技术领域Technical Field
本发明涉及医疗器械领域，具体涉及一种便携式脑部血流信号监测系统。The present invention relates to the field of medical devices, and in particular to a portable cerebral blood flow signal monitoring system.
背景技术Background technique
脑是人体的中枢神经器官，位于颅腔之内，包括大脑皮层、小脑和脑干三部分。大脑皮层是脑的重要组成部分，由巨大数量的神经元组成；据估计，人体大脑皮层的神经元总数约有140亿个。临床脑保护是临床医学中的一个重要概念，其基本方法是通过各种临床监护和治疗手段，保证大脑神经元的生命活动，从而实现大脑的正常生理功能。随着经济的发展和社会的进步，临床脑保护也受到越来越多的重视。The brain is the central nervous system organ of the human body, located in the cranial cavity, and includes three parts: the cerebral cortex, cerebellum, and brainstem. The cerebral cortex is an important part of the brain, composed of a huge number of neurons; it is estimated that the total number of neurons in the human cerebral cortex is about 14 billion. Clinical brain protection is an important concept in clinical medicine. Its basic method is to ensure the vital activities of brain neurons through various clinical monitoring and treatment methods, thereby achieving the normal physiological function of the brain. With the development of the economy and the progress of society, clinical brain protection has also received more and more attention.
卒中是我国居民死亡的主要病因。据脑血管病大数据平台统计，我国的卒中发病率、患病率、复发率和死亡率均居高不下。然而，目前针对卒中的有效检测方法主要有颅脑CT、颅脑核磁共振成像、脑血管造影等。这些设备庞大、操作复杂，无法实现随时随地的应用。专用的CT救护车成本高昂，因此无法在每一辆救护车上配备移动CT。Stroke is the leading cause of death among Chinese residents. According to statistics from the cerebrovascular disease big data platform, the incidence, prevalence, recurrence and mortality of stroke in my country remain high. However, the current effective detection methods for stroke mainly include cranial CT, cranial magnetic resonance imaging, cerebral angiography, etc. These devices are large and complex to operate, and cannot be used anytime and anywhere. Dedicated CT ambulances are expensive, so it is impossible to equip every ambulance with a mobile CT.
在急诊中，经常会出现由于急性心肌梗死、意外事故、医疗事件等导致的心脏骤停病例，随着心脏停搏，脑部也会出现缺血，而脑细胞停止供血五分钟以上即可能导致脑死亡，此后患者将失去任何意识，只维持心肺功能，生存质量丧失。因此，在急救过程中，是否进行了有效的胸外按压，从而及时恢复脑部供血这一结果至关重要。但目前临床上尚无便携式可实时监测颅内血流的设备，且现有的PPG信号监测装置无法准确判断监测到的血流信号是来自颅外皮肤还是颅内大脑，且难以在心肺复苏极低脑血流的情况下准确判断脑血流信号。In the emergency department, there are often cases of cardiac arrest caused by acute myocardial infarction, accidents, medical events, etc. As the heart stops beating, the brain will also suffer from ischemia, and brain cells may die if they stop supplying blood for more than five minutes. After that, the patient will lose any consciousness, only maintain cardiopulmonary function, and lose the quality of life. Therefore, in the process of first aid, whether effective chest compressions are performed to restore blood supply to the brain in time is crucial. However, there is currently no portable device that can monitor intracranial blood flow in real time in clinical practice, and the existing PPG signal monitoring device cannot accurately determine whether the monitored blood flow signal comes from the extracranial skin or the intracranial brain, and it is difficult to accurately judge the cerebral blood flow signal in the case of extremely low cerebral blood flow during cardiopulmonary resuscitation.
发明内容Summary of the invention
本发明提供一种便携式脑部血流信号监测系统，以解决上述至少一个技术问题。The present invention provides a portable cerebral blood flow signal monitoring system to solve at least one of the above technical problems.
本发明解决上述技术问题的技术方案如下：一种便携式脑部血流信号监测系统，包括控制器以及与所述控制器电连接的双波长LED光源和双路探测器；The technical solution of the present invention to solve the above technical problems is as follows: a portable brain blood flow signal monitoring system, comprising a controller and a dual-wavelength LED light source and a dual-path detector electrically connected to the controller;
所述控制器用于消除所述双波长LED光源的频闪；The controller is used to eliminate the flicker of the dual-wavelength LED light source;
所述双波长LED光源用于在所述控制器的控制下向脑部待测部位发出频闪消除的双波长光信号；The dual-wavelength LED light source is used to emit a dual-wavelength light signal with stroboscopic elimination to the part of the brain to be tested under the control of the controller;
所述双路探测器用于接收从所述脑部待测部位反射的双波长光信号，并从所述双波长光信号中提取出双波长光强信号，且依次对所述双波长光强信号进行电信号转换处理、滤波处理、信号跟随及放大处理和AD转换处理，得到双波长原始PPG信号；The dual-path detector is used to receive the dual-wavelength light signal reflected from the brain part to be tested, and extract the dual-wavelength light intensity signal from the dual-wavelength light signal, and sequentially perform electrical signal conversion processing, filtering processing, signal following and amplification processing and AD conversion processing on the dual-wavelength light intensity signal to obtain a dual-wavelength original PPG signal;
所述控制器还用于接收所述双路探测器探测到的双波长原始PPG信号，并对所述双波长原始PPG信号进行降噪滤波处理，以消除所述双波长原始PPG信号中的工频噪声、基线漂移和运动伪影噪声，得到双波长优化PPG信号，且对所述双波长优化PPG信号进行特征参数提取处理，以根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号。The controller is also used to receive the dual-wavelength original PPG signal detected by the dual-path detector, and perform noise reduction filtering on the dual-wavelength original PPG signal to eliminate power frequency noise, baseline drift and motion artifact noise in the dual-wavelength original PPG signal to obtain a dual-wavelength optimized PPG signal, and perform feature parameter extraction processing on the dual-wavelength optimized PPG signal to distinguish intracranial and extracranial brain blood flow signals from the dual-wavelength optimized PPG signal according to the feature parameters.
在上述技术方案的基础上，本发明还可以做如下改进。Based on the above technical solution, the present invention can also be improved as follows.
进一步，所述双路探测器中的任一路探测器均包括：Furthermore, any one of the two-path detectors includes:
APD，输入端设置有滤光片，电源端接入直流电源；APD, the input end is provided with a filter, and the power supply end is connected to a DC power supply;
电感L，电连接于所述APD的接地端与地之间；an inductor L, electrically connected between a ground terminal of the APD and the ground;
电容C1，电连接于所述APD的电源端与地之间；A capacitor C1 is electrically connected between a power supply terminal of the APD and a ground;
电阻R1，电连接于所述APD的输出端与地之间；A resistor R1 is electrically connected between the output terminal of the APD and ground;
电容C2，一端电连接于所述APD的输出端；A capacitor C2, one end of which is electrically connected to the output end of the APD;
电阻R2，一端电连接所述电容C2的另一端，另一端接地；A resistor R2, one end of which is electrically connected to the other end of the capacitor C2, and the other end of which is grounded;
运算放大器U1，正向输入端电连接所述电阻R2的一端；The operational amplifier U1 has a positive input terminal electrically connected to one end of the resistor R2;
电阻R3，电连接于所述运算放大器U1的负向输入端与输出端之间；The resistor R3 is electrically connected between the negative input terminal and the output terminal of the operational amplifier U1;
运算放大器U2，正向输入端电连接所述运算放大器U1的输出端；An operational amplifier U2, a positive input terminal electrically connected to the output terminal of the operational amplifier U1;
电阻R4，电连接于所述运算放大器U2的负向输入端与地之间；The resistor R4 is electrically connected between the negative input terminal of the operational amplifier U2 and the ground;
电阻R5，电连接于所述运算放大器U2的负向输入端与输出端之间；The resistor R5 is electrically connected between the negative input terminal and the output terminal of the operational amplifier U2;
AD转换器，输入端电连接所述运算放大器U2的输出端，输出端电连接所述控制器。The AD converter has an input end electrically connected to the output end of the operational amplifier U2, and an output end electrically connected to the controller.
进一步，所述控制器内设置有：Furthermore, the controller is provided with:
PWM模块，其用于输入电源电压，并将所述电源电压调整成PWM脉冲电压且输出；A PWM module, which is used to input a power supply voltage, and adjust the power supply voltage into a PWM pulse voltage and output it;
PWM转电压模块，其用于将所述PWM模块输出的PWM脉冲电压转换成稳定的LED驱动电压，以驱动所述双波长LED光源中任一波长LED光源发出频闪消除的一波长光信号。The PWM to voltage module is used to convert the PWM pulse voltage output by the PWM module into a stable LED driving voltage to drive any wavelength LED light source in the dual-wavelength LED light source to emit a single-wavelength light signal with flicker eliminated.
