技术领域Technical Field
本发明涉及信号处理领域，尤其涉及的是一种基于信号质量评估的脉搏波预处理方法。The invention relates to the field of signal processing, and in particular to a pulse wave preprocessing method based on signal quality evaluation.
背景技术Background technique
计算机辅助脉诊需要借助脉象传感器获取脉搏信号，然而脉象传感器获取的脉搏信号是一种微弱的生理信号，并且经常受到外部环境、人体和电路系统的干扰。当这些干扰导致采集到的脉搏波信号产生形态畸变，形态畸变严重的脉搏波信号甚至完全不能用于后续确定脉诊结果。然而目前在对脉象传感器采集到的脉搏波信号进行处理时，默认脉象传感器采集到的脉搏波信号均为可用脉搏波信号，从而导致最后脉诊结果不准确。Computer-aided pulse diagnosis requires the use of a pulse sensor to obtain a pulse signal. However, the pulse signal obtained by the pulse sensor is a weak physiological signal and is often interfered by the external environment, the human body, and the circuit system. When these interferences cause the collected pulse wave signal to produce morphological distortion, the pulse wave signal with severe morphological distortion cannot even be used to determine the pulse diagnosis result in the future. However, when processing the pulse wave signal collected by the pulse sensor, it is assumed that the pulse wave signal collected by the pulse sensor is a usable pulse wave signal, which leads to inaccurate pulse diagnosis results.
因此，现有技术还有待改进和发展。Therefore, the existing technology still needs to be improved and developed.
发明内容Summary of the invention
本发明要解决的技术问题在于，针对现有技术的上述缺陷，提供一种基于信号质量评估的脉搏波预处理方法，旨在解决现有技术中将脉象传感器采集到的脉搏波信号均作为可用脉搏波信号用于生成脉诊结果，忽略了采集到的脉搏波信号中存在一些形态畸形的脉搏波信号，从而导致脉诊结果不准确。The technical problem to be solved by the present invention is that, in view of the above-mentioned defects of the prior art, a pulse wave preprocessing method based on signal quality evaluation is provided, aiming to solve the problem in the prior art that all pulse wave signals collected by the pulse sensor are used as available pulse wave signals to generate pulse diagnosis results, ignoring the presence of some morphologically deformed pulse wave signals in the collected pulse wave signals, thereby causing inaccurate pulse diagnosis results.
本发明解决问题所采用的技术方案如下：The technical solution adopted by the present invention to solve the problem is as follows:
第一方面，本发明实施例提供一种基于信号质量评估的脉搏波预处理方法，其中，所述方法包括：In a first aspect, an embodiment of the present invention provides a pulse wave preprocessing method based on signal quality assessment, wherein the method comprises:
获取原始脉搏波信号的波形特征信息，根据所述波形特征信息调节脉象传感器的位置信息，并根据调节后的所述脉象传感器的位置信息，获取第一脉搏波信号；所述波形特征信息用于反映所述脉搏波信号中全部周期的幅值变化规律；Acquire waveform characteristic information of the original pulse wave signal, adjust the position information of the pulse sensor according to the waveform characteristic information, and acquire a first pulse wave signal according to the adjusted position information of the pulse sensor; the waveform characteristic information is used to reflect the amplitude variation law of all cycles in the pulse wave signal;
获取所述第一脉搏波信号的幅度特征信息，根据所述幅度特征信息调节所述脉象传感器的参数信息，并根据调节后的所述脉象传感器的参数信息，获取第二脉搏波信号；所述幅度特征信息用于反映所述第一脉搏波信号在时域上的幅值的强度信息；Acquire amplitude characteristic information of the first pulse wave signal, adjust parameter information of the pulse sensor according to the amplitude characteristic information, and acquire a second pulse wave signal according to the adjusted parameter information of the pulse sensor; the amplitude characteristic information is used to reflect the intensity information of the amplitude of the first pulse wave signal in the time domain;
获取所述第二脉搏波信号的周期差异信息，根据所述周期差异信息对所述第二脉搏波信号进行处理；所述周期差异信息用于反映所述第二脉搏波信号中单个周期之间在幅值变化上的差异信息。Obtaining period difference information of the second pulse wave signal, and processing the second pulse wave signal according to the period difference information; the period difference information is used to reflect the difference information in amplitude change between single periods in the second pulse wave signal.
本发明的有益效果：本发明实施例可以根据采集到的脉搏波信号的波形特征信息、幅度特征信息以及周期差异信息及时调整脉象传感器的摆放位置或者参数，从而减少了获取到无效的脉搏波信号的概率。解决了现有技术中将脉象传感器采集到的脉搏波信号均作为可用脉搏波信号用于生成脉诊结果，忽略了采集到的脉搏波信号中存在一些形态畸形的脉搏波信号，从而导致脉诊结果不准确。Beneficial effects of the present invention: The embodiment of the present invention can timely adjust the placement position or parameters of the pulse sensor according to the waveform characteristic information, amplitude characteristic information and period difference information of the collected pulse wave signal, thereby reducing the probability of obtaining invalid pulse wave signals. It solves the problem in the prior art that all pulse wave signals collected by the pulse sensor are used as available pulse wave signals to generate pulse diagnosis results, ignoring some morphologically deformed pulse wave signals in the collected pulse wave signals, thereby causing inaccurate pulse diagnosis results.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明中记载的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
图1是本发明实施例提供的一种基于信号质量评估的脉搏波预处理方法的流程示意图。FIG1 is a flow chart of a pulse wave preprocessing method based on signal quality assessment provided by an embodiment of the present invention.
图2是本发明实施例提供的采集得到目标脉搏波信号的流程示意图。FIG. 2 is a schematic diagram of a process of acquiring a target pulse wave signal according to an embodiment of the present invention.
图3是本发明实施例提供的进行周期分割的流程示意图。FIG. 3 is a schematic diagram of a process of performing period segmentation according to an embodiment of the present invention.
图4是本发明实施例提供的对单周期数据进行归一化处理的算法参考图。FIG. 4 is a reference diagram of an algorithm for normalizing single-cycle data provided by an embodiment of the present invention.
图5是本发明实施例提供的对第二周期信号进行边采集边检测的流程示意图。FIG. 5 is a schematic diagram of a flow chart of collecting and detecting a second periodic signal according to an embodiment of the present invention.
图6是本发明实施例提供的五种类型的脉搏波信号分别对应的时域描述图像。FIG. 6 is a time domain description image corresponding to five types of pulse wave signals provided by an embodiment of the present invention.
图7是本发明实施例提供的饱和信号对应的时域描述图像。FIG. 7 is a time domain description image corresponding to a saturation signal provided by an embodiment of the present invention.
图8是本发明实施例提供的全局波形异常信号对应的时域描述图像和频域描述图像。FIG. 8 is a time domain description image and a frequency domain description image corresponding to a global waveform abnormal signal provided by an embodiment of the present invention.
图9是本发明实施例提供的正常脉搏波信号对应的时域描述图像和频域描述图像。FIG. 9 is a time domain description image and a frequency domain description image corresponding to a normal pulse wave signal provided by an embodiment of the present invention.
图10是本发明实施例提供的局部波形异常信号和正常脉搏波信号分别对应的时域描述图像。FIG. 10 is a time domain description image corresponding to a local waveform abnormal signal and a normal pulse wave signal provided by an embodiment of the present invention.
图11是本发明实施例提供部波形异常信号和正常脉搏波信号分别对应的展示单周期集合的聚集度的图像。FIG. 11 is an image showing the concentration of a single-cycle set corresponding to an abnormal waveform signal and a normal pulse wave signal, respectively, provided by an embodiment of the present invention.
具体实施方式Detailed ways
为使本发明的目的、技术方案及优点更加清楚、明确，以下参照附图并举实施例对本发明进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。In order to make the purpose, technical solution and advantages of the present invention clearer and more specific, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
需要说明，若本发明实施例中有涉及方向性指示(诸如上、下、左、右、前、后……)，则该方向性指示仅用于解释在某一特定姿态(如附图所示)下各部件之间的相对位置关系、运动情况等，如果该特定姿态发生改变时，则该方向性指示也相应地随之改变。It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.
