and risk of suddenunexpected death in infancy[J].Arch Dis Child,1980,55(8):595-604.
[10] Micropower Impulse Radar[R]. Science and Technology Review.UCRL, 1996.
[11] Amy D.Droitcour, O.Boric -Lubecke, V.M.Lubecke, et al. 0.25 ìm CMOS and BiCMOS single-chip direct-conversion Doppler radars for remote sensing of vital signs [J]. Solid-State Circuits Conference, Digest of Technical Papers. ISSCC. 2002 IEEE International, 2002,348-349.
[12] Droitcour A, Boric-Lubecke O, Lubecke VM, et.al. Range Correlation Effect on ISM Band I/Q CMOS Radar for Non-Contact Cardiopulmonary Monitoring[J]. 2003 IEEE MTT-S IMS Digest, Philadelphis,2003,3:1945-1948.
[13] Florian Michahelles, Ramon Wicki , Bernt Schiele. Less Contact: Heart-Rate Detection Without Even Touching the User [J]. Eighth IEEE International Symposium on Wearable Computers Conference (ISWC), 2004.
[14] 尹秋艳,樊明捷,黄勇. 用微波频谱分析仪检测人体心动信号[J]. 2003年全国微波毫米波会议论文集,2003,1061-1064.
[15] 黄莉,史林,姜敏.基于提升算法的低速目标信号提取与生命信号检测应用[J].电子科技,2004,5:18-21.
[16] 史林,姜敏,黄莉.基于谐波模型的生命探测雷达人体状态识别方法[J].西安电子科技大学学报(自然科学版),2005,32(2):179-183.
[17] 王海滨,倪安胜,王健琪等.LMS 算法在非接触生命参数信号检测中的消噪应用[J].中国医疗器械杂志,2003,27(1):21-24.
[18] 杨冬.基于非接触生命参数检测系统的信号处理技术研究[D].第四军医大学硕士学位论文,2004,38-52.
[19] Jutten C,Herault J. Blind separation of source,part I: An adaptive algorithmbased on neuromimetic architecture [J] .Signal Processing,l99l,24:1-10.
[20] Common P. Independent component analysis, a new concept? [J]. Signal Processing,1994,36:287-314.
[21] A J Bell, T J Sejnowski. An information-maximisation approach to blind separation and blind deconvolution [J].Neural Computation,1995,7:1129-1159.
[22] Lee TW. Independent component analysis using an extended infomax algorithm for mixed Subgaussian and Supergaussian sources [J]. Neural Computation,1999,11(2):409-433.
[23] Lieven De Lathauwer, Bart De Moor, and Joos Banewale. Fetal ECG extraction by blind source subspace separation [J]. IEEE Transactions on Biomedical Engineering,2000,47(5):567-572.
[24] Ricardo Vigario, Jaako Sarela, Beikko Jousmiiki etl. Independent component approach to the analysis of EEG and MEG recordings [J]. IEEE Tansactions on Biomedical Engineering,2000,47(5):589-593.
[25] 周卫东, 贾磊. 小波变换和独立分量分析去除脑电信号中的噪声和干扰[J]. 山东大学学报(医学版),2003,41(2):116-122.
[26] Burel G. Blind separation of sources: a nonlinear neural algorithm [J]. Neural Networks,1992,5(6):937-947.
[27] Deco G, Brauer W. Nonlinear higher-order statistical decorrelation by volume-conserving[J]. Neural Networks,1995,8:525-535.
[28] Yang HH,Amari SI,Cichocki A. Information-theorectic approach to blind separation of sources in nonlinear mixture[J]. Signal Processing,1998,64(3):291-300.
[29] 席涛,杨国胜等. 基于自适应滤波的心电图中呼吸信号的提取方法[J].第四军医大学学报,2005,26(9):852-854.
[30] 王健琪,王海滨,杨国胜等. 心冲击图的雷达式非接触检测技术研究[J].北京生物医学工程,2001,20(2):112.
[31] Starr I, Karreman G, et al. Blind study on the relations between the extent of coronary arteriosclerosis and the strength of the myocardial contraction as measured by invasive and noninvasive tests[J]. Am Heart J, 1978,96: 37-46.
[32] 张唯真.生物医学电子学[M].北京:清华大学出版社,1990,11.
[33] Pavlov S.N.,Samkov S.V. Algorithm of Signal Processing in Ultra-Wideband Radar Designed for Remote Measuring Parameters of Patient’s Cardiac Activity[C]. Sevastopol, Ukraine:Ultra Wideband and Ultra Short Impulse Signals,19-22 September,2003,1-3.