进一步，所述控制器内设置有两个降噪滤波处理模块，其中一个所述降噪滤波处理模块与所述双路探测器中的一路探测器电连接，另一个所述降噪滤波处理模块与所述双路探测器中的另一路探测器电连接；任一个所述降噪滤波处理模块均具体包括：Furthermore, two noise reduction filter processing modules are provided in the controller, one of which is electrically connected to one of the two-path detectors, and the other is electrically connected to the other of the two-path detectors; any one of the noise reduction filter processing modules specifically includes:
微分处理单元，其用于接收任一路探测器探测到的一波长原始PPG信号，并对该波长原始PPG信号进行微分处理，得到微分信号；A differential processing unit, which is used to receive an original PPG signal of a wavelength detected by any detector, and perform differential processing on the original PPG signal of the wavelength to obtain a differential signal;
均值和标准差计算单元，其用于计算出所述微分信号的均值和标准差；A mean and standard deviation calculation unit, which is used to calculate the mean and standard deviation of the differential signal;
突变噪声去除单元，其用于根据所述微分信号的均值和标准差，标记出所述微分信号中的突变噪声点并剔除，并在所述微分信号中利用所述突变噪声点的前后点的平均值补齐因剔除所述突变噪声点而形成的缺失点位，得到新微分信号；a mutation noise removal unit, which is used to mark and remove mutation noise points in the differential signal according to the mean and standard deviation of the differential signal, and to fill the missing points in the differential signal formed by removing the mutation noise points using the average values of the points before and after the mutation noise points to obtain a new differential signal;
积分单元，其用于对所述新微分信号进行积分还原处理，得到积分信号；An integration unit, which is used to perform integration restoration processing on the new differential signal to obtain an integrated signal;
高频噪声去除单元，其用于对所述积分信号进行平滑滤波处理，得到平滑滤波信号；A high-frequency noise removal unit, which is used to perform smoothing filtering on the integrated signal to obtain a smoothing filtered signal;
基线漂移去除单元，其用于利用经验模态分解方法对所述平滑滤波信号进行处理，得到多个本征模态函数，并将多个本征模态函数进行相加处理，得到重构信号；A baseline drift removal unit, which is used to process the smoothing filter signal using an empirical mode decomposition method to obtain a plurality of intrinsic mode functions, and add the plurality of intrinsic mode functions to obtain a reconstructed signal;
信号提取单元，其用于从所述重构信号中提取出感兴趣的本征模态函数并相加，得到一波长优化PPG信号。A signal extraction unit is used to extract the intrinsic mode function of interest from the reconstructed signal and add them to obtain a wavelength optimized PPG signal.
进一步，在所述突变噪声去除单元中，标记所述微分信号中的突变噪声点的公式为：Furthermore, in the mutation noise removal unit, the formula for marking the mutation noise point in the differential signal is:
其中，d(i)表示所述微分信号中的第i个微分信号值，mean表示所述微分信号的均值，std表示所述微分信号的标准差，mark(d(i))表示所述微分信号中的第i个微分信号值所对应点位的标记值，且当mark(d(i))＝0时表示所述微分信号中的第i个微分信号值所对应的点位为异常的突变噪声点位，当mark(d(i))＝1时表示所述微分信号中的第i个微分信号值所对应的点位为正常的点位。Wherein, d(i) represents the i-th differential signal value in the differential signal, mean represents the mean of the differential signal, std represents the standard deviation of the differential signal, mark(d(i)) represents the mark value of the point corresponding to the i-th differential signal value in the differential signal, and when mark(d(i))=0, it means that the point corresponding to the i-th differential signal value in the differential signal is an abnormal mutation noise point, and when mark(d(i))=1, it means that the point corresponding to the i-th differential signal value in the differential signal is a normal point.
进一步，所述高频噪声去除单元具体用于对所述积分信号进行Savitzky-Golay平滑滤波处理，得到平滑滤波信号。Furthermore, the high-frequency noise removal unit is specifically used to perform Savitzky-Golay smoothing filtering on the integrated signal to obtain a smoothed filtered signal.
进一步，所述控制器内设置有特征参数提取模块，所述特征参数提取模块用于：Furthermore, a characteristic parameter extraction module is provided in the controller, and the characteristic parameter extraction module is used to:
根据所述双波长优化PPG信号中的各波长优化PPG信号，提取出与各波长优化PPG信号的脉图面积相关的PPG波形特征量；According to each wavelength optimized PPG signal in the dual-wavelength optimized PPG signal, a PPG waveform feature quantity related to the pulse graph area of each wavelength optimized PPG signal is extracted;
将与各波长优化PPG信号的脉图面积相关的PPG波形特征量均作为所述特征参数，并根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号；The PPG waveform feature quantities related to the pulse graph area of each wavelength-optimized PPG signal are used as the feature parameters, and the intracranial and extracranial brain blood flow signals are distinguished from the dual-wavelength-optimized PPG signal according to the feature parameters;
其中，所述PPG波形特征量的提取公式为：Among them, the extraction formula of the PPG waveform feature is:
具体的，K表示与任一波长优化PPG信号的脉图面积相关的PPG波形特征量，Pm表示所述双波长优化PPG信号中的任一波长优化PPG信号的平均动脉压，Ps和Pd分别表示所述双波长优化PPG信号中的任一波长优化PPG信号的收缩压和舒张压。Specifically, K represents a PPG waveform feature related to the pulse graph area of any wavelength-optimized PPG signal,Pm represents the mean arterial pressure of any wavelength-optimized PPG signal in the dual-wavelength-optimized PPG signals, andPs andPd respectively represent the systolic pressure and diastolic pressure of any wavelength-optimized PPG signal in the dual-wavelength-optimized PPG signals.
进一步，所述控制器内设置有特征参数提取模块，所述特征参数提取模块用于：Furthermore, a characteristic parameter extraction module is provided in the controller, and the characteristic parameter extraction module is used to:
根据所述双波长优化PPG信号中的各波长优化PPG信号，绘制出各波长优化PPG信号的PPG波图；According to each wavelength optimized PPG signal in the dual-wavelength optimized PPG signal, a PPG wave graph of each wavelength optimized PPG signal is drawn;
从各波长优化PPG信号的PPG波图中提取出曲线拐点或/和曲线斜率；Extracting a curve inflection point and/or a curve slope from the PPG wave graph of each wavelength-optimized PPG signal;
将从各波长优化PPG信号的PPG波图中提取出的曲线拐点或/和曲线斜率均作为所述特征参数，并根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号。The curve inflection point and/or the curve slope extracted from the PPG wave graph of each wavelength optimized PPG signal are used as the characteristic parameters, and the intracranial and extracranial brain blood flow signals are distinguished from the dual-wavelength optimized PPG signal based on the characteristic parameters.
进一步，所述控制器内设置有特征参数提取模块，所述特征参数提取模块用于：Furthermore, a characteristic parameter extraction module is provided in the controller, and the characteristic parameter extraction module is used to:
将所述双波长优化PPG信号中的各波长优化PPG信号均分解成分别与三个高斯函数一一对应的三个钟形波；Decomposing each wavelength-optimized PPG signal in the dual-wavelength-optimized PPG signal into three bell-shaped waves corresponding to three Gaussian functions respectively;
确定出各波长优化PPG信号中的三个钟形波的幅度、时间和宽度；Determine the amplitude, time and width of the three bell-shaped waves in each wavelength-optimized PPG signal;
将各波长优化PPG信号中的三个钟形波的幅度、时间和宽度均作为所述特征参数，并根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号；The amplitude, time and width of the three bell-shaped waves in each wavelength-optimized PPG signal are used as the characteristic parameters, and the intracranial and extracranial brain blood flow signals are distinguished from the dual-wavelength-optimized PPG signal according to the characteristic parameters;
或，所述特征参数提取模块用于：Or, the feature parameter extraction module is used for:
对所述双波长优化PPG信号中的各波长优化PPG信号进行傅里叶变换，得到各波长优化PPG信号的频域信息；Performing Fourier transform on each wavelength optimized PPG signal in the dual-wavelength optimized PPG signal to obtain frequency domain information of each wavelength optimized PPG signal;
获取各波长优化PPG信号的频域信息中的特征峰值，并进行高斯拟合，得到各波长优化PPG信号的高斯拟合函数；Obtaining characteristic peaks in the frequency domain information of each wavelength optimized PPG signal, and performing Gaussian fitting to obtain the Gaussian fitting function of each wavelength optimized PPG signal;
获取各波长优化PPG信号的高斯拟合函数的半高全宽参数；Obtaining the full width at half maximum parameters of the Gaussian fitting function of the optimized PPG signal at each wavelength;
将各波长优化PPG信号的高斯拟合函数的半高全宽参数均作为所述特征参数，并根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号。The half-width parameters of the Gaussian fitting function of each wavelength-optimized PPG signal are used as the characteristic parameters, and the intracranial and extracranial brain blood flow signals are distinguished from the dual-wavelength optimized PPG signal according to the characteristic parameters.
进一步，所述控制器内设置有特征参数提取模块，所述特征参数提取模块用于：Furthermore, a characteristic parameter extraction module is provided in the controller, and the characteristic parameter extraction module is used to:
将所述双波长优化PPG信号中的各波长优化PPG信号划分为多个独立的心动周期；Dividing each wavelength-optimized PPG signal in the dual-wavelength-optimized PPG signal into a plurality of independent cardiac cycles;
分别对各波长优化PPG信号中的每一个心动周期进行傅里叶拟合，得到各波长优化PPG信号的多个傅里叶拟合函数；Performing Fourier fitting on each cardiac cycle in each wavelength-optimized PPG signal to obtain a plurality of Fourier fitting functions of each wavelength-optimized PPG signal;
获取各波长优化PPG信号的多个傅里叶拟合函数的系数，并进行概率分布统计，得到各波长优化PPG信号的多个傅里叶拟合函数的系数的概率分布情况；Obtaining coefficients of multiple Fourier fitting functions of each wavelength optimized PPG signal, and performing probability distribution statistics to obtain probability distribution of coefficients of multiple Fourier fitting functions of each wavelength optimized PPG signal;
将各波长优化PPG信号的多个傅里叶拟合函数的系数的概率分布情况均作为所述特征参数，并根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号。The probability distribution of the coefficients of the multiple Fourier fitting functions of each wavelength-optimized PPG signal is used as the characteristic parameter, and the intracranial and extracranial brain blood flow signals are distinguished from the dual-wavelength optimized PPG signal according to the characteristic parameter.