随着图形处理单元(GPU)硬件计算力和深度学习算法的发展，人工智能(Artificial Intelligence，AI)在图像数据、语音等方面的识别准确度越来越高。可穿戴智能设备和医疗设备获得的人体健康信息也可被深度学习算法处理，用以辅助诊断和预防疾病，比如脉搏波、心电图、组织切片图等。其中，脉诊是中国传统中医的重要组成部分，计算机辅助脉诊有利于解决传统中医脉诊的客观性和标准化问题。With the development of graphics processing unit (GPU) hardware computing power and deep learning algorithms, artificial intelligence (AI) has become more and more accurate in recognizing image data, voice, etc. Human health information obtained by wearable smart devices and medical devices can also be processed by deep learning algorithms to assist in the diagnosis and prevention of diseases, such as pulse waves, electrocardiograms, tissue slices, etc. Among them, pulse diagnosis is an important part of traditional Chinese medicine, and computer-assisted pulse diagnosis is conducive to solving the objectivity and standardization problems of traditional Chinese medicine pulse diagnosis.
计算机辅助脉诊需要借助脉象传感器获取脉搏信号，然而脉象传感器获取的脉搏信号是一种微弱的生理信号，并且经常受到外部环境、人体和电路系统的影响，导致包含多种噪声源产生的低频和高频噪声。其中，干扰因素主要有以下三点：Computer-aided pulse diagnosis requires the use of a pulse sensor to obtain a pulse signal. However, the pulse signal obtained by the pulse sensor is a weak physiological signal and is often affected by the external environment, the human body, and the circuit system, resulting in low-frequency and high-frequency noise generated by a variety of noise sources. Among them, the interference factors are mainly the following three points:
1)由50HZ交流供电电源引起的高频干扰噪声；1) High-frequency interference noise caused by 50HZ AC power supply;
2)由于身体抖动(如咳嗽、肢体运动)、肌肉颤抖、说话等原因引起的不规则噪声；2) Irregular noise caused by body shaking (such as coughing, limb movement), muscle tremors, talking, etc.;
3)人体呼吸产生的肺部活动会在皮肤表面产生低频呼吸波，被压力脉象传感器采集到以后，会在原始信号中形成基线漂移。3) The lung activity generated by human breathing will produce low-frequency respiratory waves on the skin surface. After being collected by the pressure pulse sensor, it will form a baseline drift in the original signal.
这些干扰因素会产生一些影响后续分析精度的形态畸变，甚至导致脉象传感器采集到的脉搏波信号不可用。然而目前在对采集到的脉搏波信号进行处理时，默认信号均为可用脉搏波信号，忽略了采集到的脉搏波信号中存在一些形态畸形的脉搏波信号，从而导致脉诊结果不准确。These interference factors will produce some morphological distortions that affect the accuracy of subsequent analysis, and even make the pulse wave signal collected by the pulse sensor unusable. However, when processing the collected pulse wave signals, the default signals are all usable pulse wave signals, ignoring the presence of some morphologically deformed pulse wave signals in the collected pulse wave signals, resulting in inaccurate pulse diagnosis results.
针对现有技术的上述缺陷，本发明提供了一种基于信号质量评估的脉搏波预处理方法，所述方法可以根据采集到的脉搏波信号的波形特征信息、幅度特征信息以及周期差异信息及时调整脉象传感器的摆放位置或者参数，从而减少了获取到无效的脉搏波信号的概率。解决了现有技术中将脉象传感器采集到的脉搏波信号均作为可用脉搏波信号用于生成脉诊结果，忽略了采集到的脉搏波信号中存在一些形态畸形的脉搏波信号，从而导致脉诊结果不准确。In view of the above-mentioned defects of the prior art, the present invention provides a pulse wave preprocessing method based on signal quality assessment, which can timely adjust the placement position or parameters of the pulse sensor according to the waveform characteristic information, amplitude characteristic information and period difference information of the collected pulse wave signal, thereby reducing the probability of obtaining invalid pulse wave signals. The problem that the pulse wave signals collected by the pulse sensor are all used as available pulse wave signals to generate pulse diagnosis results in the prior art is solved, and some morphologically deformed pulse wave signals in the collected pulse wave signals are ignored, thereby causing inaccurate pulse diagnosis results.
如图1所示，本实施例提供一种基于信号质量评估的脉搏波预处理方法，所述方法包括如下步骤：As shown in FIG1 , this embodiment provides a pulse wave preprocessing method based on signal quality assessment, the method comprising the following steps:
步骤S100、获取原始脉搏波信号的波形特征信息，根据所述波形特征信息调节脉象传感器的位置信息，并根据调节后的所述脉象传感器的位置信息，获取第一脉搏波信号；所述波形特征信息用于反映所述脉搏波信号中全部周期的幅值变化规律。Step S100, obtaining waveform characteristic information of the original pulse wave signal, adjusting the position information of the pulse sensor according to the waveform characteristic information, and obtaining a first pulse wave signal according to the adjusted position information of the pulse sensor; the waveform characteristic information is used to reflect the amplitude variation law of all cycles in the pulse wave signal.
具体地，本实施例在初始阶段获取到脉搏波信号时，首先需要获取所述脉搏波信号的波形特征信息，所述波形特征信息用于反映所述脉搏波信号中全部周期的幅值变化规律。根据所述波形特征信息可以判断出采集脉搏波信号的脉象传感器的摆放位置是否正确。在实际应用过程中，脉象传感器是通过对桡动脉管施加不同压力采集到脉搏波博动力，进而生成脉搏波信号，而脉象传感器的摆放位置不同采集到的脉搏波信号也会不同，错误地摆放脉象传感器甚至会导致采集到完全异常的脉搏波信号。因此，如图2所示，本实施例首先需要根据脉搏波信号的波形特征信息确定脉象传感器的摆放位置是否正确，当摆放位置错误时对脉象传感器的位置信息进行调节后，重新采集用户的脉搏波信号，得到第一脉搏波信号。Specifically, when the present embodiment obtains the pulse wave signal in the initial stage, it is first necessary to obtain the waveform characteristic information of the pulse wave signal, and the waveform characteristic information is used to reflect the amplitude variation law of all cycles in the pulse wave signal. According to the waveform characteristic information, it can be judged whether the placement position of the pulse sensor for collecting the pulse wave signal is correct. In the actual application process, the pulse sensor collects the pulse wave power by applying different pressures to the radial artery tube, and then generates the pulse wave signal, and the pulse wave signal collected by the different placement positions of the pulse sensor will also be different, and the wrong placement of the pulse sensor may even lead to the collection of a completely abnormal pulse wave signal. Therefore, as shown in Figure 2, the present embodiment first needs to determine whether the placement position of the pulse sensor is correct according to the waveform characteristic information of the pulse wave signal, and when the placement position is wrong, the position information of the pulse sensor is adjusted, and the pulse wave signal of the user is recollected to obtain the first pulse wave signal.
在一种实现方式中，所述步骤S100具体包括如下步骤：In one implementation, step S100 specifically includes the following steps:
步骤S110、获取预设的截取时长数据和原始脉搏波信号，根据所述截取时长数据对所述原始脉搏波信号进行截取，得到检测信号段；Step S110, obtaining preset interception time data and an original pulse wave signal, intercepting the original pulse wave signal according to the interception time data to obtain a detection signal segment;
步骤S120、获取所述检测信号段的频域描述图像数据，根据所述频域描述图像数据获取所述检测信号段对应的幅度最大值出现的区间信息，并将所述区间信息作为所述波形特征信息；所述频域描述图像数据为反映脉搏波信号的幅值随频率变化的关系图像数据；Step S120, obtaining frequency domain description image data of the detection signal segment, obtaining interval information of the maximum amplitude corresponding to the detection signal segment according to the frequency domain description image data, and using the interval information as the waveform feature information; the frequency domain description image data is image data reflecting the relationship between the amplitude of the pulse wave signal and the frequency;
步骤S130、将所述区间信息与标准区间信息进行比较，当所述区间信息位于所述预设的标准区间信息之外时，确定所述脉搏波信号为全局波形异常信号；Step S130, comparing the interval information with the standard interval information, and when the interval information is outside the preset standard interval information, determining that the pulse wave signal is a global waveform abnormality signal;
步骤S140、对所述脉象传感器的摆放位置进行调节，并根据调节后的所述脉象传感器的位置信息，第一脉搏波信号。Step S140, adjusting the placement position of the pulse sensor, and generating a first pulse wave signal based on the adjusted position information of the pulse sensor.