[34] 倪安胜,王健琪,王海滨等. 非接触生命参数检测系统软件的研制[J]. 医疗卫生装备,2003,24(8):8-9.
[35] Ralph M.Myerson. M.D..丁孝宏等译.心脏的功能[M].广州:广东人民出版社,1995.
[36] Donna M.Mooney,Lynn J. Groome,J.Doug Wilson,Dennis L. Stearns,Lynn S.Bentz.A PC-controlled data acquisition system for transabdominal recording of cardiac activity in the human fetus[J].Proceedings of the 1993 ACM/SIGAPP symposium on Applied computing,March 1993.
[37] Reguig,F.Bereksi,Kirk,D.L. Design of an FECG scalp electrode fetal heart rate monitor[J]. Medical Engineering & Physics,1996,18(2):150-160.
[38] Jean M.Darniede,Dean C. Jeutter etc.Miniature Microcontroller-based Heart Rate Telemeter processes Single Precordial Lead[J].IEEE.1994,900-901.
[39] Morlet,D.,Couderc,J.ph.,Touboul,P.,Rubel,P.. Wavelet analysis ofhigh-resolution ECG in Post-infarction Patients:role of the basic wavelet and of the analyzed lead[J].International Journal of Bio-Medical Computing,1995,39(3):311-325.
[40] Bahoura,M..Hassani,M..Hubin,M.. DSP implementation of wavelet transform for real time ECG wave forms detection and heart rate analysis[J]. Computer Methods and Programs in Biomedicine,1997,52(1):35-44.
[41] Khobragade,Kalyani S.,Deshmukh,R.B.,ECG analysis using wavelet transforms[J].Computer Standards and Interfaces,1999,20(6-7):466.
[42] Cherkassky,Vladimir,Kilts,Steven.Myopotential denoising of ECG signals using wavelet thresholding methods[J]. Neural Networks,2001,14(8):1129-1137.
[43] 方晓颖.微波人体心动信号的研究[D].华东师范大学硕士学位论文,2004,14.
[44] LIU Shaoying,LU Jilai,HA0 Li,et a1.Detection of QRS complex using mathematical morphology and wavelet transform[J].Journal of Tsinghua University(Science and Technology),2004,44(6):852-855.
[45] G.Jamous etc. Optimal time-window duration for computing time/frequency representations of normal phonocardiograms in dogs[J]. Med.Biol.Eng.Comput, 1992,(7):502.
[46] Oskiper T, Watrous R. Results on the time-frequency characterization of the first heart sound in normal man[A].Proceeding of the Second Joint EMBS/BMES Conference[C].USA:Houston,TX,2002,126-127.
[47] Livanos G,Ranganathan N,Jiang J.Heart sound analysis using the S transform[J].Computers in Cardiology,2002,27:587-590.
[48] Ozgur S,Zumray D,Tamer O.Classification of heart sounds by usingwavelet Transform[A] . Proceeding of the Second Joint EMBS / BMES Conterence[C].USA:Houston,TX,2002,128-129.
[49] Kurnaz MN.Olmez T.Determination of features for heart sounds by using wavelet transforms[A].Proceedings of the 15th IEEE Symposium on Computer-Based Medical Systems[C].2002,156-159.
[50] Oskiper T, Watrous R.Detection of the first heart sound using a time-delay neural network[J].Computers in Cardiology,2002,29:537-540.
[51] 齐颁扬.医学仪器[M].北京:高等教育出版社,1989,61-160.
[52] Scharf SM,Cassidy SS.Heart-lung interactions in health and disease.Lung Biology in Health and Disease[M].The United States of America:Marecel Dekker,Inc,1989,251-253,315-330.
[53] Braunwald E. Heart Disease[M] . A Textbook of Cardiovascular Medicine.Philadelphia,PA.Saunders WB,2001.409-411,1845-1847.
[54] Cao T , Yuan L , Duan Y . Mechanism study of respiration-related hemodynamics using in vitro and in vivo models[J].Circulation,2003,108(17):1035.
[55] Riggs TW,Snider AR.Respiratory influence on right and left ventricular diastolic function in normal children[J].Am J Cardiol,1989,63:858-861.
[56] Appleton CP,Hatle L K,Popp R L.Cardiac tamponade and pericardial effusion:respiratory variation in transvalvular flow velocities studied by Doppler echocardiography[J].J Am Coll Cardiol,1988,11:1020-1030.
[57] 袁丽君,曹铁生,段云友等.平静呼吸对正常人心内血流速度影响的超声心动图研究[J].中华超声影像学杂志,2002,11(12):736-739.
[58] Tabata T,Kabbani SS,Murray RD,et al.Difference in the respiratory variation between pulmonary venous and mitral inflow Doppler velocities inpatients with constrictive pericarditis with and without atrial fibrillation[J].J Am Coll Cardiol,2001,37(7):1936-1942.
[59] 汪源源.现代信号处理理论和方法[M].上海:复旦大学出版社,2002.
[60] 杨 冬,王健琪等.雷达式非接触生命参数检测系统中心跳信号提取方法研究[J].医疗卫生装备,2005,26(8):78-80.
[61] 刘雪红,吴爱平等.小波变换去除心电信号中呼吸信号干扰[J].生物医学工程与临床,2003,7(2):78-80.
[62] 梁 霖,徐光华,侯成刚.基于奇异值分解的连续小波消噪方法[J].西安交通大学学报,2004,38(9):904-908.
[63] 曹 毅,张榆锋,蒋丽华.用小波变换的模极大值提取胎儿心率的方法[J].北京生物医学工程,2005,24(1): 56-59.
[64] 杨丰,岳小荣.基于三次 B-样条函数心电图数据滤波[J].北京生物医学工程,1994,13(4):193-196.
[65] 夏恒超,詹永麒.一种新的基于三次样条插值技术的心电图基线漂移消除方法[J].生物医学工程学杂志,2003,20(3):524-526.
[2] 本村和磨,荒井郁男.24GHz マィクロ波心拍モニタの開発[J]. 医用電子と生体工学,1997,35(3):17-23.
[3] D.K.Misra. Scattering of electromagnetic waves by human body and its applications[D]. Ph.D. dissertation, Michigan State Univ, East Lansing, 1984
[4] Alihanka J, Vaahtoranta K. A new method for long-term monitoring of the ballistocardiogram,heart rate, and respiration[J]. AMJ Physio,1981,240(5):384-392.
[5] K.M.Chen, D.K.Misra. An X-band microwave life detection system[C]. Presented at the 6th Annu.Meet. Bioelectromagn.Soc,Atlanta,GA,July 1984,15-19.
[6] Yamaguchi,Mitsumoto M,Sengoku M,et al.Synthetic Aperture FM-CW Radar Applied to the Detection of Objects Buried in Snowpack [J]. IEEE Trans Geoscience and Remote Sensing,1994,32(1):11-18.
[7] Chen KM,Huang Y,Zhang JP,et a1.Microwave life-detection system for searching human subjects under earthquake rubble or behind barrier[J]. IEEE Trans BME,2000,27(1):105-114.
[8] Caro CG,Bloice JA.Contactless apnea detector based on radar[J]. Lancet,1971,2(7731):959-961.
[9] Franks CI,Watson JB,Foster EF.Respiration Patterns and risk of suddenunexpected death in infancy[J].Arch Dis Child,1980,55(8):595-604.
[10] Micropower Impulse Radar[R]. Science and Technology Review.UCRL, 1996.
[11] Amy D.Droitcour, O.Boric -Lubecke, V.M.Lubecke, et al. 0.25 ìm CMOS and BiCMOS single-chip direct-conversion Doppler radars for remote sensing of vital signs [J]. Solid-State Circuits Conference, Digest of Technical Papers. ISSCC. 2002 IEEE International, 2002,348-349.
[12] Droitcour A, Boric-Lubecke O, Lubecke VM, et.al. Range Correlation Effect on ISM Band I/Q CMOS Radar for Non-Contact Cardiopulmonary Monitoring[J]. 2003 IEEE MTT-S IMS Digest, Philadelphis,2003,3:1945-1948.
[13] Florian Michahelles, Ramon Wicki , Bernt Schiele. Less Contact: Heart-Rate Detection Without Even Touching the User [J]. Eighth IEEE International Symposium on Wearable Computers Conference (ISWC), 2004.
[14] 尹秋艳,樊明捷,黄勇. 用微波频谱分析仪检测人体心动信号[J]. 2003年全国微波毫米波会议论文集,2003,1061-1064.
[15] 黄莉,史林,姜敏.基于提升算法的低速目标信号提取与生命信号检测应用[J].电子科技,2004,5:18-21.
[16] 史林,姜敏,黄莉.基于谐波模型的生命探测雷达人体状态识别方法[J].西安电子科技大学学报(自然科学版),2005,32(2):179-183.
[17] 王海滨,倪安胜,王健琪等.LMS 算法在非接触生命参数信号检测中的消噪应用[J].中国医疗器械杂志,2003,27(1):21-24.
[18] 杨冬.基于非接触生命参数检测系统的信号处理技术研究[D].第四军医大学硕士学位论文,2004,38-52.
[19] Jutten C,Herault J. Blind separation of source,part I: An adaptive algorithmbased on neuromimetic architecture [J] .Signal Processing,l99l,24:1-10.
[20] Common P. Independent component analysis, a new concept? [J]. Signal Processing,1994,36:287-314.
[21] A J Bell, T J Sejnowski. An information-maximisation approach to blind separation and blind deconvolution [J].Neural Computation,1995,7:1129-1159.
[22] Lee TW. Independent component analysis using an extended infomax algorithm for mixed Subgaussian and Supergaussian sources [J]. Neural Computation,1999,11(2):409-433.
[23] Lieven De Lathauwer, Bart De Moor, and Joos Banewale. Fetal ECG extraction by blind source subspace separation [J]. IEEE Transactions on Biomedical Engineering,2000,47(5):567-572.
[24] Ricardo Vigario, Jaako Sarela, Beikko Jousmiiki etl. Independent component approach to the analysis of EEG and MEG recordings [J]. IEEE Tansactions on Biomedical Engineering,2000,47(5):589-593.
[25] 周卫东, 贾磊. 小波变换和独立分量分析去除脑电信号中的噪声和干扰[J]. 山东大学学报(医学版),2003,41(2):116-122.
[26] Burel G. Blind separation of sources: a nonlinear neural algorithm [J]. Neural Networks,1992,5(6):937-947.
[27] Deco G, Brauer W. Nonlinear higher-order statistical decorrelation by volume-conserving[J]. Neural Networks,1995,8:525-535.
[28] Yang HH,Amari SI,Cichocki A. Information-theorectic approach to blind separation of sources in nonlinear mixture[J]. Signal Processing,1998,64(3):291-300.
[29] 席涛,杨国胜等. 基于自适应滤波的心电图中呼吸信号的提取方法[J].第四军医大学学报,2005,26(9):852-854.
[30] 王健琪,王海滨,杨国胜等. 心冲击图的雷达式非接触检测技术研究[J].北京生物医学工程,2001,20(2):112.
[31] Starr I, Karreman G, et al. Blind study on the relations between the extent of coronary arteriosclerosis and the strength of the myocardial contraction as measured by invasive and noninvasive tests[J]. Am Heart J, 1978,96: 37-46.
[32] 张唯真.生物医学电子学[M].北京:清华大学出版社,1990,11.
[33] Pavlov S.N.,Samkov S.V. Algorithm of Signal Processing in Ultra-Wideband Radar Designed for Remote Measuring Parameters of Patient’s Cardiac Activity[C]. Sevastopol, Ukraine:Ultra Wideband and Ultra Short Impulse Signals,19-22 September,2003,1-3.
[34] 倪安胜,王健琪,王海滨等. 非接触生命参数检测系统软件的研制[J]. 医疗卫生装备,2003,24(8):8-9.
[35] Ralph M.Myerson. M.D..丁孝宏等译.心脏的功能[M].广州:广东人民出版社,1995.
[36] Donna M.Mooney,Lynn J. Groome,J.Doug Wilson,Dennis L. Stearns,Lynn S.Bentz.A PC-controlled data acquisition system for transabdominal recording of cardiac activity in the human fetus[J].Proceedings of the 1993 ACM/SIGAPP symposium on Applied computing,March 1993.
[37] Reguig,F.Bereksi,Kirk,D.L. Design of an FECG scalp electrode fetal heart rate monitor[J]. Medical Engineering & Physics,1996,18(2):150-160.
[38] Jean M.Darniede,Dean C. Jeutter etc.Miniature Microcontroller-based Heart Rate Telemeter processes Single Precordial Lead[J].IEEE.1994,900-901.
[39] Morlet,D.,Couderc,J.ph.,Touboul,P.,Rubel,P.. Wavelet analysis ofhigh-resolution ECG in Post-infarction Patients:role of the basic wavelet and of the analyzed lead[J].International Journal of Bio-Medical Computing,1995,39(3):311-325.
[40] Bahoura,M..Hassani,M..Hubin,M.. DSP implementation of wavelet transform for real time ECG wave forms detection and heart rate analysis[J]. Computer Methods and Programs in Biomedicine,1997,52(1):35-44.
[41] Khobragade,Kalyani S.,Deshmukh,R.B.,ECG analysis using wavelet transforms[J].Computer Standards and Interfaces,1999,20(6-7):466.
[42] Cherkassky,Vladimir,Kilts,Steven.Myopotential denoising of ECG signals using wavelet thresholding methods[J]. Neural Networks,2001,14(8):1129-1137.
[43] 方晓颖.微波人体心动信号的研究[D].华东师范大学硕士学位论文,2004,14.
[44] LIU Shaoying,LU Jilai,HA0 Li,et a1.Detection of QRS complex using mathematical morphology and wavelet transform[J].Journal of Tsinghua University(Science and Technology),2004,44(6):852-855.
[45] G.Jamous etc. Optimal time-window duration for computing time/frequency representations of normal phonocardiograms in dogs[J]. Med.Biol.Eng.Comput, 1992,(7):502.
[46] Oskiper T, Watrous R. Results on the time-frequency characterization of the first heart sound in normal man[A].Proceeding of the Second Joint EMBS/BMES Conference[C].USA:Houston,TX,2002,126-127.
[47] Livanos G,Ranganathan N,Jiang J.Heart sound analysis using the S transform[J].Computers in Cardiology,2002,27:587-590.
[48] Ozgur S,Zumray D,Tamer O.Classification of heart sounds by usingwavelet Transform[A] . Proceeding of the Second Joint EMBS / BMES Conterence[C].USA:Houston,TX,2002,128-129.
[49] Kurnaz MN.Olmez T.Determination of features for heart sounds by using wavelet transforms[A].Proceedings of the 15th IEEE Symposium on Computer-Based Medical Systems[C].2002,156-159.
[50] Oskiper T, Watrous R.Detection of the first heart sound using a time-delay neural network[J].Computers in Cardiology,2002,29:537-540.
[51] 齐颁扬.医学仪器[M].北京:高等教育出版社,1989,61-160.
[52] Scharf SM,Cassidy SS.Heart-lung interactions in health and disease.Lung Biology in Health and Disease[M].The United States of America:Marecel Dekker,Inc,1989,251-253,315-330.
[53] Braunwald E. Heart Disease[M] . A Textbook of Cardiovascular Medicine.Philadelphia,PA.Saunders WB,2001.409-411,1845-1847.
[54] Cao T , Yuan L , Duan Y . Mechanism study of respiration-related hemodynamics using in vitro and in vivo models[J].Circulation,2003,108(17):1035.
[55] Riggs TW,Snider AR.Respiratory influence on right and left ventricular diastolic function in normal children[J].Am J Cardiol,1989,63:858-861.
[56] Appleton CP,Hatle L K,Popp R L.Cardiac tamponade and pericardial effusion:respiratory variation in transvalvular flow velocities studied by Doppler echocardiography[J].J Am Coll Cardiol,1988,11:1020-1030.
[57] 袁丽君,曹铁生,段云友等.平静呼吸对正常人心内血流速度影响的超声心动图研究[J].中华超声影像学杂志,2002,11(12):736-739.
[58] Tabata T,Kabbani SS,Murray RD,et al.Difference in the respiratory variation between pulmonary venous and mitral inflow Doppler velocities inpatients with constrictive pericarditis with and without atrial fibrillation[J].J Am Coll Cardiol,2001,37(7):1936-1942.
[59] 汪源源.现代信号处理理论和方法[M].上海:复旦大学出版社,2002.
[60] 杨 冬,王健琪等.雷达式非接触生命参数检测系统中心跳信号提取方法研究[J].医疗卫生装备,2005,26(8):78-80.
[61] 刘雪红,吴爱平等.小波变换去除心电信号中呼吸信号干扰[J].生物医学工程与临床,2003,7(2):78-80.
[62] 梁 霖,徐光华,侯成刚.基于奇异值分解的连续小波消噪方法[J].西安交通大学学报,2004,38(9):904-908.
[63] 曹 毅,张榆锋,蒋丽华.用小波变换的模极大值提取胎儿心率的方法[J].北京生物医学工程,2005,24(1): 56-59.
[64] 杨丰,岳小荣.基于三次 B-样条函数心电图数据滤波[J].北京生物医学工程,1994,13(4):193-196.
[65] 夏恒超,詹永麒.一种新的基于三次样条插值技术的心电图基线漂移消除方法[J].生物医学工程学杂志,2003,20(3):524-526.