本发明的有益效果是：本发明采用由控制器以及与所述控制器电连接的双波长LED光源和双路探测器构成的脑部血流信号监测系统，可以对脑部血流情况的实时无创监测功能，获得了清晰明确的脑部血流信号，为后续的相关研究提供了原始数据支撑；同时，本发明体积小，操作简便，对使用环境要求低，对操作人员无额外要求；另外，本发明可以对运动状态下极低(≤30％正常脑血流)脑部血流信号进行的提取，可以在心肺复苏等治疗过程中对患者的情况进行实时监测，实现了对信号的滤波去噪功能，得到了干净可辨识的脑部PPG信号；以及对PPG信号进行特征参数提取，实现了对颅内外信号的辨别，可以对监测到的信号成分进行分析，判别信号中是否含有颅内和颅外的相关成分。The beneficial effects of the present invention are as follows: the present invention adopts a brain blood flow signal monitoring system composed of a controller and a dual-wavelength LED light source and a dual-path detector electrically connected to the controller, which can realize the real-time non-invasive monitoring function of the brain blood flow condition, obtain clear and definite brain blood flow signals, and provide original data support for subsequent related research; at the same time, the present invention is small in size, easy to operate, has low requirements for the use environment, and has no additional requirements for operators; in addition, the present invention can extract extremely low (≤30% normal cerebral blood flow) brain blood flow signals in a motion state, can monitor the patient's condition in real time during treatment such as cardiopulmonary resuscitation, realize the filtering and denoising function of the signal, and obtain a clean and identifiable brain PPG signal; and extract characteristic parameters of the PPG signal, realize the discrimination of intracranial and extracranial signals, and can analyze the monitored signal components to determine whether the signal contains relevant intracranial and extracranial components.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
图1为本发明一种便携式脑部血流信号监测系统的结构示意图；FIG1 is a schematic structural diagram of a portable brain blood flow signal monitoring system of the present invention;
图2为本发明一种便携式脑部血流信号监测系统的实物及使用场景图；FIG2 is a diagram of a portable brain blood flow signal monitoring system and its use scenario according to the present invention;
图3为双路探测器的部分电路原理图；FIG3 is a partial circuit schematic diagram of a dual-path detector;
图4为控制器内的部分结构框图；FIG4 is a partial structural block diagram of the controller;
图5为EEMD的处理流程图；FIG5 is a processing flow chart of EEMD;
图6为光在脑部的传输路径示意图；FIG6 is a schematic diagram of the transmission path of light in the brain;
图7为血管内血液容积变化图；FIG7 is a diagram showing changes in blood volume in blood vessels;
图8为脉搏波信号的波形图；FIG8 is a waveform diagram of a pulse wave signal;
图9为一个脉搏波周期的示意图；FIG9 is a schematic diagram of a pulse wave cycle;
图10为PPG的特征点的示意图；FIG10 is a schematic diagram of characteristic points of PPG;
图11为高斯函数拟合示意图；Fig. 11 is a schematic diagram of Gaussian function fitting;
图12为在频域对一个优化PPG信号进行高斯拟合的处理结果示意图；FIG12 is a schematic diagram of the processing result of Gaussian fitting of an optimized PPG signal in the frequency domain;
图13为基于独立的心动周期对一个优化PPG信号进行傅里叶拟合的结果示意图；FIG13 is a schematic diagram showing the result of Fourier fitting of an optimized PPG signal based on an independent cardiac cycle;
图14为实例中按照独立的心动周期进行傅里叶拟合得到的部分傅里叶系数的概率分布示意图。FIG. 14 is a schematic diagram showing the probability distribution of some Fourier coefficients obtained by Fourier fitting according to independent cardiac cycles in the example.
具体实施方式Detailed ways
以下结合附图对本发明的原理和特征进行描述，所举实例只用于解释本发明，并非用于限定本发明的范围。The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.
如图1所示，一种便携式脑部血流信号监测系统，包括控制器1以及与所述控制器1电连接的双波长LED光源2和双路探测器3；As shown in FIG1 , a portable brain blood flow signal monitoring system includes a controller 1 and a dual-wavelength LED light source 2 and a dual-path detector 3 electrically connected to the controller 1;
所述控制器1用于消除所述双波长LED光源2的频闪；The controller 1 is used to eliminate the flicker of the dual-wavelength LED light source 2;
所述双波长LED光源2用于在所述控制器1的控制下向脑部待测部位4发出频闪消除的双波长光信号；The dual-wavelength LED light source 2 is used to emit a dual-wavelength light signal with stroboscopic elimination to the brain test site 4 under the control of the controller 1;
所述双路探测器3用于接收从所述脑部待测部位4反射的双波长光信号，并从所述双波长光信号中提取出双波长光强信号，且依次对所述双波长光强信号进行电信号转换处理、滤波处理、信号跟随及放大处理和AD转换处理，得到双波长原始PPG信号；The dual-path detector 3 is used to receive the dual-wavelength light signal reflected from the brain test site 4, and extract the dual-wavelength light intensity signal from the dual-wavelength light signal, and sequentially perform electrical signal conversion processing, filtering processing, signal following and amplification processing and AD conversion processing on the dual-wavelength light intensity signal to obtain a dual-wavelength original PPG signal;
所述控制器1还用于接收所述双路探测器3探测到的双波长原始PPG信号，并对所述双波长原始PPG信号进行降噪滤波处理，以消除所述双波长原始PPG信号中的工频噪声、基线漂移和运动伪影噪声，得到双波长优化PPG信号，且对所述双波长优化PPG信号进行特征参数提取处理，以根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号。The controller 1 is also used to receive the dual-wavelength original PPG signal detected by the dual-path detector 3, and perform noise reduction filtering on the dual-wavelength original PPG signal to eliminate power frequency noise, baseline drift and motion artifact noise in the dual-wavelength original PPG signal to obtain a dual-wavelength optimized PPG signal, and perform feature parameter extraction processing on the dual-wavelength optimized PPG signal to distinguish intracranial and extracranial brain blood flow signals from the dual-wavelength optimized PPG signal according to the feature parameters.
本发明一种便携式脑部血流信号监测系统还包括直流电源5，所述直流电源5与所述双路探测器3电连接，用于为所述双路探测器3提供电源。The portable brain blood flow signal monitoring system of the present invention further includes a DC power supply 5 , which is electrically connected to the dual-path detector 3 and is used to provide power to the dual-path detector 3 .
图2为发明一种便携式脑部血流信号监测系统的实物图，从图2可以看出，双波长LED光源2和双路探测器3集成在一起；当脑部待测部位4存在毛发时，也可以准确测到颅内血流信号。FIG2 is a physical picture of a portable brain blood flow signal monitoring system. As can be seen from FIG2 , a dual-wavelength LED light source 2 and a dual-path detector 3 are integrated together; when hair exists in the brain area to be measured 4, the intracranial blood flow signal can also be accurately measured.
在一些实施例中，如图3所示，所述双路探测器中的任一路探测器均包括：In some embodiments, as shown in FIG3 , any one of the two-path detectors includes:
APD，输入端设置有滤光片，电源端接入直流电源；APD, the input end is provided with a filter, and the power supply end is connected to a DC power supply;
电感L，电连接于所述APD的接地端与地之间；an inductor L, electrically connected between a ground terminal of the APD and the ground;
电容C1，电连接于所述APD的电源端与地之间；A capacitor C1 is electrically connected between a power supply terminal of the APD and a ground;
电阻R1，电连接于所述APD的输出端与地之间；A resistor R1 is electrically connected between the output terminal of the APD and ground;
电容C2，一端电连接于所述APD的输出端；A capacitor C2, one end of which is electrically connected to the output end of the APD;
电阻R2，一端电连接所述电容C2的另一端，另一端接地；A resistor R2, one end of which is electrically connected to the other end of the capacitor C2, and the other end of which is grounded;
运算放大器U1，正向输入端电连接所述电阻R2的一端；The operational amplifier U1 has a positive input terminal electrically connected to one end of the resistor R2;
电阻R3，电连接于所述运算放大器U1的负向输入端与输出端之间；The resistor R3 is electrically connected between the negative input terminal and the output terminal of the operational amplifier U1;
运算放大器U2，正向输入端电连接所述运算放大器U1的输出端；An operational amplifier U2, a positive input terminal electrically connected to the output terminal of the operational amplifier U1;
电阻R4，电连接于所述运算放大器U2的负向输入端与地之间；The resistor R4 is electrically connected between the negative input terminal of the operational amplifier U2 and the ground;
电阻R5，电连接于所述运算放大器U2的负向输入端与输出端之间；The resistor R5 is electrically connected between the negative input terminal and the output terminal of the operational amplifier U2;
AD转换器(图3中未示出)，输入端电连接所述运算放大器U2的输出端，输出端电连接所述控制器。An AD converter (not shown in FIG. 3 ) has an input end electrically connected to the output end of the operational amplifier U2 , and an output end electrically connected to the controller.