本实施例需要在尽可能短的时间内及时检测出脉象传感器摆放位置错误的情况。在一种实现方式中，为了加快对脉搏波信号的处理速度，还可以对原始的脉搏波信号进行每隔预设数量点重新采样的操作，构建一个长度比原始的脉搏波信号长度更短的脉搏波信号，然后再对得到的短的该脉搏波信号进行处理。举例说明，可以对原始的脉搏波信号每隔10个点重新采一次样，构建出一个长度只有原始的脉搏波信号1/10长度的脉搏波信号。This embodiment needs to detect the situation that the pulse sensor is placed in the wrong position in a timely manner in the shortest possible time. In one implementation, in order to speed up the processing speed of the pulse wave signal, the original pulse wave signal can also be resampled every preset number of points to construct a pulse wave signal with a length shorter than the original pulse wave signal, and then the obtained short pulse wave signal is processed. For example, the original pulse wave signal can be resampled every 10 points to construct a pulse wave signal with a length of only 1/10 of the original pulse wave signal.
首先，本实施例会根据预设的截取时长数据对原始脉搏波信号进行截取，得到检测信号段。举例说明，所述截取时长数据可以设置为3秒，即计算机只需要通过对长度为3秒的检测信号段来判断采集的脉搏波信号是否异常，进而判断出脉象传感器的摆放位置是否异常，从而大大缩短检测时长。具体地，为了判断出采集的脉搏波信号是否异常，本实施例还需要获取所述检测信号段的频域描述图像数据，然后根据所述频域描述图像数据获取所述检测信号段的幅度最大值出现的区间信息。举例说明，如图8和图9所示，正常脉搏波信号的频域描述图像数据与异常脉搏波信号的频域描述图像数据中幅值随频率变化的规律明显不同。具体而言，正常的脉搏波信号具有与心跳对应的变化规律，所以正常脉搏波信号的频域描述图像数据中幅度最大值出现的区间与心跳频率相符。在一种实现方式中，本实施例可以根据一般人的心率范围设定标准区间为0.9-2.5HZ，当所述检测信号段的幅度最大值出现的区间与标准区间相符时，则判断采集到的脉搏波信号是正常的。而当所述检测信号段的幅度最大值出现的区间位于标准区间之外时，则判断采集到的脉搏波信号是全局波形异常信号，即波形整体异常的信号。在一种实现方式中，由于异常的脉搏波信号种类多样，有些异常的脉搏波信号的幅度最大值出现的区间也会与标准区间相符，为了降低误判率，因此本实施例还需要对脉搏波信号的幅度最大值出现的位置信息增加一个约束条件。简单来说，如图8和图9所示，对于异常的脉搏波信号来讲，其幅度最大值出现的区间在横坐标上会越靠近零点，即最小频率的幅值最大，原因是异常的脉搏波信号通常是相对平滑的信号，具有不规则的高频噪声。因此本实施例在判断出所述区间信息位于所述预设的标准区间信息之内时，根据所述频域描述图像数据确定所述检测信号段的幅度最大值对应的频率值。然后根据所述频域描述图像数据确定所述检测信号段的频率最小值，并将所述检测信号段的幅度最大值对应的频率值与所述频率最小值进行比较，当所述检测信号段的幅度最大值对应的频率值与所述频率最小值相等时，确定所述脉搏波信号为全局波形异常信号。First, this embodiment will intercept the original pulse wave signal according to the preset interception time data to obtain the detection signal segment. For example, the interception time data can be set to 3 seconds, that is, the computer only needs to judge whether the collected pulse wave signal is abnormal by the detection signal segment with a length of 3 seconds, and then judge whether the placement position of the pulse sensor is abnormal, thereby greatly shortening the detection time. Specifically, in order to judge whether the collected pulse wave signal is abnormal, this embodiment also needs to obtain the frequency domain description image data of the detection signal segment, and then obtain the interval information of the maximum amplitude of the detection signal segment according to the frequency domain description image data. For example, as shown in Figures 8 and 9, the law of amplitude variation with frequency in the frequency domain description image data of the normal pulse wave signal and the frequency domain description image data of the abnormal pulse wave signal is obviously different. Specifically, the normal pulse wave signal has a variation law corresponding to the heartbeat, so the interval where the maximum amplitude appears in the frequency domain description image data of the normal pulse wave signal is consistent with the heartbeat frequency. In one implementation, the present embodiment can set the standard interval to 0.9-2.5HZ according to the heart rate range of an average person. When the interval in which the maximum amplitude of the detection signal segment appears is consistent with the standard interval, the collected pulse wave signal is judged to be normal. When the interval in which the maximum amplitude of the detection signal segment appears is outside the standard interval, the collected pulse wave signal is judged to be a global waveform abnormal signal, that is, a signal with an overall abnormal waveform. In one implementation, since there are various types of abnormal pulse wave signals, the interval in which the maximum amplitude of some abnormal pulse wave signals appears will also be consistent with the standard interval. In order to reduce the misjudgment rate, the present embodiment also needs to add a constraint condition to the position information of the maximum amplitude of the pulse wave signal. In short, as shown in Figures 8 and 9, for an abnormal pulse wave signal, the interval in which the maximum amplitude appears will be closer to the zero point on the horizontal axis, that is, the amplitude of the minimum frequency will be the largest. The reason is that the abnormal pulse wave signal is usually a relatively smooth signal with irregular high-frequency noise. Therefore, when the present embodiment determines that the interval information is within the preset standard interval information, the frequency value corresponding to the maximum amplitude of the detection signal segment is determined according to the frequency domain description image data. Then, the minimum frequency value of the detection signal segment is determined according to the frequency domain description image data, and the frequency value corresponding to the maximum amplitude of the detection signal segment is compared with the minimum frequency value. When the frequency value corresponding to the maximum amplitude of the detection signal segment is equal to the minimum frequency value, the pulse wave signal is determined to be a global waveform abnormality signal.
由于全局波形异常信号通常是由于脉象传感器的摆放位置错误造成的，因此本实施例在判断出采集到的脉搏波信号为全局波形异常信号时，为了获取到有效的脉搏波信号，需要对脉象传感器的摆放位置进行调节后重新采集脉搏波信号，得到第一脉搏波信号。Since the global waveform abnormal signal is usually caused by the incorrect placement of the pulse sensor, when this embodiment determines that the collected pulse wave signal is a global waveform abnormal signal, in order to obtain a valid pulse wave signal, it is necessary to adjust the placement of the pulse sensor and re-collect the pulse wave signal to obtain a first pulse wave signal.
可以理解的是，若根据原始脉搏波信号的波形特征信息判断出所述原始脉搏波信号为正常的脉搏波信号，则不需要对脉象传感器的摆放位置进行调节，也不需要重新采集脉搏波信号，所述原始脉搏波信号可以进入下一个检测阶段。It is understandable that if the original pulse wave signal is judged to be a normal pulse wave signal based on its waveform characteristic information, there is no need to adjust the placement of the pulse sensor or re-collect the pulse wave signal, and the original pulse wave signal can enter the next detection stage.
如图1所示，所述方法还包括如下步骤：As shown in FIG1 , the method further comprises the following steps:
步骤S200、获取所述第一脉搏波信号的幅度特征信息，根据所述幅度特征信息调节所述脉象传感器的参数信息，并根据调节后的所述脉象传感器的参数信息，获取第二脉搏波信号；所述幅度特征信息用于反映所述第一脉搏波信号在时域上的幅值的强度信息。Step S200, obtaining the amplitude characteristic information of the first pulse wave signal, adjusting the parameter information of the pulse sensor according to the amplitude characteristic information, and obtaining the second pulse wave signal according to the adjusted parameter information of the pulse sensor; the amplitude characteristic information is used to reflect the intensity information of the amplitude of the first pulse wave signal in the time domain.
具体地，除了在采集脉搏波信号的初始阶段会出现采集到形态畸变的脉搏波信号之外，在采集脉搏波信号的中间阶段中也有可能会采集到形态畸变的脉搏波信号，从而导致采集到质量较差的脉搏波信号，进而对后续的脉诊结果的准确性产生影响。因此，本实施例还需要获取所述第一脉搏波信号的幅度特征信息，通过幅度特征信息判断所述脉象传感器的参数信息设置的是否合理，例如所述脉象传感器的采集压力或者电压等参数信息，当判断出脉象传感器的参数信息有问题时，及时对参数信息进行调节，然后重新采集用户的脉搏波信号，得到第二脉搏波信号，以减少采集到形态畸变的脉搏波信号的概率。Specifically, in addition to the morphologically distorted pulse wave signal being collected in the initial stage of collecting the pulse wave signal, the morphologically distorted pulse wave signal may also be collected in the intermediate stage of collecting the pulse wave signal, thereby collecting a pulse wave signal of poor quality, which in turn affects the accuracy of the subsequent pulse diagnosis results. Therefore, the present embodiment also needs to obtain the amplitude characteristic information of the first pulse wave signal, and judge whether the parameter information of the pulse sensor is reasonably set through the amplitude characteristic information, such as the parameter information of the pulse sensor such as the pressure or voltage collected. When it is judged that there is a problem with the parameter information of the pulse sensor, the parameter information is adjusted in time, and then the pulse wave signal of the user is recollected to obtain the second pulse wave signal, so as to reduce the probability of collecting the morphologically distorted pulse wave signal.
在一种实现方式中，所述步骤S200具体包括如下步骤：In one implementation, step S200 specifically includes the following steps:
步骤S210、获取所述第一脉搏波信号的时域描述图像数据，并获取所述时域描述图像数据中的幅度最大值以及幅度最小值，得到所述第一脉搏波信号的幅度特征信息；所述时域描述图像数据为反映所述脉搏波信号的幅值随时间变化的关系图像数据；Step S210, obtaining time domain description image data of the first pulse wave signal, and obtaining the maximum amplitude value and the minimum amplitude value in the time domain description image data, to obtain amplitude characteristic information of the first pulse wave signal; the time domain description image data is image data reflecting the relationship between the amplitude of the pulse wave signal and time;
步骤S220、将所述时域描述图像数据中幅值与所述幅值最大值的差距小于第一阈值的采样点作为波峰点，并将所述时域描述图像数据中幅值与所述幅值最小值的差距小于所述第一阈值的采样点作为波谷点；Step S220, taking the sampling points where the difference between the amplitude in the time domain description image data and the maximum amplitude is less than a first threshold as the peak points, and taking the sampling points where the difference between the amplitude in the time domain description image data and the minimum amplitude is less than the first threshold as the trough points;
步骤S230、根据所述波峰点和所述波谷点确定所述脉搏波信号为幅度异常信号，并根据所述幅度异常信号对应的信号类型调节所述脉象传感器的参数信息；所述幅度异常信号为所述波峰点和所述波谷点的特征信息与标准脉搏波信号不同的信号；Step S230, determining that the pulse wave signal is an abnormal amplitude signal according to the peak point and the trough point, and adjusting the parameter information of the pulse sensor according to the signal type corresponding to the abnormal amplitude signal; the abnormal amplitude signal is a signal whose characteristic information of the peak point and the trough point is different from that of the standard pulse wave signal;
步骤S240、根据调节后的所述脉象传感器的参数信息，获取第二脉搏波信号。Step S240: acquiring a second pulse wave signal according to the adjusted parameter information of the pulse sensor.
具体地，为了获取所述第一脉搏波信号的幅度特征信息，本实施例需要首先获取所述第一脉搏波信号对应的时域描述图像数据，并在所述时域描述图像数据中确定波峰点和波谷点，可以理解的是，时域描述图像数据中有可能存在多个波峰点和多个波谷点，波峰点之间/波谷点之间对应的幅值也并非严格相等，因此在判定波峰点和波谷点时，不能采用严格的判别标准。所以本实施例设置了第一阈值，作为判定波峰点和波谷点的标准(根据对数据库中饱和信号的统计，可以将所述第一阈值设置为10)。具体地，将所述时域描述图像数据中幅值与所述幅值最大值的差距小于第一阈值的采样点作为波峰点，并将所述时域描述图像数据中幅值与所述幅值最小值的差距小于所述第一阈值的采样点作为波谷点。之后，根据所述波峰点和所述波谷点可以确定脉搏波信号的幅度是否异常，并同时确定脉搏波信号对应的幅度异常信号的信号类型。并根据确定出的信号类型对所述脉象传感器的参数信息进行调节后重新采集到第二脉搏波信号。Specifically, in order to obtain the amplitude characteristic information of the first pulse wave signal, the present embodiment needs to first obtain the time domain description image data corresponding to the first pulse wave signal, and determine the peak point and the trough point in the time domain description image data. It can be understood that there may be multiple peak points and multiple trough points in the time domain description image data, and the amplitudes corresponding to the peak points/trough points are not strictly equal. Therefore, when determining the peak point and the trough point, a strict discrimination standard cannot be used. Therefore, the present embodiment sets a first threshold as a standard for determining the peak point and the trough point (according to the statistics of the saturated signal in the database, the first threshold can be set to 10). Specifically, the sampling point whose amplitude difference with the maximum amplitude in the time domain description image data is less than the first threshold is taken as the peak point, and the sampling point whose amplitude difference with the minimum amplitude in the time domain description image data is less than the first threshold is taken as the trough point. Afterwards, whether the amplitude of the pulse wave signal is abnormal can be determined according to the peak point and the trough point, and the signal type of the amplitude abnormal signal corresponding to the pulse wave signal can be determined at the same time. The parameter information of the pulse sensor is adjusted according to the determined signal type, and the second pulse wave signal is collected again.
在一种实现方式中，幅度异常信号的信号类型有很多种，为了确定脉搏波信号对应的幅度异常信号的信号类型，本实施例还需要在所述时域描述图像数据中获取所述波峰点的时域位置信息，根据所述波峰点的时域位置信息确定所述脉搏波信号的幅值达到波峰值的持续时长数据，得到波峰持续时长数据，所述波峰值为所述波峰点的幅值。并且还需要在所述时域描述图像数据中获取所述波谷点的时域位置信息，根据所述波谷点的时域位置信息确定所述脉搏波信号的幅值达到波谷值的持续时长数据，得到波谷持续时长数据，所述波谷值为所述波谷点的幅值。然后将所述波峰持续时长数据和所述波谷持续时长数据分别与第二阈值进行比较，当所述波谷持续时长数据或者所述波谷持续时长数据大于所述第二阈值，确定所述第一脉搏波信号对应的幅度异常信号类型为饱和信号。举例说明，如图6或者图7所示，与正常的脉搏波信号相比，饱和信号在时域描述图像数据上有明显的特征，即脉搏波信号达到了幅值最大值或者幅值最小值，并且这种状态会持续一定的时长。换言之，在时域描述图像数据中，连续多个点达到幅值最大值或者幅值最小值时，表示脉搏波信号已出现饱和状态。脉搏波信号出现饱和状态的直接原因是因为信号的电压值超出了模数转换器的转换上限和下限，根本原因是由于采样压力过大。并且饱和信号对于局部信息的丢失是不可逆的，虽然可以通过拟合的或者估计的方法对信号进行补偿，但是在采集时应该尽量避免采集到饱和信号。因此当判断所述第一脉搏波信号为饱和信号时，可以对所述脉象传感器的参数信息中的采样压力值进行调节，避免后续采集到的脉搏波信号再出现信号饱和的情况。In one implementation, there are many types of amplitude abnormal signals. In order to determine the signal type of the amplitude abnormal signal corresponding to the pulse wave signal, this embodiment also needs to obtain the time domain position information of the peak point in the time domain description image data, determine the duration data of the amplitude of the pulse wave signal reaching the peak value according to the time domain position information of the peak point, and obtain the peak duration data, wherein the peak value is the amplitude of the peak point. It is also necessary to obtain the time domain position information of the trough point in the time domain description image data, determine the duration data of the amplitude of the pulse wave signal reaching the trough value according to the time domain position information of the trough point, and obtain the trough duration data, wherein the trough value is the amplitude of the trough point. Then, the peak duration data and the trough duration data are respectively compared with the second threshold value. When the trough duration data or the trough duration data is greater than the second threshold value, it is determined that the type of the amplitude abnormal signal corresponding to the first pulse wave signal is a saturation signal. For example, as shown in Fig. 6 or Fig. 7, compared with the normal pulse wave signal, the saturation signal has obvious characteristics in the time domain description image data, that is, the pulse wave signal reaches the maximum amplitude or the minimum amplitude, and this state will last for a certain period of time. In other words, in the time domain description image data, when multiple points reach the maximum amplitude or the minimum amplitude, it means that the pulse wave signal has appeared in the saturation state. The direct reason for the saturation state of the pulse wave signal is that the voltage value of the signal exceeds the upper and lower limits of the conversion of the analog-to-digital converter, and the fundamental reason is that the sampling pressure is too large. And the saturation signal is irreversible for the loss of local information. Although the signal can be compensated by fitting or estimation methods, it should be avoided to collect the saturation signal as much as possible during collection. Therefore, when it is judged that the first pulse wave signal is a saturation signal, the sampling pressure value in the parameter information of the pulse sensor can be adjusted to avoid the situation of signal saturation in the subsequent collected pulse wave signal.
在一种实现方式中，还可以在所述时域描述图像数据中计算波峰值与波谷值的差值，其中，所述波峰值为所述波峰点的幅值，所述波谷值为所述波谷点的幅值。然后，将所述波峰值与所述波谷值的差值和第三阈值进行比较，当所述波峰值与所述波谷值的差值小于或者等于所述第三阈值时，确定所述第一脉搏波信号对应的幅度异常信号类型为幅值过小信号。举例说明，如图6所示，幅度过小信号的时域描述图像数据中波谷点与波峰点在幅值方向的垂直距离过小，因此对幅度过小信号的判别需要至少有一个完整的周期。幅度过小信号难以判断脉搏波的变化趋势，使得波峰点和波谷点这类关键区域变得难以识别或者丢失，从而导致有效信号强度不足，最终也会造成采集的脉搏波信号不可用。由于脉搏波信号的幅值强度可以通过提高脉象传感器对应的电机的电压值来增强，因此当判断出所述第一脉搏波信号为幅度过小信号时，可以对所述脉象传感器的参数信息中的电压值进行调节，调节完毕后重新采集用户的脉搏波信号，得到第二脉搏波信号。In one implementation, the difference between the peak value and the trough value can also be calculated in the time domain description image data, wherein the peak value is the amplitude of the peak point, and the trough value is the amplitude of the trough point. Then, the difference between the peak value and the trough value is compared with a third threshold value, and when the difference between the peak value and the trough value is less than or equal to the third threshold value, it is determined that the amplitude abnormal signal type corresponding to the first pulse wave signal is a signal with too small amplitude. For example, as shown in FIG6, the vertical distance between the trough point and the peak point in the amplitude direction in the time domain description image data of the signal with too small amplitude is too small, so the determination of the signal with too small amplitude requires at least one complete cycle. The signal with too small amplitude makes it difficult to judge the changing trend of the pulse wave, making key areas such as the peak point and the trough point difficult to identify or lost, resulting in insufficient effective signal strength, and ultimately making the collected pulse wave signal unusable. Since the amplitude strength of the pulse wave signal can be enhanced by increasing the voltage value of the motor corresponding to the pulse sensor, when it is determined that the first pulse wave signal is a signal with too small an amplitude, the voltage value in the parameter information of the pulse sensor can be adjusted. After the adjustment, the user's pulse wave signal can be collected again to obtain a second pulse wave signal.
可以理解的是，若所述第一脉搏波信号判断为正常的脉搏波信号，则不需要对脉象传感器的参数信息中的电压值进行调节，也不需要重新采集用户的脉搏波信号，所述第一脉搏波信号可以进入下一个检测环节。It is understandable that if the first pulse wave signal is judged to be a normal pulse wave signal, there is no need to adjust the voltage value in the parameter information of the pulse sensor, nor is there any need to re-collect the user's pulse wave signal, and the first pulse wave signal can enter the next detection link.
如图1所示，所述方法还包括如下步骤：As shown in FIG1 , the method further comprises the following steps:
步骤S300、获取所述第二脉搏波信号的周期差异信息，根据所述周期差异信息对所述第二脉搏波信号进行处理；所述周期差异信息用于反映所述第二脉搏波信号中单个周期之间在幅值变化上的差异信息。Step S300, obtaining period difference information of the second pulse wave signal, and processing the second pulse wave signal according to the period difference information; the period difference information is used to reflect the difference information in amplitude change between single periods in the second pulse wave signal.
具体地，由于脉搏波信号在采集的过程中，也有可能会受到不规则噪声的影响，例如由于身体抖动(如咳嗽、肢体运动)、肌肉颤抖、说话等原因引起的不规则噪声，又或者是由于人体呼吸产生的肺部活动会在皮肤表面产生低频呼吸波，被脉象传感器采集到以后，在脉搏波信号中形成基线漂移。可以理解的是，脉搏波信号是一种周期信号，即幅度值存在周期性变化的信号，因此这类不规则噪声有可能影响到脉搏波信号中的少数周期，从而形成局部异常的脉搏波信号；也有可能会影响到脉搏波信号中的大部分周期，从而导致采集到的脉搏波信号完全异常。因此本实施例需要获取所述第二脉搏波信号的周期差异信息，根据所述周期差异信息确定第二脉搏波信号单个周期之间在幅值变化上的差异信息，进而判断其受不规则噪声干扰的程度，并确定相应的处理方法。Specifically, the pulse wave signal may be affected by irregular noise during the acquisition process, such as irregular noise caused by body shaking (such as coughing, limb movement), muscle tremor, talking, etc., or the lung activity generated by human breathing will generate low-frequency respiratory waves on the skin surface, which will be collected by the pulse sensor and form a baseline drift in the pulse wave signal. It can be understood that the pulse wave signal is a periodic signal, that is, a signal with periodic changes in amplitude value. Therefore, this type of irregular noise may affect a few cycles in the pulse wave signal, thereby forming a locally abnormal pulse wave signal; it may also affect most of the cycles in the pulse wave signal, resulting in the collected pulse wave signal being completely abnormal. Therefore, this embodiment needs to obtain the period difference information of the second pulse wave signal, determine the difference information in amplitude change between single cycles of the second pulse wave signal according to the period difference information, and then judge the degree of interference by irregular noise, and determine the corresponding processing method.
在一种实现方式中，所述步骤S300具体包括如下步骤：In one implementation, step S300 specifically includes the following steps:
步骤S310、在所述时域描述图像数据中，以所述波峰点作为分割点对所述第二脉搏波信号进行分割，得到若干个单周期数据；Step S310, in the time domain description image data, segment the second pulse wave signal using the wave crest point as a segmentation point to obtain a plurality of single cycle data;
步骤S320、将所述若干个单周期数据中每一个单周期数据轮流作为第一信号，将除所述第一信号之外的单周期数据作为第二信号；Step S320: taking each of the plurality of single-cycle data as a first signal in turn, and taking the single-cycle data other than the first signal as a second signal;
步骤S330、获取所述第一信号和所述第二信号的方差信息，根据所述方差信息生成所述第二脉搏波信号的周期差异信息；Step S330, obtaining variance information of the first signal and the second signal, and generating period difference information of the second pulse wave signal according to the variance information;
步骤S340、根据所述周期差异信息确定所述第二脉搏波信号为波形异常信号，并根据所述波形异常信号对应的信号类型对所述第二脉搏波信号进行处理。Step S340: determine that the second pulse wave signal is a waveform abnormality signal according to the period difference information, and process the second pulse wave signal according to the signal type corresponding to the waveform abnormality signal.
具体地，如图5所示，本实施例对第二脉搏波信号的检测是持续性的，即边采集边检测。可以理解的是，正常的脉搏波信号，其不同周期之间在形态上的相似程度是极高的，即正常的脉搏波信号的幅值变化是规则地周期性变化。而对于局部受损的脉搏波信号而言，其受不规则噪声干扰的周期与其他周期在形态上肯定会有所差异。举例说明，如图10和图11所示，异常脉搏波信号单周期长度差异性较大，周期之间在形态上的聚集度也不高，整体分布较为发散。而对于正常脉搏波信号而言，大部分单周期在形态上和长度上都比较一致，少数位于干扰处的信号段有明显的离群现象。因此本实施例需要利用第二脉搏波信号中单个周期之间在幅值变化上的差异信息来判断第二脉搏波信号局部受不规则噪声影响的程度。Specifically, as shown in FIG5 , the detection of the second pulse wave signal in this embodiment is continuous, that is, the detection is performed while the pulse wave signal is collected. It can be understood that the morphological similarity between different cycles of a normal pulse wave signal is extremely high, that is, the amplitude change of a normal pulse wave signal is a regular periodic change. However, for a partially damaged pulse wave signal, the cycle interfered by irregular noise will definitely be different in morphology from other cycles. For example, as shown in FIG10 and FIG11 , the length of a single cycle of an abnormal pulse wave signal is quite different, the morphological aggregation between cycles is not high, and the overall distribution is relatively divergent. For a normal pulse wave signal, most of the single cycles are relatively consistent in morphology and length, and a few signal segments located at the interference point have obvious outliers. Therefore, this embodiment needs to use the difference information in the amplitude change between single cycles in the second pulse wave signal to determine the degree to which the second pulse wave signal is locally affected by irregular noise.
为了获取第二脉搏波信号的周期差异信息，本实施例需要对第二脉搏波信号按照步骤S310的方法进行周期分割，得到若干个单周期数据。由于波峰点具有易于检测和准确性高的特性，因此本实施例使用波峰点作为分割点可以提高检测算法的实时性。然后获取每一个单周期数据与其他单周期数据之间的方差信息，并根据这些方差信息生成所述第二脉搏波信号的周期差异信息。In order to obtain the period difference information of the second pulse wave signal, the present embodiment needs to perform period segmentation on the second pulse wave signal according to the method of step S310 to obtain a plurality of single-period data. Since the peak point has the characteristics of being easy to detect and having high accuracy, the present embodiment uses the peak point as the segmentation point to improve the real-time performance of the detection algorithm. Then, the variance information between each single-period data and other single-period data is obtained, and the period difference information of the second pulse wave signal is generated based on the variance information.
具体地，本实施例需要根据所述第一信号和所述第二信号之间的方差信息计算所述第一信号和所述第二信号之间的相关性系数，得到所述第一信号对应的相关性系数集合，然后将所述若干个单周期数据中每一个单周期数据的所述相关性系数集合作为所述第二脉搏波信号的周期差异信息。在一种实现方式中，由于单周期数据之间存在长度差异，直接对各个单周期数据之间的距离进行计算是不现实的。因此，如图3和图4所示，在计算相关性系数之前，可以先对各个单周期数据进行归一化操作，使之具有相同的长度。简言之，本实施例使用相关性系数来度量不同单周期数据之间的相似程度，并以此作为脉搏波信号的周期差异信息。举例说明，假设需要度量周期X和周期Y的相似程度，则计算周期X和周期Y的相关性系数的算法如下所示：Specifically, this embodiment needs to calculate the correlation coefficient between the first signal and the second signal based on the variance information between the first signal and the second signal, obtain the correlation coefficient set corresponding to the first signal, and then use the correlation coefficient set of each single-cycle data in the plurality of single-cycle data as the period difference information of the second pulse wave signal. In one implementation, due to the length difference between the single-cycle data, it is unrealistic to directly calculate the distance between each single-cycle data. Therefore, as shown in Figures 3 and 4, before calculating the correlation coefficient, each single-cycle data can be normalized to have the same length. In short, this embodiment uses the correlation coefficient to measure the similarity between different single-cycle data, and uses this as the period difference information of the pulse wave signal. For example, assuming that the similarity between cycle X and cycle Y needs to be measured, the algorithm for calculating the correlation coefficient between cycle X and cycle Y is as follows:
其中，r表示周期X和周期Y的相关性系数；σ表示信号的方差信息，和分别为信号X和Y的幅度值的均值。可以理解的是，如果脉搏波信号有n个周期，则每个周期都会有n-1个不同的相关性系数r，这n-1个不同的相关性系数r即组成该周期对应的相关性系数集合。并且r∈[-1，1]。当r越接近于1，表示周期X和周期Y的相似程度越高。Among them, r represents the correlation coefficient between period X and period Y; σ represents the variance information of the signal, and are the mean values of the amplitude values of signals X and Y, respectively. It can be understood that if the pulse wave signal has n cycles, each cycle will have n-1 different correlation coefficients r, and these n-1 different correlation coefficients r constitute the set of correlation coefficients corresponding to the cycle. And r∈[-1,1]. The closer r is to 1, the higher the similarity between cycle X and cycle Y.
在一种实现方式中，在计算单周期数据之间的相似性系数之前，还可以对单周期数据进行初步筛选操作，以去除一些长度过长或者过短的周期。具体地，由于脉搏波信号是一种生理信号，因此脉搏波信号中的每一个单周期数据的长度并不会完全相等，本实施例使用变量ξT来表示差异容许范围，Tp表示脉搏波信号的平均周期长度，可以得到每个单周期数据的长度限制：In one implementation, before calculating the similarity coefficient between single-cycle data, a preliminary screening operation can be performed on the single-cycle data to remove some cycles that are too long or too short. Specifically, since the pulse wave signal is a physiological signal, the length of each single-cycle data in the pulse wave signal is not completely equal. This embodiment uses the variable ξT to represent the difference tolerance range, and Tp represents the average cycle length of the pulse wave signal. The length limit of each single-cycle data can be obtained:
Tp-ξT≤T≤Tp+ξTTp -ξT ≤T ≤Tp +ξT
本实施例将获得的每一个单周期数据的相关性系数集合作为所述第二脉搏波信号的周期差异信息。完全异常的脉搏波信号绝大部分周期的相关性系数都远离1。局部异常的脉搏波信号的受干扰周期与其他正常周期之间的相关性系数远离1，而其它正常周期之间的相关性系数都接近1。因此本实施例利用相关系数的数值和分布，就可以确定第二脉搏波信号是否异常，并确定其对应的波形异常信号的信号类型。This embodiment uses the correlation coefficient set of each single-cycle data obtained as the period difference information of the second pulse wave signal. The correlation coefficients of most periods of the completely abnormal pulse wave signal are far from 1. The correlation coefficients between the disturbed period and other normal periods of the partially abnormal pulse wave signal are far from 1, while the correlation coefficients between other normal periods are close to 1. Therefore, this embodiment uses the value and distribution of the correlation coefficient to determine whether the second pulse wave signal is abnormal and determine the signal type of the corresponding waveform abnormal signal.
具体地，本实施例会预先设置一个相关性系数阈值，用于判断每一个单周期数据与其他单周期数据之间的相似程度。并计算每一个单周期数据的相关性系数集合中数值大于所述相关性系数阈值的相关性系数的个数比例。然后获取第一比例阈值，当所述个数比例小于所述第一比例阈值时，确定所述相关性系数集合对应的单周期数据为异常周期数据。再计算所述异常周期数据的数量占所述单周期数据的总数量的比例，得到异常周期数量比例，然后获取第二比例阈值，当所述异常周期数量比例大于所述第二比例阈值时，确定所述第二脉搏波信号为全局波形异常信号。举例说明，令Rx＝{ri|i＝1,...,n-1}表示周期X的相关性系数集合，定义一个参数Cx，用来表示集合Rx中相似性系数大于相关性系数阈值θ1的个数比例：Specifically, this embodiment pre-sets a correlation coefficient threshold value to determine the similarity between each single-cycle data and other single-cycle data. And calculate the ratio of the number of correlation coefficients in the correlation coefficient set of each single-cycle data whose values are greater than the correlation coefficient threshold. Then obtain a first ratio threshold value. When the number ratio is less than the first ratio threshold value, determine that the single-cycle data corresponding to the correlation coefficient set is abnormal cycle data. Then calculate the ratio of the number of abnormal cycle data to the total number of single-cycle data to obtain the ratio of the number of abnormal cycles, and then obtain a second ratio threshold value. When the ratio of the number of abnormal cycles is greater than the second ratio threshold value, determine that the second pulse wave signal is a global waveform abnormal signal. For example, let Rx ={ri |i=1,...,n-1} represent the correlation coefficient set of period X, and define a parameter Cx to represent the ratio of the number of similarity coefficients in the set Rx that is greater than the correlation coefficient threshold θ1 :
可以理解的是，若X是一个异常周期，则Cx的数值较低；若X是一个正常周期，则Cx的数值较高。当Cx的个数在该相关性系数集合中的比例小于一定阈值时，表示该相关性系数集合对应的单周期与其他绝大多数单周期的相似度低，则该单周期为异常周期。当脉搏波信号中异常周期占周期总数量的比例大于一定阈值的时候，则说明该脉搏波信号中的异常周期过多，则该脉搏波信号被判定为全局波形异常信号，是无效信号，因此将所述第二脉搏波信号删除，并重新采集用户的脉搏波信号。It can be understood that if X is an abnormal cycle, the value of Cx is low; if X is a normal cycle, the value of Cx is high. When the proportion of the number of Cx in the correlation coefficient set is less than a certain threshold, it means that the single cycle corresponding to the correlation coefficient set has a low similarity with the vast majority of other single cycles, and the single cycle is an abnormal cycle. When the proportion of abnormal cycles in the pulse wave signal to the total number of cycles is greater than a certain threshold, it means that there are too many abnormal cycles in the pulse wave signal, and the pulse wave signal is determined to be a global waveform abnormal signal, which is an invalid signal, so the second pulse wave signal is deleted and the user's pulse wave signal is re-collected.
在一种实现方式中，当所述异常周期数量比例小于或者等于所述第二比例阈值时，表示第二脉搏波信号中虽然存在异常周期，但是数量不算太多，只有局部的周期出现异常。然而由于正常信号中也有可能出现极少的异常周期，因此本实施例还需要将所述异常周期数量比例与第三比例阈值进行比较，所述第三比例阈值小于所述第二比例阈值，通过第三比例阈值判断第二脉搏波信号是正常的信号还是局部受损的信号。当所述异常周期数量比例大于所述第三比例阈值时，表示第二脉搏波信号受到一定程度的干扰，存在一定数量的异常周期信号，因此判定所述第二脉搏波信号为波形异常信号，且具体为局部波形异常信号，例如采集者出现咳嗽等原因都可能导致这种情况发生。为了确保后续获得准确的脉诊结果，本实施例需要确定所述局部波形异常信号对应的干扰因素，并排除所述干扰因素，然后重新采集用户的脉搏波信号，得到目标脉搏波信号。在一种实现方式中，当确定第二脉搏波信号为局部波形异常信号时，可以基于得到的波形差异信息定位出受到干扰的异常周期，并将该部分异常周期去除后得到目标脉搏波信号。In one implementation, when the ratio of the number of abnormal cycles is less than or equal to the second ratio threshold, it means that although there are abnormal cycles in the second pulse wave signal, the number is not too large, and only local cycles are abnormal. However, since there may be very few abnormal cycles in normal signals, this embodiment also needs to compare the ratio of the number of abnormal cycles with the third ratio threshold, and the third ratio threshold is less than the second ratio threshold. The third ratio threshold is used to determine whether the second pulse wave signal is a normal signal or a locally damaged signal. When the ratio of the number of abnormal cycles is greater than the third ratio threshold, it means that the second pulse wave signal is interfered to a certain extent, and there are a certain number of abnormal cycle signals. Therefore, the second pulse wave signal is determined to be a waveform abnormal signal, and specifically a local waveform abnormal signal. For example, the collector coughs and other reasons may cause this situation. In order to ensure that accurate pulse diagnosis results are obtained later, this embodiment needs to determine the interference factors corresponding to the local waveform abnormal signal, eliminate the interference factors, and then re-collect the user's pulse wave signal to obtain the target pulse wave signal. In one implementation, when the second pulse wave signal is determined to be a local waveform abnormality signal, the disturbed abnormal period can be located based on the obtained waveform difference information, and the target pulse wave signal can be obtained after removing the part of the abnormal period.
总的来讲，本实施例在采集到目标脉搏波之前，已经排除了传感器的摆放位置错误、压力过大、电压过小以及采集过程存在干扰等问题，因此最终采集到的目标脉搏波信号具有较强的周期性，且大部分周期之间在幅值强度和持续时长上也有较好的一致性。而且在幅值方向上，强度也保持了较为合理的范围，既不会因为过小导致幅值方向的分辨率不足，也不会因为过大而导致有效信息丢失。因此可以将目标脉搏波信号进行保存，用于后续计算机辅助脉诊。In general, before the target pulse wave is collected, the present embodiment has eliminated the problems of wrong placement of the sensor, excessive pressure, too low voltage, and interference in the collection process. Therefore, the target pulse wave signal collected in the end has strong periodicity, and most of the periods have good consistency in amplitude intensity and duration. Moreover, in the amplitude direction, the intensity also maintains a relatively reasonable range, which will not lead to insufficient resolution in the amplitude direction due to being too small, nor will it lead to loss of effective information due to being too large. Therefore, the target pulse wave signal can be saved for subsequent computer-assisted pulse diagnosis.
综上所述，本发明公开了一种基于信号质量评估的脉搏波预处理方法，本发明可以根据采集到的脉搏波信号的波形特征信息、幅度特征信息以及周期差异信息及时调整脉象传感器的摆放位置或者参数，从而减少了获取到无效的脉搏波信号的概率。解决了现有技术中将脉象传感器采集到的脉搏波信号均作为可用脉搏波信号用于生成脉诊结果，忽略了采集到的脉搏波信号中存在一些形态畸形的脉搏波信号，从而导致脉诊结果不准确。In summary, the present invention discloses a pulse wave preprocessing method based on signal quality assessment, which can timely adjust the placement position or parameters of the pulse sensor according to the waveform characteristic information, amplitude characteristic information and period difference information of the collected pulse wave signal, thereby reducing the probability of obtaining invalid pulse wave signals. The problem in the prior art that the pulse wave signals collected by the pulse sensor are all used as available pulse wave signals to generate pulse diagnosis results, ignoring the presence of some morphologically deformed pulse wave signals in the collected pulse wave signals, thereby causing inaccurate pulse diagnosis results is solved.
应当理解的是，本发明的应用不限于上述的举例，对本领域普通技术人员来说，可以根据上述说明加以改进或变换，所有这些改进和变换都应属于本发明所附权利要求的保护范围。It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110693033.9ACN113558584B (en) | 2021-06-22 | 2021-06-22 | Pulse wave preprocessing method based on signal quality evaluation |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110693033.9ACN113558584B (en) | 2021-06-22 | 2021-06-22 | Pulse wave preprocessing method based on signal quality evaluation |
| Publication Number | Publication Date |
|---|---|
| CN113558584A CN113558584A (en) | 2021-10-29 |
| CN113558584Btrue CN113558584B (en) | 2024-08-06 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110693033.9AActiveCN113558584B (en) | 2021-06-22 | 2021-06-22 | Pulse wave preprocessing method based on signal quality evaluation |
| Country | Link |
|---|---|
| CN (1) | CN113558584B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114052677B (en)* | 2021-11-19 | 2023-07-21 | 武汉联影智融医疗科技有限公司 | Pulse wave beat quality evaluation method, device, computer equipment and storage medium |
| CN114469017B (en)* | 2022-01-27 | 2025-01-17 | 上海国民集团健康科技有限公司 | Pulse condition data detection method and system |
| TWI838937B (en)* | 2022-11-01 | 2024-04-11 | 友達光電股份有限公司 | Pulse manifestation determining method |
| CN115886764B (en)* | 2022-11-18 | 2025-05-27 | 隽颐智能科技(珠海)有限公司 | Acquisition position determining method and device for PPG (Point-to-Point) signals |
| CN115919274B (en)* | 2022-12-30 | 2024-05-31 | 深圳市捷美瑞科技有限公司 | Pulse state detection method and device, electronic equipment and medium |
| CN119366877B (en)* | 2024-12-30 | 2025-03-14 | 雅安市人民医院 | Multi-parameter fusion pulse wave signal quality assessment method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112089405A (en)* | 2020-08-26 | 2020-12-18 | 上海大学 | Pulse wave characteristic parameter measuring and displaying device |
| CN112971740A (en)* | 2021-02-08 | 2021-06-18 | 上海掌门科技有限公司 | Method and equipment for diagnosing pulse through pulse diagnosing equipment |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE69833656T2 (en)* | 1997-08-26 | 2006-08-17 | Seiko Epson Corp. | DEVICE FOR DIAGNOSIS OF PULSE WAVES |
| JP2004135733A (en)* | 2002-10-16 | 2004-05-13 | Nippon Colin Co Ltd | Biomedical information measuring apparatus |
| CN103565428B (en)* | 2012-07-18 | 2015-12-09 | 株式会社日立制作所 | Pulse detecting device and pulse detection method |
| CN103815891B (en)* | 2012-09-18 | 2016-02-17 | 卡西欧计算机株式会社 | Pulse data checkout gear and pulse data detection method |
| US9949892B2 (en)* | 2013-10-11 | 2018-04-24 | Peking Union Medical College Hospital | Pulse oximetry-based cardio-pulmonary resuscitation (CPR) quality feedback systems and methods |
| CN203953636U (en)* | 2014-03-28 | 2014-11-26 | 深圳市大富网络技术有限公司 | A kind of blood pressure detector, Intelligent spire lamella, intelligent watch and communication system |
| KR101593412B1 (en)* | 2014-06-24 | 2016-02-15 | 주식회사 메디코아 | Acceleration plethysmography analysis apparatus and method using wave form frequency distribution |
| US20160100765A1 (en)* | 2014-10-10 | 2016-04-14 | Koninkl Philips Nv | Non-invasive blood pressure monitor and method of operating the same |
| CN104605939B (en)* | 2015-02-05 | 2019-07-16 | 腾讯科技（深圳）有限公司 | Physiological information processing method and information processing device |
| JP6729704B2 (en)* | 2016-09-02 | 2020-07-22 | 株式会社村田製作所 | Blood pressure estimation device |
| JP6934309B2 (en)* | 2017-03-30 | 2021-09-15 | ルネサスエレクトロニクス株式会社 | Pulse measuring device, pulse measuring method, and program |
| JP2019180666A (en)* | 2018-04-05 | 2019-10-24 | Ｗｉｎフロンティア株式会社 | Pulse wave detection terminal program, pulse wave analysis server program and pulse wave detection analysis terminal program |
| WO2020012793A1 (en)* | 2018-07-10 | 2020-01-16 | 国立大学法人香川大学 | Pulse-wave signal analysis device, pulse-wave signal analysis method and computer program |
| CN209932724U (en)* | 2019-01-08 | 2020-01-14 | 研和智能科技(杭州)有限公司 | Multi-wavelength signal fusion heart rate detection system based on wearable equipment |
| CN109512405B (en)* | 2019-01-08 | 2021-05-11 | 哈尔滨工业大学（深圳） | Pulse condition signal segmentation method based on local point detection |
| CN110251094A (en)* | 2019-05-14 | 2019-09-20 | 周常安 | Finger-worn physiological detection device |
| CN110755055A (en)* | 2019-10-18 | 2020-02-07 | 上海掌门科技有限公司 | A method and apparatus for determining waveform evaluation information of a pulse waveform |
| CN111839494A (en)* | 2020-09-04 | 2020-10-30 | 广东电网有限责任公司电力科学研究院 | Heart rate monitoring method and system |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112089405A (en)* | 2020-08-26 | 2020-12-18 | 上海大学 | Pulse wave characteristic parameter measuring and displaying device |
| CN112971740A (en)* | 2021-02-08 | 2021-06-18 | 上海掌门科技有限公司 | Method and equipment for diagnosing pulse through pulse diagnosing equipment |
| Publication number | Publication date |
|---|---|
| CN113558584A (en) | 2021-10-29 |
| Publication | Publication Date | Title |
|---|---|---|
| CN113558584B (en) | Pulse wave preprocessing method based on signal quality evaluation | |
| CN110664390B (en) | Heart rate monitoring system and method based on wristband PPG and deep learning | |
| Vadrevu et al. | A robust pulse onset and peak detection method for automated PPG signal analysis system | |
| CN110269600B (en) | Non-contact video heart rate detection method based on multivariate empirical mode decomposition and combined blind source separation | |
| CN112998690B (en) | A respiratory rate extraction method based on pulse wave multi-feature fusion | |
| CN105662345B (en) | heartbeat signal processing method, device and system | |
| CN104545818A (en) | Sleep apnea syndrome detection method based on pulse and blood oxygen signals | |
| Jang et al. | A robust method for pulse peak determination in a digital volume pulse waveform with a wandering baseline | |
| CN117828511B (en) | A method for processing anesthesia depth EEG signal data | |
| WO2023226223A1 (en) | Ppg signal quality evaluation method and apparatus and ppg signal processing method and system | |
| CN112155560A (en) | Apnea detection method and system based on real-time cardiac shock signal | |
| CN110313903A (en) | A kind of pulse wave frequency domain character parameter extracting method and device | |
| Suresh et al. | End-to-end deep learning for reliable cardiac activity monitoring using seismocardiograms | |
| CN114469131A (en) | Adaptive real-time ECG signal quality assessment method | |
| CN117918805A (en) | Pulse wave quality automatic evaluation system in oscillography blood pressure measurement based on multiple indexes | |
| Zhang et al. | Using rear smartphone cameras as sensors for measuring heart rate variability | |
| CN113729653A (en) | Human body pulse wave signal acquisition method | |
| CN115381424A (en) | Heart rate detection method based on double-Gaussian function model and linear regression iterative algorithm | |
| WO2023108331A1 (en) | Adaptive real-time electrocardiogram signal quality evaluation method | |
| CN116616727A (en) | Pulse positioning, collecting and analyzing method and device | |
| TWI815546B (en) | Establishing method of sleep apnea assessment program, sleep apnea assessment system, and sleep apnea assessment method | |
| CN116473539A (en) | Electrocardiosignal-based composite empirical mode decomposition respiration extraction method and system | |
| US20220039753A1 (en) | Modulation model of photoplethysmography signal for vital sign extraction | |
| CN119949802B (en) | MRI real-time triggering method, system, terminal and medium based on SCG-PPG signal | |
| CN111905229A (en) | A 5G-based piano music hypnotherapy control system and method |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |