基于扩展Infomax 独立分量分析算法的脑电信号消噪

背景：脑电信号能够反映大脑不同的生理病理状态,但在采集和分析处理过程中极易受到各种噪声的干扰,如眼球运动、眨眼、心电、肌电等,这些噪声的存在严重影响了脑电信号的分析和处理。 目的：介绍了一种基于扩展Infomax的独立分量分析方法,并用于脑电信号消噪。 方法：通过扩展Infomax算法的迭代求得分离矩阵,采用去除噪声分量后的独立成分重构需要记录的脑电信号,观察Matlab仿真得到的去噪后的脑电信号,同时比较去噪前后各导联脑电信号与眼电信号的相关性。 结果与结论：使用扩展Infomax 独立分量分析算法能够成功地去除多导脑电信号中的眼电干扰。再比较去噪前后各导联脑电信号的功率谱,可以发现使用扩展Infomax独立分量分析算法同时也能够成功地去除多导脑电信号中的工频干扰,且对脑电信号中的其他有用信号几乎没有破坏。     

采集表面肌电信号应用于动作识别的可行性

背景：文献表明上肢前臂运动时所产生的表面肌电信号具有非线性特征,而肢体运动时肌电信号又呈现出非平稳特性。 目的：设计一种简单的拾取电路采集表面肌电信号,拟应用于动作肌电信号的特征识别。 方法：根据表面肌电信号的特点,设计高共模抑制比的前端放大电路,抑制共模干扰;采用低通滤波电路,有源双T带阻滤波器对信号进行去噪处理;对采集得到的信号进行小波包变换,得到信号的特征量。 结果与结论：所设计的表面肌电信号检测电路具有较高共模抑制比,并能有效地滤除50 Hz工频信号,可以满足肌电信号采集电路的基本要求。肌电信号的处理结果表明采用子频段能量值的方法可以区分手部4种不同动作。     

基于变分模态分解的眼电伪迹去除

脑电信号可以反映人体大脑活动状态,精确地将脑内信息传递向外界,对脑科学研究具有重要的意义。在实际情况中,脑电信号采集的同时会带有一些噪声,而眼电伪迹的存在会严重干扰脑电信号。本研究尝试了一种基于变分模态分解的眼电伪迹去除方法。通过变分模态分解将采集到的脑电信号分解成K组模态分量;根据眼电伪迹的频率特点,选择出眼电伪迹所对应的模态分量,并将其去除后重新构建剩余的模态分量。结果表明通过对实验数据的处理,变分模态分解可以有效地将眼电伪迹去除,并维持脑电信号的特征。     

基于独立分量分析的脑电中眼电伪迹消除

利用独立分最分析的方法对脑电中眼电伪迹成分进行剔除。针对扩腮熵最大算法能够同时分离超高斯和亚高斯信号的特点，将脑电信号分解成独立分量，利用伪迹脑地形图的特征，将伪迹分最分离，得到不含伪迹的脑电信号。实验结果表明。该算法具有较强的稳健性和实用性。     

基于小波变换和独立分量分析相结合的癫痫脑电分析

临床上分析癫痫脑电信号非常重要。由于临床记录的癫痫脑电信号中含有大量的伪迹干扰,特别是肌电伪迹,所采集的脑电信号无法正确反映大脑的生理及病理状况。本研究利用小波变换的多分辨率特性和独立分量分析(ICA)的盲源分离特性,把用连续小波变换分解的脑电子带信号作为ICA输入,经ICA分离后,有效地消除了癫痫脑电中的肌电伪迹,并分离出了癫痫样特征波,效果理想。     

基于遗传算法的多通道癫痫脑电信号盲源分离

目的：研究一种将心电噪声信号从脑电信号中分离出来的算法及其DSP硬件实现。方法：癫痫是一种中枢神经系统疾病，该病的诊断主要依靠脑电监测，但由于人体是一个复杂网络，临床采集到的脑电通常会混有其他噪声如心电干扰，这为后续的处理引入不可控制的误差。本文采用基于遗传算法的独立分量方法实现多通道脑电信号的盲源分离。结果：通过相关临床专家检验，认为该方法基本能够去除心电噪声，和参考心电信号对比具有一致性。结论：通过从北京某三甲医院癫痫中心采集到的患者脑电数据进行测试，对比试验表明，该方法是一种稳健高效的处理方法，符合并行运算的特点，整套算法可以移植到基于DSP的嵌入式系统架构上，具有一定的实用价值。     

基于形态学滤波器的棘波提取技术

本研究提出了一种基于形态学滤波器的脑电棘波检测方法。首先，选择三角形为结构元素，采用数学形态学中的级联开闭和闭开运算分别对脑电信号进行处理，去除癫痫样瞬态信号；然后对级联开闭和闭开运算采用平均加权的方法，消除统计偏倚现象，得到背景脑电信号；最后将处理结果和原始脑电信号做差，提取出脑电信号中的棘波。通过仿真信号的对比分析和临床癫痫脑电信号的应用，表明该方法适用于双向棘波的提取，与以往的方法相比，背景脑电的抑制能力更强，提取出的信号特征成分更完整。     

基于典型相关分析和小波变换的眼龟伪迹去除

目的针对脑电信号中眼电伪迹去除尚存在的问题，提出一种基于典型相关分析与小波变换的（wavelet—enhanced canonical correlation analysis，wCCA）自动去除眼电伪迹的算法。方法首先，充分利用脑电信号和眼电伪迹的空间分布特征，将基于典型相关分析的盲源分离算法分别应用于左右脑区的混合信号中，从而保证典型相关分析分解得到的第一个典型相关变量（即左右脑区之间的最公共成分），就是眼电伪迹分量。然后为了恢复泄漏在该伪迹分量中的脑电成分，对伪迹分量进行小波阈值滤波，将高于某一阈值的小波系数置零，而保留低于阈值的系数。结果与其他三种基于盲源分离去除眼电伪迹的方法相比较，该方法在有效地自动去除眼电伪迹的同时，很好地保留了潜在的脑电信号，去除效果明显优于其他三种方法。结论由于该算法简单，处理速度较快，因此应用于实时的脑机接口系统中更具优越性，为后续脑电信号的特征提取和分类分析提供了良好的基础。     

脑电计算机实时监护系统的研制

本文介绍了脑电信号的特点和采集方法,以及基于８０５８６ 微机的脑电实时监护系统的硬件结构,阐述了系统硬件和软件的设计思想和实现方法。该系统的软件具有良好的兼容性,可靠性和实时性。系统功能齐全,可适用于脑电监护和脑电反馈的研究。     

基于固有模态分解和深度学习的抑郁症脑电信号分类分析

以采集到的抑郁症患者和正常人的脑电信号为基础,采用固有模态分解算法对原始信号去噪处理,通过卷积神经网络对抑郁症患者和正常人进行分类分析。首先通过脑电信号的采集实验,采集15位抑郁症患者和15位正常人对照组Fp1的静息态脑电信号;之后对采集到的静息态脑电进行去噪处理,脑电去噪处理主要包括固有模态分解算法对原始信号的分解获得不同层次的IMF分量,对IMF分量进行频域分析,通过硬阈值的方法剔除原始信号中的噪声信号;最后采用卷积神经网络对抑郁症患者和正常人对照组进行二值分类,结果相较于传统的特征提取-机器学习算法,分类准确率明显提高。     

Wireless biopotential acquisition system for portable healthcare monitoring

A complete biopotential acquisition system with an analogue front-end (AFE) chip is proposed for portable healthcare monitoring. A graphical user interface (GUI) is also implemented to display the extracted biopotential signals in real-time on a computer for patients or in a hospital via the internet for doctors. The AFE circuit defines the quality of the acquired biosignals. Thus, an AFE chip with low power consumption and a high common-mode rejection ratio (CMRR) was implemented in the TSMC 0.18-μm CMOS process. The measurement results show that the proposed AFE, with a core area of 0.1 mm(2), has a CMRR of 90 dB, and power consumption of 21.6 μW. Biopotential signals of electroencephalogram (EEG), electrocardiogram (ECG) and electromyogram (EMG) were measured to verify the proposed system. The board size of the proposed system is 6 cm × 2.5 cm and the weight is 30 g. The total power consumption of the proposed system is 66 mW.     

Wireless biopotential acquisition system for portable healthcare monitoring

A complete biopotential acquisition system with an analogue front-end (AFE) chip is proposed for portable healthcare monitoring. A graphical user interface (GUI) is also implemented to display the extracted biopotential signals in real-time on a computer for patients or in a hospital via the internet for doctors. The AFE circuit defines the quality of the acquired biosignals. Thus, an AFE chip with low power consumption and a high common-mode rejection ratio (CMRR) was implemented in the TSMC 0.18-μm CMOS process. The measurement results show that the proposed AFE, with a core area of 0.1 mm², has a CMRR of 90 dB, and power consumption of 21.6?μW. Biopotential signals of electroencephalogram (EEG), electrocardiogram (ECG) and electromyogram (EMG) were measured to verify the proposed system. The board size of the proposed system is 6 cm×2.5 cm and the weight is 30 g. The total power consumption of the proposed system is 66 mW.     

低噪声、低纹波头皮脑电信号采集的前端电路

基于SMIC 0.18 μm CMOS工艺设计了一种由前置放大器、纹波抑制电路、增益可编程放大器和外置低通滤波器组成的低噪声头皮脑电信号采集前端电路。其中前置放大器采用斩波技术来降低低频噪声和失调电压,其结构为电容耦合的斩波稳定放大器,其有源放大部分由折叠式共源共栅级和共源级构成的两级全差分放大器构成,以获得较高的开环增益;同时在前置放大器后引入了阻容耦合电路来抑制在斩波频率处产生的纹波。可编程增益放大器采用可调电容阵列构成的电容负反馈放大器以实现增益逐级可调。Spectre后仿真结果表明,前置放大器的增益为40 dB,共模抑制比可达131 dB,电源抑制比90 dB,输入等效噪声772 nV/sqrt(Hz)@100 Hz;可编程增益放大器的增益分别为12、20、25 dB;低通滤波器的截止频率为1 kHz;纹波抑制电路对在斩波频率处的纹波具有400倍的抑制效果。该前端电路的总增益为40~65 dB,通频带为1 Hz~1 kHz。     

A single supply biopotential amplifier.

A biopotential amplifier for single supply operation is presented. It uses a Driven Right Leg Circuit (DRL) to drive the patient's body to a DC common mode voltage, centering biopotential signals with respect to the amplifier's input voltage range. This scheme ensures proper range operation when a single power supply is used. The circuit described is especially suited for low consumption, battery-powered applications, requiring a single battery and avoiding switching voltage inverters to achieve dual supplies. The generic circuit is described and, as an example, a biopotential amplifier with a gain of 60 dB and a DC input range of +/-200 mV was implemented using low power operational amplifiers. A Common Mode Rejection Ratio (CMRR) of 126 dB at 50 Hz was achieved without trimming.     

Two-electrode low supply voltage electrocardiogram signal amplifier

Portable biomedical instrumentation has become an important part of diagnostic and treatment instrumentation, including telemedicine applications. Lowvoltage and low-power design tendencies prevail. Modern battery cell voltages in the range of 3–3.6V require appropriate circuit solutions. A two-electrode biopotential amplifier design is presented, with a high common-mode rejection ratio (CMRR), high input voltage tolerance and standard first-order high-pass characteristic. Most of these features are due to a high-gain first stage design. The circuit makes use of passive components of popular values and tolerances. Powered by a single 3V source, the amplifier tolerates ±1V common mode voltage, ±50μA common mode current and 2V input DC voltage, and its worst-case CMRR is 60 dB. The amplifier is intended for use in various applications, such as Holter-type monitors, defibrillators, ECG monitors, biotelemetry devices etc.     

面向脑机接口的脑电采集设备硬件系统综述

脑机接口技术(BCI)可为人脑和外界建立一种全新的直接的交互方式,具有非常广阔的应用前景。脑电采集设备作为脑机接口采集信号的重要手段和途径,是其技术的关键和基础,已得到广泛关注。近年脑机接口研究呈爆炸式增长,各种脑电采集技术与应用不断涌现。未来,脑电采集设备在科学、医疗、军事、生活等领域具有巨大的应用潜力。为理清目前脑电采集设备硬件系统的发展现状和发展方向,从基本组成结构、性能优化电路以及现有产品等方面进行剖析。归纳脑电采集设备的4个主要组成部分,进一步分类并讨论脑电采集设备性能优化方法;对比现有的主流产品的关键指标,探讨它们的功能特性;分析现有脑电采集设备的不足之处,并对其发展趋势进行展望。     

A low noise isolated amplifier system for electrophysiological measurements: basic considerations and design

An instrumentation amplifier for medical and biological research applications is described. It consists of an isolated preamplifier with a high signal/ noise ratio and a computer or microprocessor controlled main amplifier that incorporates variable highpass and lowpass filtering. The preamplifier includes an input stage, a compensation circuit for electrode offset voltage and an isolation amplifier. Recently introduced ultra low noise operational amplifiers used in the input stage ensure good noise behaviour. Isolation is obtained with an inductive isolation amplifier, which also delivers the supply voltages for the input stage. Accordingly, battery power is not needed, in contrast with optically isolated amplifiers. The main amplifier has computer control for gain, lowpass cut-off frequency and highpass cut-off frequency settings. This features dynamic changes in signal conditioning and data acquisition by the same computer that collects the data. Data acquisition and noise behaviour are discussed. The filter configuration is designed in such a way that several user-defined options can be implemented on the same circuit board. The trade-off between the order of the lowpass filters, their roll-off, and the sampling rate which is used in the subsequent computing system, are of major importance and will be discussed. The amplifier system is at present in sue in laboratories of neurology (BEP, EEG), opthalmology (VEP, EOG, ERG) and cardiology (noninvasive His bundle recordings) from which some results will be presented. The amplifiers form a part of a microprocessor-controlled data acquisition system.     

基于薄膜压力传感器的脉搏信号测试系统

人体脉搏信号携带有丰富的与健康相关的生理信息.为了方便脉搏信号的采集和研究,本文设计出一种采用轻触式薄膜按键面板工艺的自制脉搏压力传感器,并使用了常见的555和V-f转换器件将传感器压力信号转换为电信号,再通过后续设计的滤波、陷波放大电路以及电路的干扰抑制处理,系统就可以获取高质量的脉搏波信号.传感器的性能测试以及示波器的应用结果表明,基于薄膜压力传感器的脉搏信号测试系统工作性能稳定,并通过示波器的串口通信扩展模块或者LABVIEW采集卡可以在计算机上显示出采集的脉搏信号,此测试系统可用于脉象信息的采集与研究.     

Acquiring research‐grade ERPs on a shoestring budget: A comparison of a modified Emotiv and commercial SynAmps EEG system

This study compared the performance of a low‐cost wireless EEG system to a research‐grade EEG system on an auditory oddball task designed to elicit N200 and P300 ERP components. Participants were 15 healthy adults (6 female) aged between 19 and 40 (M = 28.56; SD = 6.38). An auditory oddball task was presented comprising 1,200 presentations of a standard tone interspersed by 300 trials comprising a deviant tone. EEG was simultaneously recorded from a modified Emotiv EPOC and a NeuroScan SynAmps RT EEG system. The modifications made to the Emotiv system included attaching research grade electrodes to the Bluetooth transmitter. Additional modifications enabled the Emotiv system to connect to a portable impedance meter. The cost of these modifications and portable impedance meter approached the purchase value of the Emotiv system. Preliminary analyses revealed significantly more trials were rejected from data acquired by the modified Emotiv compared to the SynAmps system. However, the ERP waveforms captured by the Emotiv system were found to be highly similar to the corresponding waveform from the SynAmps system. The latency and peak amplitude of N200 and P300 components were also found to be similar between systems. Overall, the results indicate that, in the context of an oddball task, the ERP acquired by a low‐cost wireless EEG system can be of comparable quality to research‐grade EEG acquisition equipment.