在图3中，APD(雪崩光电二极管探测器)将光强的变化信号(即光强信号)转变为电流的变化信号(即电流信号)；电流的变化信号通过电阻R1转变为电压信号来进行表征；APD通过电感L1接地，用于降低部分直流分量；在APD上并联一个电容C1，用于滤除直流电源的部分交流分量；通过电容C2和电阻R2组成的RC高通滤波器滤除电压信号中的直流分量；将滤除了直流分量的电压信号输入运算放大器U1，其型号为LM324，通过运算放大器U1进行信号跟随，目的是使在后续的过程中阻抗匹配；将运算放大器U2输出的信号按照需求，调整电阻R4和电阻R5这两个电阻的大小，选择合适的放大倍数进行放大，然后通过端口OUT输出，端口OUT输出的电压信号输入AD转换器，通过AD转换器转换后得到一波长原始PPG信号。In Figure 3, the APD (avalanche photodiode detector) converts the light intensity change signal (i.e., light intensity signal) into a current change signal (i.e., current signal); the current change signal is converted into a voltage signal through a resistor R1 for characterization; the APD is grounded through an inductor L1 to reduce part of the DC component; a capacitor C1 is connected in parallel to the APD to filter out part of the AC component of the DC power supply; the DC component in the voltage signal is filtered out by an RC high-pass filter composed of a capacitor C2 and a resistor R2; the voltage signal with the DC component filtered out is input into an operational amplifier U1, whose model is LM324, and the operational amplifier U1 performs signal following in order to achieve impedance matching in the subsequent process; the signal output by the operational amplifier U2 is adjusted according to the requirements, the sizes of the two resistors R4 and R5 are adjusted, and the appropriate amplification factor is selected for amplification, and then output through the port OUT, and the voltage signal output from the port OUT is input into the AD converter, and a one-wavelength original PPG signal is obtained after conversion by the AD converter.
在一些实施例中，如图4所示，所述控制器内设置有：In some embodiments, as shown in FIG4 , the controller is provided with:
PWM模块，其用于输入电源电压，并将所述电源电压调整成PWM脉冲电压且输出；A PWM module, which is used to input a power supply voltage, and adjust the power supply voltage into a PWM pulse voltage and output it;
PWM转电压模块，其用于将所述PWM模块输出的PWM脉冲电压转换成稳定的LED驱动电压，以驱动所述双波长LED光源中任一波长LED光源发出频闪消除的一波长光信号。The PWM to voltage module is used to convert the PWM pulse voltage output by the PWM module into a stable LED driving voltage to drive any wavelength LED light source in the dual-wavelength LED light source to emit a single-wavelength light signal with flicker eliminated.
目前市面上大部分的LED光强控制为直接通过PWM调整脉冲宽度来调光，缺点是存在频闪。由于本发明是通过对光强的变化进行测量来反映心率信号的，所以光源的频闪对测量存在很大影响。故本发明通过PWM模块调整脉冲宽度来调节输入的电源电压，再将调控好的带有脉冲信号的电压通过PWM转电压模块输出稳定的电压，实现了对LED电源的调节功能，有效的消除了LED光源的频闪，使光源更加稳定和干净。At present, most of the LED light intensity controls on the market are dimming by directly adjusting the pulse width through PWM, and the disadvantage is that there is flicker. Since the present invention reflects the heart rate signal by measuring the change of light intensity, the flicker of the light source has a great influence on the measurement. Therefore, the present invention adjusts the input power supply voltage by adjusting the pulse width through the PWM module, and then outputs a stable voltage through the PWM to voltage module of the regulated voltage with the pulse signal, thereby realizing the regulation function of the LED power supply, effectively eliminating the flicker of the LED light source, and making the light source more stable and clean.
在一些实施例中，对PPG信号信噪比造成影响的噪声种类大概分为以下几种：由交流电源引起的工频噪声，是一种高频噪声，频率一般在50Hz；由呼吸或者轻微抖动造成的基线漂移，一般频率较低，小于1Hz；由人体肌肉颤动引起的肌电噪声(随机发生)，频率范围在5-2000Hz之间；受试者在测量信号时突然发生的微移动，对信号产生了突然的干扰，使得信号产生奇异噪声。本发明采用下述降噪滤波处理模块可以对以上噪声进行去除。In some embodiments, the types of noise that affect the signal-to-noise ratio of the PPG signal can be roughly divided into the following categories: power frequency noise caused by AC power supply, which is a high-frequency noise, generally at 50Hz; baseline drift caused by breathing or slight shaking, generally at a lower frequency, less than 1Hz; myoelectric noise caused by human muscle tremor (random occurrence), with a frequency range of 5-2000Hz; sudden micro-movement of the subject when measuring the signal, which suddenly interferes with the signal, causing the signal to generate strange noise. The present invention uses the following noise reduction filter processing module to remove the above noise.
所述控制器内设置有两个降噪滤波处理模块，其中一个所述降噪滤波处理模块与所述双路探测器中的一路探测器电连接，另一个所述降噪滤波处理模块与所述双路探测器中的另一路探测器电连接；任一个所述降噪滤波处理模块均具体包括：The controller is provided with two noise reduction filter processing modules, one of which is electrically connected to one of the two-way detectors, and the other is electrically connected to the other of the two-way detectors; any one of the noise reduction filter processing modules specifically includes:
微分处理单元，其用于接收任一路探测器探测到的一波长原始PPG信号，并对该波长原始PPG信号进行微分处理，得到微分信号；具体的，d＝gradient(s)，其中，d代表微分信号，s代表任一路探测器探测到的一波长原始PPG信号，gradient()代表微分函数。The differential processing unit is used to receive a single-wavelength original PPG signal detected by any detector, and perform differential processing on the single-wavelength original PPG signal to obtain a differential signal; specifically, d=gradient(s), wherein d represents the differential signal, s represents the single-wavelength original PPG signal detected by any detector, and gradient() represents the differential function.
均值和标准差计算单元，其用于计算出所述微分信号的均值和标准差；具体的，其中，d(i)表示所述微分信号d中的第i个微分信号值，mean表示所述微分信号d的均值，std表示所述微分信号d的标准差。A mean and standard deviation calculation unit, which is used to calculate the mean and standard deviation of the differential signal; specifically, Wherein, d(i) represents the i-th differential signal value in the differential signal d, mean represents the mean of the differential signal d, and std represents the standard deviation of the differential signal d.
突变噪声去除单元，其用于根据所述微分信号的均值和标准差，标记出所述微分信号中的突变噪声点并剔除，并在所述微分信号中利用所述突变噪声点的前后点的平均值补齐因剔除所述突变噪声点而形成的缺失点位，得到新微分信号d′；具体的，标记所述微分信号中的突变噪声点的公式为：The mutation noise removal unit is used to mark and remove the mutation noise points in the differential signal according to the mean and standard deviation of the differential signal, and to fill the missing points formed by removing the mutation noise points in the differential signal using the average value of the points before and after the mutation noise points to obtain a new differential signal d′; specifically, the formula for marking the mutation noise points in the differential signal is:
其中，mark(d(i))表示所述微分信号d中的第i个微分信号值所对应点位的标记值，且当mark(d(i))＝0时表示所述微分信号d中的第i个微分信号值所对应的点位为异常的突变噪声点位，当mark(d(i))＝1时表示所述微分信号d中的第i个微分信号值所对应的点位为正常的点位；将标记值为零的异常点直接去除，去除突变噪声后数据点缺失，为了补充数据值方便将信号进行还原，用缺失点前后点的平均值c(i)将缺失点代替,其代替公式为Wherein, mark(d(i)) represents the mark value of the point corresponding to the ith differential signal value in the differential signal d, and when mark(d(i))=0, it indicates that the point corresponding to the ith differential signal value in the differential signal d is an abnormal mutation noise point, and when mark(d(i))=1, it indicates that the point corresponding to the ith differential signal value in the differential signal d is a normal point; the abnormal point with a mark value of zero is directly removed, and the data point is missing after removing the mutation noise. In order to supplement the data value and restore the signal conveniently, the missing point is replaced by the average value c(i) of the points before and after the missing point, and the replacement formula is:
积分单元，其用于对所述新微分信号进行积分还原处理，得到积分信号；具体的，D＝integrl(d′)；其中，D代表积分信号，integrl()代表积分函数。The integration unit is used to perform integration reduction processing on the new differential signal to obtain an integral signal; specifically, D=integrl(d′); wherein D represents the integral signal, and integerl() represents the integral function.
高频噪声去除单元，其用于对所述积分信号进行行Savitzky-Golay平滑滤波处理，得到平滑滤波信号；具体的，T＝savgol-filter(D)；其中，T代表平滑滤波信号，savgol-filter()代表Savitzky-Golay平滑滤波处理；Savitzky-Golay(以下简称S-G)滤波器是一种在时域内基于局域多项式最小二乘法拟合的滤波方法。S-G滤波器的核心思想是在信号的每个局部窗口上进行多项式拟合。这个局部窗口是一个滑动的窗口，它在信号上移动；其本质是在时域上对窗口长度内的数据进行多项式拟合，而从频域上看，相当于通过低频数据，滤掉高频数据，实现高频去噪的目的；S-G滤波器的优势在于在滤除噪声的同时可以确保信号的形状、宽度不变；这对PPG信号来讲很重要，因为信号上包含了很多特征点，保证信号的形状宽度不变可以保留信号上的重要信息，S-G滤波步骤具体如下：A high-frequency noise removal unit is used to perform Savitzky-Golay smoothing filtering on the integrated signal to obtain a smoothed filtered signal; specifically, T = savgol-filter (D); wherein T represents the smoothed filtered signal, and savgol-filter () represents the Savitzky-Golay smoothing filtering process; the Savitzky-Golay (hereinafter referred to as S-G) filter is a filtering method based on local polynomial least squares fitting in the time domain. The core idea of the S-G filter is to perform polynomial fitting on each local window of the signal. This local window is a sliding window that moves on the signal; its essence is to perform polynomial fitting on the data within the window length in the time domain, and from the frequency domain, it is equivalent to filtering out high-frequency data through low-frequency data to achieve the purpose of high-frequency denoising; the advantage of the S-G filter is that it can ensure that the shape and width of the signal remain unchanged while filtering out noise; this is very important for PPG signals, because the signal contains many feature points, and ensuring that the shape and width of the signal remain unchanged can retain important information on the signal. The specific steps of the S-G filter are as follows:
(1)选择窗口长度和多项式阶数：根据积分信号D的采样率和采样时间选择适当的窗口长度和多项式阶数；窗口长度决定了每次拟合的数据点数，多项式的阶数决定了拟合的阶数；通常，窗口长度应该大于多项式阶数；(1) Select window length and polynomial order: Select appropriate window length and polynomial order according to the sampling rate and sampling time of the integrated signal D. The window length determines the number of data points for each fitting, and the order of the polynomial determines the order of the fitting. Usually, the window length should be larger than the polynomial order.
(2)构建局部窗口：在积分信号D上移动一个长度为窗口长度的滑动窗口，对于窗口内的数据点，进行多项式拟合。(2) Constructing a local window: Move a sliding window with a length equal to the window length on the integrated signal D, and perform polynomial fitting on the data points within the window.
(3)计算加权平均：使用拟合的多项式对窗口内的数据进行拟合，然后计算加权平均值；(3) Calculate the weighted average: fit the data in the window using the fitted polynomial, and then calculate the weighted average;
(4)滑动窗口：将窗口在积分信号D上滑动，重复(2)到(3)，直到整个积分信号D都被处理，得到平滑滤波信号T；(4) Sliding window: Slide the window on the integrated signal D, repeating (2) to (3) until the entire integrated signal D is processed to obtain the smoothed filtered signal T;
高频噪声去除单元的具体参数设置取决于原始PPG信号的采样率，在实际应用中需要反复观察和调试，即保证信号噪声的去除，又保留了细节信号。The specific parameter setting of the high-frequency noise removal unit depends on the sampling rate of the original PPG signal. In practical applications, it requires repeated observation and debugging to ensure that the signal noise is removed while retaining the detail signal.
基线漂移去除单元，其用于利用经验模态分解方法(EEMD)对所述平滑滤波信号进行处理，得到多个本征模态函数，并将多个本征模态函数进行相加处理，得到重构信号；具体的，E＝EEMD(Ti)＝imf[0]+imf[1]+imf[2]+…+imf[m],其中，E代表重构信号，Ti代表平滑滤波信号T中的第i个平滑滤波信号值，m代表本征模态函数的总个数；EEMD的主要优点是它的基函数是根据信号本身自适应提取的，按照信号固有时间尺度相关联的能量的顺序，从高频时间尺度到低频的时间尺度；分解后的信号成分称为本征模式函数(IntrinsicMode Function，IMF)，原始信最终会被分解为各阶IMF与残差信号之和。EEMD有以下几点优势：(1)通过自适应的方法来分解信号，分解后得到的IMF分量具有明确的物理含义，可以更好地保留信号的局部信息；(2)可以通过信号分量重构，滤除以高频噪声信号和运动伪影为代表的IMF分量，实现对非平稳信号的平滑处理，从而缓解由PPG运动伪影引起的PPG导数信号的形态畸形问题；IMF分量需要满足以下两个条件：(1)在整个时间范围内，函数的局部极值点和零点的个数必须相同或最多相差一个；(2)在任意时间点，函数局部最大值的包络(上包络)和局部最小值的包络(下包络)的平均值必须为零。EEMD的处理流程如图5所示。A baseline drift removal unit is used to process the smoothing filter signal using an empirical mode decomposition method (EEMD) to obtain multiple intrinsic mode functions, and add the multiple intrinsic mode functions to obtain a reconstructed signal; specifically, E=EEMD(Ti)=imf[0]+imf[1]+imf[2]+…+imf[m], wherein E represents the reconstructed signal, Ti represents the i-th smoothing filter signal value in the smoothing filter signal T, and m represents the total number of intrinsic mode functions; the main advantage of EEMD is that its basis function is adaptively extracted based on the signal itself, in the order of energy associated with the inherent time scale of the signal, from high-frequency time scale to low-frequency time scale; the decomposed signal component is called an intrinsic mode function (IMF), and the original signal will eventually be decomposed into the sum of each order IMF and the residual signal. EEMD has the following advantages: (1) The signal is decomposed by an adaptive method, and the IMF components obtained after decomposition have clear physical meanings and can better retain the local information of the signal; (2) The IMF components represented by high-frequency noise signals and motion artifacts can be filtered out through signal component reconstruction to achieve smoothing of non-stationary signals, thereby alleviating the morphological deformity of the PPG derivative signal caused by PPG motion artifacts; the IMF components need to meet the following two conditions: (1) In the entire time range, the number of local extreme points and zero points of the function must be the same or differ by at most one; (2) At any time point, the average value of the envelope of the local maximum value (upper envelope) and the envelope of the local minimum value (lower envelope) of the function must be zero. The processing flow of EEMD is shown in Figure 5.
信号提取单元，其用于从所述重构信号中提取出感兴趣的本征模态函数并相加，得到一波长优化PPG信号；具体的，在一些实施例中，感兴趣的本征模态函数为第2个本征模态函数imf[1]和第3个本征模态函数imf[2]，则P＝imf[1]+imf[2],其中，P代表一波长优化PPG信号。由于经验模态分解方法的原理是把信号按照不同的频率分开，因此在信号提取过程中，是将我们感兴趣的频率对应的本征模态函数相加得到优化PPG信号；此处选择P＝imf[1]+imf[2]是因为在所选数据的计算过程中，imf[1]+imf[2]所得到的信号效果更好，如果更换别的原始PPG数据进行计算，在感兴趣频率不同时，也存在P＝imf[1]+imf[2]+imf[3]这种情况，具体提取哪几个本征模态函数相加取决于原始PPG信号的主要兴趣频率。A signal extraction unit is used to extract the intrinsic mode function of interest from the reconstructed signal and add them to obtain a wavelength optimized PPG signal; specifically, in some embodiments, the intrinsic mode function of interest is the second intrinsic mode function imf[1] and the third intrinsic mode function imf[2], then P=imf[1]+imf[2], where P represents a wavelength optimized PPG signal. Since the principle of the empirical mode decomposition method is to separate the signal according to different frequencies, in the signal extraction process, the intrinsic mode functions corresponding to the frequencies of interest are added to obtain the optimized PPG signal; P=imf[1]+imf[2] is selected here because in the calculation process of the selected data, the signal obtained by imf[1]+imf[2] has a better effect. If other original PPG data are replaced for calculation, when the frequencies of interest are different, there is also a situation of P=imf[1]+imf[2]+imf[3]. Which intrinsic mode functions are extracted and added depends on the main frequencies of interest of the original PPG signal.
在一些实施例，通过以上降噪滤波处理后得到血流的优化PPG信号之后，需要从优化PPG信号中提取出我们所需要的各种信息。如图6所示，脑部可以大致分为四层，由上到下一次为毛发层、皮肤组织层、颅骨层、颅内组织层，两种不同波长的LED光源的传输深度是不同的，短波光的穿透能力差，长波的穿透能力强，如图6所示；基于此原理，同时使用两种波长测量时，短波无法测得颅内信号，而长波可以测得颅内信号。本发明可以通过以下五种方案来对采集到的优化PPG信号进行定量化的特征参数描述，便于区分不同波长采集到的信号的差异性。In some embodiments, after the optimized PPG signal of blood flow is obtained through the above noise reduction and filtering process, it is necessary to extract the various information we need from the optimized PPG signal. As shown in Figure 6, the brain can be roughly divided into four layers, from top to bottom, the hair layer, skin tissue layer, skull layer, and intracranial tissue layer. The transmission depths of two LED light sources with different wavelengths are different. The short-wave light has poor penetration ability, and the long-wave light has strong penetration ability, as shown in Figure 6; based on this principle, when two wavelengths are used for measurement at the same time, the short-wave cannot measure the intracranial signal, while the long-wave can measure the intracranial signal. The present invention can quantitatively describe the characteristic parameters of the collected optimized PPG signal through the following five schemes, which is convenient for distinguishing the differences of signals collected at different wavelengths.
第一种方案，脉图面积法：The first solution is the pulse area method:
所述控制器内设置有特征参数提取模块，所述特征参数提取模块用于：The controller is provided with a characteristic parameter extraction module, and the characteristic parameter extraction module is used to:
根据所述双波长优化PPG信号中的各波长优化PPG信号，提取出与各波长优化PPG信号的脉图面积相关的PPG波形特征量；According to each wavelength optimized PPG signal in the dual-wavelength optimized PPG signal, a PPG waveform feature quantity related to the pulse graph area of each wavelength optimized PPG signal is extracted;
将与各波长优化PPG信号的脉图面积相关的PPG波形特征量均作为所述特征参数，并根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号；The PPG waveform feature quantities related to the pulse graph area of each wavelength-optimized PPG signal are used as the feature parameters, and the intracranial and extracranial brain blood flow signals are distinguished from the dual-wavelength-optimized PPG signal according to the feature parameters;
其中，所述PPG波形特征量的提取公式为：Among them, the extraction formula of the PPG waveform feature is:
具体的，K表示与任一波长优化PPG信号的脉图面积相关的PPG波形特征量，Pm表示所述双波长优化PPG信号中的任一波长优化PPG信号的平均动脉压，Ps和Pd分别表示所述双波长优化PPG信号中的任一波长优化PPG信号的收缩压和舒张压。Specifically, K represents a PPG waveform feature related to the pulse graph area of any wavelength-optimized PPG signal,Pm represents the mean arterial pressure of any wavelength-optimized PPG signal in the dual-wavelength-optimized PPG signals, andPs andPd respectively represent the systolic pressure and diastolic pressure of any wavelength-optimized PPG signal in the dual-wavelength-optimized PPG signals.
具体的，在正常的生理状态下，血管壁的弹性可以使得血管在心脏的搏动血液的脉动作用下，产生周期性的扩张和收缩。在血液循环过程中，心室快速射血期时，动脉血压迅速上升，引起血管壁扩张，形成脉搏波形中的上升支；上升支的斜率和幅度受心输出量、射血速度以及射血所遇的动脉阻力的影响，如果射血遇到的阻力大，导致心输出量小，射血速度慢，则脉搏波形中上升速度慢幅值小；反之，如果射血所遇的阻力小，导致心输出量大，射血速度快，则上升速度快幅值也较大。动脉血管壁的顺应性也影响着上升支波形的上升速度，如果血管的顺应性降低，会使血流速度增加，即上升支斜率增大。在心室射血的后期，射血速度减慢，进入主动脉的血量小于由主动脉流向外周的血量，之前扩张的动脉开始回缩，形成脉搏波形中下降支的前段。心室收缩期结束舒张期开始的标志是主动脉瓣关闭，在主动脉瓣关闭的瞬间，心室舒张使室内压下降，主动脉内的血液向心室方向返流，返流的血液受到闭合的主动脉瓣阻挡，会在下降支的前段后产生一个切迹，称为降中峡，在降中峡的后面形成一个短暂的向上的小波称为重搏波。随后，心室舒张，血液继续向前流动，动脉血压持续下降，形成下降支的其余部分，如图7所示和图8所示。Specifically, under normal physiological conditions, the elasticity of the blood vessel wall can cause the blood vessel to expand and contract periodically under the pulsation of the heart's pulsating blood. During the blood circulation process, during the rapid ventricular ejection period, the arterial blood pressure rises rapidly, causing the blood vessel wall to expand, forming the ascending branch in the pulse waveform; the slope and amplitude of the ascending branch are affected by the cardiac output, ejection velocity, and the arterial resistance encountered by the ejection. If the resistance encountered by the ejection is large, resulting in a small cardiac output and a slow ejection velocity, the pulse waveform has a slow ascending velocity and a small amplitude; conversely, if the resistance encountered by the ejection is small, resulting in a large cardiac output and a fast ejection velocity, the ascending velocity is fast and the amplitude is also large. The compliance of the arterial blood vessel wall also affects the rising velocity of the ascending branch waveform. If the compliance of the blood vessel decreases, the blood flow velocity will increase, that is, the slope of the ascending branch increases. In the late stage of ventricular ejection, the ejection velocity slows down, the amount of blood entering the aorta is less than the amount of blood flowing from the aorta to the periphery, and the previously expanded artery begins to shrink, forming the front section of the descending branch in the pulse waveform. The sign of the end of ventricular systole and the beginning of diastole is the closure of the aortic valve. At the moment of closure of the aortic valve, the ventricular diastole causes the pressure inside the chamber to drop, and the blood in the aorta flows back toward the ventricle. The refluxed blood is blocked by the closed aortic valve, and a notch is formed behind the front section of the descending branch, which is called the descending middle isthmus. A short upward small wave called the dicrotic wave is formed behind the descending middle isthmus. Subsequently, the ventricle relaxes, the blood continues to flow forward, and the arterial blood pressure continues to drop, forming the rest of the descending branch, as shown in Figures 7 and 8.
由上面所记载的PPG的产生与传播的机理可知，随着血管阻力和动脉弹性等生理变化，PPG波形特征变化首先反映在PPG波形面积的变化上，这里用了一个以脉图面积变化为基础的PPG波形特征量K，图9为一个脉搏波周期的示意图，由此可见，K值的大小仅决定于PPG的脉图面积，与收缩压和舒张压的绝对值无关，是一个无量纲值，它相当于PPG压力脉动分量的平均值在脉动分量最大值中所占的百分比。不同生理病理状态下脉图波形和面积都会有很大变化，这个变化可以用K值表示。当外周阻力较低，血管壁弹性较好时，PPG波形特征是主波窄而高，潮波不明显，重搏波峰和波谷都很突出，这时脉图面积很小，K值在0.35左右。随着外周阻力和血管壁硬化程度增加，波形的动态变化首先反映在潮波由不明显变为明显，它相对于主波的位置也逐渐升高，并自后向前与主波接近和呈不同程度的融合，甚至超过主波。与此同时，重搏波波峰与波谷对主波的位置亦逐渐抬高，且混为一体不易区分，使整个PPG波形呈馒头型，这时脉图的面积也逐渐增加，K值也将相应有规律地增加到0.5左右。From the mechanism of PPG generation and propagation recorded above, it can be seen that with physiological changes such as vascular resistance and arterial elasticity, the change of PPG waveform characteristics is first reflected in the change of PPG waveform area. Here, a PPG waveform characteristic quantity K based on the change of pulse area is used. Figure 9 is a schematic diagram of a pulse wave cycle. It can be seen that the size of the K value is only determined by the pulse area of the PPG, and has nothing to do with the absolute values of systolic and diastolic pressures. It is a dimensionless value, which is equivalent to the percentage of the average value of the PPG pressure pulsation component in the maximum value of the pulsation component. The pulse waveform and area will change greatly under different physiological and pathological conditions, and this change can be expressed by the K value. When the peripheral resistance is low and the elasticity of the vascular wall is good, the PPG waveform characteristics are that the main wave is narrow and high, the tidal wave is not obvious, and the dicrotic peak and trough are very prominent. At this time, the pulse area is very small, and the K value is about 0.35. As the peripheral resistance and vascular wall sclerosis increase, the dynamic changes in the waveform are first reflected in the tidal wave changing from inconspicuous to obvious, and its position relative to the main wave gradually increases, and approaches and merges with the main wave from back to front to varying degrees, and even exceeds the main wave. At the same time, the position of the peak and trough of the dicrotic wave relative to the main wave is gradually raised, and it is difficult to distinguish them as one, making the entire PPG waveform steamed bun-shaped. At this time, the area of the pulse graph gradually increases, and the K value will also increase regularly to about 0.5 accordingly.
由于优化PPG信号中只有颅外信号或同时存在颅外和颅内信号这两种情况，K值存在一定的差异性，因此K值可以作为判定监测到的优化PPG信号是否为颅内信号的特征参数。Since there are two situations in which the optimized PPG signal contains only extracranial signals or both extracranial and intracranial signals, there are certain differences in the K value. Therefore, the K value can be used as a characteristic parameter to determine whether the monitored optimized PPG signal is an intracranial signal.
第二种方案，特征点法：The second solution, feature point method:
所述控制器内设置有特征参数提取模块，所述特征参数提取模块用于：The controller is provided with a characteristic parameter extraction module, and the characteristic parameter extraction module is used to:
根据所述双波长优化PPG信号中的各波长优化PPG信号，绘制出各波长优化PPG信号的PPG波图；According to each wavelength optimized PPG signal in the dual-wavelength optimized PPG signal, a PPG wave graph of each wavelength optimized PPG signal is drawn;
从各波长优化PPG信号的PPG波图中提取出曲线拐点或/和曲线斜率；Extracting a curve inflection point and/or a curve slope from the PPG wave graph of each wavelength-optimized PPG signal;
将从各波长优化PPG信号的PPG波图中提取出的曲线拐点或/和曲线斜率均作为所述特征参数，并根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号。The curve inflection point and/or the curve slope extracted from the PPG wave graph of each wavelength optimized PPG signal are used as the characteristic parameters, and the intracranial and extracranial brain blood flow signals are distinguished from the dual-wavelength optimized PPG signal based on the characteristic parameters.
具体的，PPG波图的特征点实质上就是PPG压力曲线的拐点，是心动周期从一个力学过程转变为另一个力学过程的转变点，每个点都有明确的生理意义。一个完整的PPG应包含A、B、C、D四个主要的特征点，如图10所示，A为主波，B为潮波，C为重搏波峰，D为重搏波谷(切迹)。特征点法的特点在于由获得的PPG信号曲线中选出这些拐点或斜率作为辨别脉图的参数；这些不同的参数对应了不同情况下的血流信号，可以将这些参数作为判定监测到的优化PPG信号是否为颅内血流信号的特征参数。Specifically, the characteristic point of the PPG wave graph is essentially the inflection point of the PPG pressure curve, which is the transition point where the cardiac cycle changes from one mechanical process to another, and each point has a clear physiological meaning. A complete PPG should contain four main characteristic points, A, B, C, and D, as shown in Figure 10, A is the main wave, B is the tidal wave, C is the dicrotic wave peak, and D is the dicrotic wave trough (notch). The characteristic of the characteristic point method is that these inflection points or slopes are selected from the obtained PPG signal curve as parameters for distinguishing the pulse graph; these different parameters correspond to blood flow signals under different circumstances, and these parameters can be used as characteristic parameters to determine whether the monitored optimized PPG signal is an intracranial blood flow signal.
第三种方案，高斯函数法：The third option, Gaussian function method:
所述控制器内设置有特征参数提取模块，所述特征参数提取模块用于：The controller is provided with a characteristic parameter extraction module, and the characteristic parameter extraction module is used to:
将所述双波长优化PPG信号中的各波长优化PPG信号均分解成分别与三个高斯函数一一对应的三个钟形波；Decomposing each wavelength-optimized PPG signal in the dual-wavelength-optimized PPG signal into three bell-shaped waves corresponding to three Gaussian functions respectively;
确定出各波长优化PPG信号中的三个钟形波的幅度、时间和宽度；Determine the amplitude, time and width of the three bell-shaped waves in each wavelength-optimized PPG signal;
将各波长优化PPG信号中的三个钟形波的幅度、时间和宽度均作为所述特征参数，并根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号；The amplitude, time and width of the three bell-shaped waves in each wavelength-optimized PPG signal are used as the characteristic parameters, and the intracranial and extracranial brain blood flow signals are distinguished from the dual-wavelength-optimized PPG signal according to the characteristic parameters;
具体的，一个周期的PPG大致可以用三个高斯函数来合成(即将一个PPG信号波形拟合成多个高斯函数相加)，如图11所示，分别称为这三个高斯函数为钟形主波、钟形重搏波和钟形重播前波。根据高斯函数的定义，每个钟形波需要由三个参数确定，即幅度V、时间T和宽度U。这三个参数可以作为三个特征值，对不同情况下的血流信号的特征值不同，可以通过这三个参数来判断监测到的优化PPG信号是否为颅内血流信号。Specifically, a PPG cycle can be roughly synthesized by three Gaussian functions (i.e., fitting a PPG signal waveform into multiple Gaussian functions and adding them together), as shown in FIG11 , and these three Gaussian functions are respectively called bell-shaped main wave, bell-shaped dicrotic wave, and bell-shaped replay pre-wave. According to the definition of the Gaussian function, each bell-shaped wave needs to be determined by three parameters, namely, amplitude V, time T, and width U. These three parameters can be used as three characteristic values. The characteristic values of blood flow signals in different situations are different. These three parameters can be used to determine whether the monitored optimized PPG signal is an intracranial blood flow signal.
第四种方案，在频域进行高斯拟合：The fourth solution is to perform Gaussian fitting in the frequency domain:
所述特征参数提取模块用于：The feature parameter extraction module is used for:
对所述双波长优化PPG信号中的各波长优化PPG信号进行傅里叶变换，得到各波长优化PPG信号的频域信息；Performing Fourier transform on each wavelength optimized PPG signal in the dual-wavelength optimized PPG signal to obtain frequency domain information of each wavelength optimized PPG signal;
获取各波长优化PPG信号的频域信息中的特征峰值，并进行高斯拟合，得到各波长优化PPG信号的高斯拟合函数；Obtaining characteristic peaks in the frequency domain information of each wavelength optimized PPG signal, and performing Gaussian fitting to obtain the Gaussian fitting function of each wavelength optimized PPG signal;
获取各波长优化PPG信号的高斯拟合函数的半高全宽参数；Obtaining the full width at half maximum parameters of the Gaussian fitting function of the optimized PPG signal at each wavelength;
将各波长优化PPG信号的高斯拟合函数的半高全宽参数均作为所述特征参数，并根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号。The half-width parameters of the Gaussian fitting function of each wavelength-optimized PPG signal are used as the characteristic parameters, and the intracranial and extracranial brain blood flow signals are distinguished from the dual-wavelength optimized PPG signal according to the characteristic parameters.
具体的，在频域方法研究PPG信号，就是把时域的PPG曲线经傅里叶变换到频域，然后从中提取特征峰值。将PPG信号转化到频域后，通常会出现几个特征峰值，在将这几个峰值点连线，进行高斯拟合，会得到一个新的高斯函数(即高斯拟合函数)，这个高斯拟合函数的各项信息，可为该PPG信号的特征点。图12为对一个PPG信号的处理结果，可以看出对不同的PPG信号，拟合所得到的高斯函数各不相同。将高斯函数的半高全宽作为一个特征值，不同情况下的血流信号该特征值不同，可以用来区分监测到的优化PPG信号是否为颅内血流信号。Specifically, the frequency domain method for studying the PPG signal is to transform the PPG curve in the time domain into the frequency domain through Fourier transform, and then extract the characteristic peaks from it. After the PPG signal is converted to the frequency domain, there are usually several characteristic peaks. By connecting these peak points and performing Gaussian fitting, a new Gaussian function (i.e., Gaussian fitting function) will be obtained. The various information of this Gaussian fitting function can be the characteristic point of the PPG signal. Figure 12 shows the processing result of a PPG signal. It can be seen that the Gaussian functions obtained by fitting are different for different PPG signals. The full width at half maximum of the Gaussian function is taken as a characteristic value. The characteristic value is different for blood flow signals under different conditions, which can be used to distinguish whether the monitored optimized PPG signal is an intracranial blood flow signal.
第五种方案，对独立的心动周期进行傅里叶拟合：The fifth option is to perform Fourier fitting on independent cardiac cycles:
所述控制器内设置有特征参数提取模块，所述特征参数提取模块用于：The controller is provided with a characteristic parameter extraction module, and the characteristic parameter extraction module is used to:
将所述双波长优化PPG信号中的各波长优化PPG信号划分为多个独立的心动周期；Dividing each wavelength-optimized PPG signal in the dual-wavelength-optimized PPG signal into a plurality of independent cardiac cycles;
分别对各波长优化PPG信号中的每一个心动周期进行傅里叶拟合，得到各波长优化PPG信号的多个傅里叶拟合函数；Performing Fourier fitting on each cardiac cycle in each wavelength-optimized PPG signal to obtain a plurality of Fourier fitting functions of each wavelength-optimized PPG signal;
获取各波长优化PPG信号的多个傅里叶拟合函数的系数，并进行概率分布统计，得到各波长优化PPG信号的多个傅里叶拟合函数的系数的概率分布情况；Obtaining coefficients of multiple Fourier fitting functions of each wavelength optimized PPG signal, and performing probability distribution statistics to obtain probability distribution of coefficients of multiple Fourier fitting functions of each wavelength optimized PPG signal;
将各波长优化PPG信号的多个傅里叶拟合函数的系数的概率分布情况均作为所述特征参数，并根据所述特征参数从所述双波长优化PPG信号中区分出颅内和颅外的脑部血流信号。The probability distribution of the coefficients of the multiple Fourier fitting functions of each wavelength-optimized PPG signal is used as the characteristic parameter, and the intracranial and extracranial brain blood flow signals are distinguished from the dual-wavelength optimized PPG signal according to the characteristic parameter.
具体的，如图13所示，将一段时间内的PPG信号分开为独立的心动周期，再对每一个心动周期进行傅里叶拟合，对于每一个周期来说，都能得到一个拟合结果。对于傅里叶拟合来说，其表达式为：f(x)＝a0+a1 cos(xω)+b1 sin(xω)+a2 cos(2xω)+b2 sin(2xω)+a3cos(3xω)+b3 sin(3xω)+a4 cos(4xω)+b4 sin(4xω)+a5 cos(5xω)+b5 sin(5xω)+a6cos(6xω)+b6 sin(6xω)+...Specifically, as shown in FIG13 , the PPG signal within a period of time is separated into independent cardiac cycles, and then Fourier fitting is performed on each cardiac cycle. For each cycle, a fitting result can be obtained. For Fourier fitting, the expression is: f(x)＝a0 +a1 cos(xω)+b1 sin(xω)+a2 cos(2xω)+b2 sin(2xω)+a3 cos(3xω)+b3 sin(3xω)+a4 cos(4xω)+b4 sin(4xω)+a5 cos(5xω)+b5 sin(5xω)+a6 cos(6xω)+b6 sin(6xω)+...
其中，a0是直流分量、a1、b1、a2、b2…是傅里叶系数，表示在频率nf处的振幅。这里f是信号的基本频率，而n是正整数。这些系数可以作为特征参数来判断监测到的优化PPG信号是否为颅内血流信号。Among them,a0 is the DC component,a1 ,b1 ,a2 ,b2 ... are Fourier coefficients, representing the amplitude at the frequency nf. Here f is the fundamental frequency of the signal, and n is a positive integer. These coefficients can be used as characteristic parameters to determine whether the monitored optimized PPG signal is an intracranial blood flow signal.
例如，对大鼠做如下处理，得到以下三种不同情况下的大鼠。第一种情况下的大鼠：对大鼠的一侧颈总动脉做栓塞处理，使一侧颅内缺血，该侧只有颅外信号，无颅内信号；第二种情况下的大鼠：对大鼠另一侧的头皮进行剥离，使该侧只有颅内信号，无颅外信号；第三种情况下的大鼠：不做处理。对这三种情况下的大鼠所采集到的血流信号按照上述独立的心动周期进行傅里叶拟合方式进行处理，得到了一系列傅里叶系数，其概率分布如图14所示；其中，横坐标代表傅里叶系数a(包括a1、a2…)，纵坐标代表傅里叶系数b(包括b1、b2…)，+表示第三种工况下的大鼠同时存在颅内和颅外信号时的傅里叶系数分布情况，o表示第一种工况下的大鼠只存在颅外信号时的傅里叶系数分布情况，*表示第二种工况的大鼠只存在颅内信号时的傅里叶系数分布情况。从图14可以看出，在三种工况下，各项傅里叶系数有非常明显的概率分布差异，傅里叶系数的集中位置不同，所以可以通过傅里叶系数分布情况来区分不同波长监测到的信号存在的差异性。For example, rats are processed as follows to obtain rats in the following three different conditions. Rats in the first condition: one side of the common carotid artery of the rat is embolized to cause intracranial ischemia on one side, and only extracranial signals are present on this side, but no intracranial signals are present; rats in the second condition: the scalp of the other side of the rat is peeled off, so that only intracranial signals are present on this side, but no extracranial signals are present; rats in the third condition: no processing is performed. The blood flow signals collected from the rats in these three conditions are processed by Fourier fitting according to the above-mentioned independent cardiac cycles, and a series of Fourier coefficients are obtained, and their probability distribution is shown in FIG14 ; wherein the abscissa represents the Fourier coefficient a (including a1 , a2 , etc.), the ordinate represents the Fourier coefficient b (including b1 , b2 , etc.), + represents the Fourier coefficient distribution of the rats in the third condition when both intracranial and extracranial signals exist, o represents the Fourier coefficient distribution of the rats in the first condition when only extracranial signals exist, and * represents the Fourier coefficient distribution of the rats in the second condition when only intracranial signals exist. As can be seen from Figure 14, under the three working conditions, the probability distribution of each Fourier coefficient is very different, and the concentration positions of the Fourier coefficients are different, so the differences in signals monitored at different wavelengths can be distinguished by the distribution of the Fourier coefficients.
本发明提供的一种便携式脑部血流信号监测系统可以在运动状态下对颅内血流情况进行实时监测，通过光学的方式进行监测，完全无创，减轻了患者的痛苦；同时也解决了现有的便携脑部血流情况监测装置对颅内颅外血流情况无法区分的问题；相较于颅脑CT、颅脑核磁共振成像等方法来说成本低，操作更简便，适合在急救情况下的快速诊断；可以在心肺复苏等运动状态下进行监测，能够消除由于运动对信号造成的干扰和影响；可以在极低脑血流(≤30％)情况下进行有效测量。The portable brain blood flow signal monitoring system provided by the present invention can monitor the intracranial blood flow in real time under motion state, and monitors by optical means, which is completely non-invasive and reduces the pain of patients; at the same time, it also solves the problem that the existing portable brain blood flow monitoring devices cannot distinguish between intracranial and extracranial blood flow conditions; compared with methods such as cranial CT and cranial magnetic resonance imaging, it has low cost and simpler operation, and is suitable for rapid diagnosis in emergency situations; it can monitor under motion states such as cardiopulmonary resuscitation, and can eliminate interference and influence on the signal caused by motion; it can effectively measure under extremely low cerebral blood flow conditions (≤30%).
以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410296593.4ACN118177771B (en) | 2024-03-15 | 2024-03-15 | A portable brain blood flow signal monitoring system |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410296593.4ACN118177771B (en) | 2024-03-15 | 2024-03-15 | A portable brain blood flow signal monitoring system |
| Publication Number | Publication Date |
|---|---|
| CN118177771Atrue CN118177771A (en) | 2024-06-14 |
| CN118177771B CN118177771B (en) | 2024-11-15 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410296593.4AActiveCN118177771B (en) | 2024-03-15 | 2024-03-15 | A portable brain blood flow signal monitoring system |
| Country | Link |
|---|---|
| CN (1) | CN118177771B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090167203A1 (en)* | 2007-12-28 | 2009-07-02 | Mark Cobb Dahlman | AC-powered, microprocessor-based, dimming LED power supply |
| CN105266780A (en)* | 2015-11-24 | 2016-01-27 | 京东方科技集团股份有限公司 | Intelligent wearable device and detection method for biological characteristic information |
| CN105486921A (en)* | 2016-01-22 | 2016-04-13 | 武汉大学 | Kaiser third-order mutual convolution window triple-spectrum-line interpolation harmonic wave and inter-harmonic wave detection method |
| CN210609808U (en)* | 2019-06-06 | 2020-05-22 | 珠海市保时源光电科技有限公司 | Stroboscopic removing circuit of LED driving circuit |
| CN112043287A (en)* | 2020-09-30 | 2020-12-08 | 重庆大学 | Cerebral blood oxygen non-invasive monitoring method and monitoring device |
| CN113288092A (en)* | 2021-05-06 | 2021-08-24 | 福建师范大学 | Blood pressure detection method and terminal for extracting pulse wave based on video |
| CN117064356A (en)* | 2023-09-27 | 2023-11-17 | 南方科技大学 | Non-contact blood pressure monitoring device, method and system utilizing spatial redundancy to enhance robustness |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090167203A1 (en)* | 2007-12-28 | 2009-07-02 | Mark Cobb Dahlman | AC-powered, microprocessor-based, dimming LED power supply |
| CN105266780A (en)* | 2015-11-24 | 2016-01-27 | 京东方科技集团股份有限公司 | Intelligent wearable device and detection method for biological characteristic information |
| CN105486921A (en)* | 2016-01-22 | 2016-04-13 | 武汉大学 | Kaiser third-order mutual convolution window triple-spectrum-line interpolation harmonic wave and inter-harmonic wave detection method |
| CN210609808U (en)* | 2019-06-06 | 2020-05-22 | 珠海市保时源光电科技有限公司 | Stroboscopic removing circuit of LED driving circuit |
| CN112043287A (en)* | 2020-09-30 | 2020-12-08 | 重庆大学 | Cerebral blood oxygen non-invasive monitoring method and monitoring device |
| CN113288092A (en)* | 2021-05-06 | 2021-08-24 | 福建师范大学 | Blood pressure detection method and terminal for extracting pulse wave based on video |
| CN117064356A (en)* | 2023-09-27 | 2023-11-17 | 南方科技大学 | Non-contact blood pressure monitoring device, method and system utilizing spatial redundancy to enhance robustness |
| Title |
|---|
| 孙笑敩: "基于深度学习的多光谱容积脉搏波的血压测量技术研究", 《中国博士学位论文全文数据库 医药卫生科技辑》, 15 February 2024 (2024-02-15), pages 62 - 70* |
| 张鑫宇: "基于rPPG的在线学习者心率估计及其疲劳检测应用", 《中国优秀硕士学位论文全文数据库 社会科学Ⅱ辑》, 15 February 2023 (2023-02-15), pages 35 - 36* |
| 李强: "基于脉搏特征的智能生理信息提取与监测技术研究", 《中国优秀硕士学位论文全文数据库信息科技辑》, 15 February 2014 (2014-02-15), pages 25 - 45* |
| 李红云;魏嵘;: "局部脑氧饱和度监测在临床中的应用进展", 医学综述, no. 03, 5 February 2018 (2018-02-05)* |
| Publication number | Publication date |
|---|---|
| CN118177771B (en) | 2024-11-15 |
| Publication | Publication Date | Title |
|---|---|---|
| CN109157202B (en) | Cardiovascular disease early warning system based on multi-physiological signal deep fusion | |
| CN110537910B (en) | Coronary heart disease noninvasive screening system based on electrocardio and heart sound signal joint analysis | |
| CN110420019B (en) | A deep regression heart rate estimation method for shock cardiogram signals | |
| Nabavi et al. | A robust fusion method for motion artifacts reduction in photoplethysmography signal | |
| US7862515B2 (en) | Apparatus for detecting sleep apnea using electrocardiogram signals | |
| JPH07503380A (en) | Impedance cardiography method and heart rate determination method | |
| CN109561841A (en) | Non-invasive method and system for measuring myocardial ischemia, stenosis identification, localization, and fractional flow reserve estimation | |
| AU2011274478B2 (en) | System and method for characterizing circulatory blood flow | |
| CN105678780A (en) | Video heart rate detection method removing interference of ambient light variation | |
| Li et al. | Comparison of respiratory-induced variations in photoplethysmographic signals | |
| CN113288103B (en) | Noninvasive heart displacement monitoring system and method | |
| CN113499059A (en) | BCG signal processing system and method based on optical fiber sensing non-contact | |
| US20230055617A1 (en) | Methods and Systems for Engineering Photoplethysmographic-Waveform Features From Biophysical Signals for Use in Characterizing Physiological Systems | |
| CN118415649B (en) | A closed-loop multimodal neural stimulation system and method based on time interference | |
| CN115422976A (en) | An artificial network-based cardiopulmonary coupling relationship analysis method and monitoring system | |
| CN113288101A (en) | ICG signal processing method in motion state based on spectral subtraction and frequency domain EEMD-CCA | |
| CN115381424A (en) | Heart rate detection method based on double-Gaussian function model and linear regression iterative algorithm | |
| CN118177771B (en) | A portable brain blood flow signal monitoring system | |
| CN116509357B (en) | Continuous blood pressure estimation method based on multi-scale convolution | |
| RU182802U1 (en) | PULSE WAVE CONTOUR ANALYSIS DEVICE | |
| WO2024211033A2 (en) | Methods and devices for the non-invasive assessment of cerebro-spinal dynamics and diagnosis of brain pathophysiologies | |
| Abtahi | Feasibility of fetal EEG Recording | |
| Noorzadeh | Extraction of fetal ECG and its characteristics using multi-modality | |
| CN116982950A (en) | Cuff-free blood pressure monitoring system | |
| Chaddad et al. | Denoising fNIRS signals to enhance brain imaging diagnosis |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |