WO2018161391A1 - 一种心电信号伪差识别方法及心电信号伪差识别装置 - Google Patents

一种心电信号伪差识别方法及心电信号伪差识别装置 Download PDF

Info

Publication number
WO2018161391A1
WO2018161391A1 PCT/CN2017/079426 CN2017079426W WO2018161391A1 WO 2018161391 A1 WO2018161391 A1 WO 2018161391A1 CN 2017079426 W CN2017079426 W CN 2017079426W WO 2018161391 A1 WO2018161391 A1 WO 2018161391A1
Authority
WO
WIPO (PCT)
Prior art keywords
ecg signal
signal
target
template
data segment
Prior art date
Application number
PCT/CN2017/079426
Other languages
English (en)
French (fr)
Inventor
郑慧敏
Original Assignee
深圳竹信科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳竹信科技有限公司 filed Critical 深圳竹信科技有限公司
Publication of WO2018161391A1 publication Critical patent/WO2018161391A1/zh

Links

Images

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • A61B5/318Heart-related electrical modalities, e.g. electrocardiography [ECG]
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • A61B5/318Heart-related electrical modalities, e.g. electrocardiography [ECG]
    • A61B5/346Analysis of electrocardiograms
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7203Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal

Definitions

  • the invention relates to the field of medicine, in particular to an ECG signal artifact recognition method and an ECG signal artifact recognition device.
  • ECG signal is one of the earliest biological signals that humans have studied and applied in medical clinic. It is easier to detect than other bioelectric signals and has more intuitive regularity. Therefore, ECG analysis technology promotes the development of medicine. Electrocardiography is an important method for clinical diagnosis of cardiovascular disease.
  • the final ECG signal is mixed with noise.
  • Some of these noises have fixed laws: baseline drift, myoelectric interference, power frequency interference, and so on.
  • Another kind of noise is often not caused by human heart activity, such as the movement of the hand during the measurement process, the contact surface between the hand and the electrode is not smooth, etc.
  • These noises are called artifacts. It usually has the characteristic of mutation.
  • the artifacts in the ECG signal have a great influence on the ECG signal feature extraction and subsequent feature analysis.
  • the artifact is the part of the ECG signal that is not the heart electrocardiogram that occurs when the heart is excited.
  • Existing methods for removing artifacts in ECG signals are mostly directed to dynamic electrocardiograms.
  • Dynamic electrocardiograms generally have the characteristics of long measurement time, which facilitates the detection to some extent.
  • most of these methods are based on first extracting the non-pseudo-difference part of the ECG signal, and then obtaining the eigenvalues of the RR interval, the mean value of the QRS complex, and the variance, and then detecting the signals one by one according to these values.
  • the ECG signal that does not meet the conditions is considered to be an artifact.
  • the false positive recognition mostly faces the dynamic electrocardiogram.
  • the pseudo-difference identification algorithms require a certain prior knowledge.
  • the threshold determination process requires the learning of the normal ECG signals.
  • Embodiments of the present invention provide a pseudo-difference identification method for an electrocardiogram signal and an artifact recognition apparatus for an electrocardiogram signal for identifying an artifact in the collected ECG signal.
  • a first aspect of the embodiments of the present invention provides a method for identifying an artifact of an electrocardiogram signal, which specifically includes:
  • Reading the original ECG signal determining a target ECG signal according to the original ECG signal, wherein the target ECG signal is an ECG signal for removing baseline drift, myoelectric interference, and power frequency interference;
  • the target ECG signal determines all R points in the target ECG signal; and identifies an artifact of the target ECG signal based on all R points of the target ECG signal.
  • a second aspect of the embodiments of the present invention provides an ECG signal artifact recognition apparatus, including:
  • a reading module for reading the original ECG signal
  • a first determining module configured to determine a target ECG signal according to the original ECG signal, wherein the target ECG signal is an ECG signal that removes baseline drift, myoelectric interference, and power frequency interference;
  • a second determining module configured to determine all R points in the target ECG signal according to the target ECG signal
  • an identification module configured to identify an artifact of the target ECG signal according to all R points of the target ECG signal.
  • a third aspect of the embodiments of the present invention provides an ECG signal artifact recognition apparatus, including:
  • a central processing unit a memory, a storage medium, a power supply, a wireless network interface, and an input/output interface
  • the central processor is operative to perform the operations performed according to any one of claims 1 to 9 by invoking operational instructions stored on the memory or storage medium.
  • the embodiment of the present invention has the following advantages: reading the original ECG signal; determining the target ECG signal according to the original ECG signal, and removing the baseline ECG signal from the baseline ECG signal and the power frequency interference The ECG signal; determining all R points in the target ECG signal according to the target ECG signal; and identifying the artifact of the target ECG signal according to all R points of the target ECG signal.
  • FIG. 1 is a schematic diagram of an embodiment of a method for identifying a center electrical signal artifact according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of an embodiment of a central electrical signal artifact recognition apparatus according to an embodiment of the present invention
  • FIG. 3 is a schematic diagram of another embodiment of a central electrical signal artifact recognition apparatus according to an embodiment of the present invention.
  • FIG. 4 is a schematic diagram showing the hardware structure of a central electrical signal artifact recognition apparatus according to an embodiment of the present invention.
  • Embodiments of the present invention provide an ECG signal artifact recognition method and an ECG signal artifact recognition apparatus for quickly identifying artifacts in an ECG signal without any prior knowledge.
  • an embodiment of a method for identifying a central electrical signal artifact in an embodiment of the present invention includes:
  • the original ECG signal in the process of detecting the ECG signal of the human body, the original ECG signal may be read first.
  • the first ECG signal after reading the original ECG signal, can be determined by removing the baseline drift in the original ECG signal by the calibration formula under the constraint condition:
  • the restriction bar is:
  • Representing the estimated baseline drift signal Representing the acquired signal
  • is a non-negative parameter that controls the baseline drift to approach the true signal.
  • Indicates the secondary variation of the signal z represents the first ECG signal after removing the baseline drift
  • D represents the quadratic variation matrix
  • I represents the unit matrix of the size.
  • the first electrocardiographic signal is removed by a method of stationary wavelet transform to remove the myoelectric interference of the first electrocardiographic signal and the power frequency interference to determine the target electrocardiographic signal.
  • the stationary wavelet transform method may be adopted to remove the myoelectric interference and the power frequency interference, and the first ECG signal is subjected to stationary wavelet transform, and then the preset threshold and the attenuation coefficient are selected. Myoelectric interference and power frequency interference are separated from the first ECG signal, thereby removing myoelectric interference and power frequency interference.
  • the target ECG signal can be determined by removing the myoelectric interference and the power frequency interference in the first ECG signal by the following formula:
  • is a free factor and its value is 2.5.
  • is a threshold, and Y is the target ECG signal;
  • W i, j is decomposed wavelet coefficients
  • N is the number of sampling points
  • u the attenuation coefficient of each layer
  • i is the number of layers of the decomposition of the original signal.
  • the target ECG signal may be subjected to spline wavelet transform to determine wavelet coefficients, and a maximum value minimum value pair in the wavelet coefficient is determined, and the target ECG signal is determined according to the minimum value of the maximum value. All R points process the wavelet coefficients to determine all R points in the target ECG signal.
  • the target ECG signal can be subjected to a spline wavelet transform of 3 times, the wavelet coefficient Mj3 of the third layer is selected for analysis, and an appropriate threshold is selected to remove a relatively small extreme point, and among the remaining data values, The maximum value and the minimum value pair, and find the position of the original signal corresponding to the maximum value and the minimum value, and the point having the largest amplitude in the range of the position of the original signal X corresponding to the maximum value and the minimum value is R point.
  • the Mj3 layer is leaked and deleted. When the distance between the detected adjacent R points is too small, it can be considered that the R point is detected, and the point that is not the R point is determined as the R point.
  • the respective data segments can be formed based on the position of each R point, which can be implemented by programming:
  • Rpeak is an array that identifies the abscissa of the R wave point of the ECG signal.
  • step 106 Determine whether the data segment of each R point is the initial input. If yes, go to step 107. If no, go to step 108.
  • the data segment corresponding to each R point may be separately input to determine whether the data segment of each R point is It is the first input, if yes, step 107 is performed, and if not, step 108 is performed.
  • the data segment corresponding to the R point input for the first time is used as the first template.
  • the data segment corresponding to the input R point when it is determined that the data segment corresponding to the input R point is the initial input, the data segment corresponding to the first input R point may be used as the first template.
  • step 108 Determine, according to the DTW algorithm, whether the similarity between the data segment of each R point and each template in the template library is less than a preset threshold. If yes, execute step 112. If no, perform step 109 to step 111.
  • step 109 when it is determined that the data segment corresponding to the input R point is not the initial input, whether the similarity between the data segment of each R point and each template in the template library is determined to be less than a preset threshold according to the DTW algorithm. If yes, go to step 112. If no, go to step 109 to step 111.
  • step 109 Determine whether the number of templates in the template library is less than a preset value. If yes, go to step 110. If no, go to step 111.
  • step 110 when it is determined that the similarity with the template in the template library is not less than a preset threshold, at this time, it can be determined whether the number of templates in the template library is less than a preset value, and if yes, step 110 is performed, and if no, step 111 is performed.
  • a new template may be established for the data segment corresponding to the template whose similarity is not less than the preset threshold in the data segment corresponding to each R point.
  • the template when determining that the number of templates in the template library is not less than a preset value, the template may be used.
  • the template in the library is deleted from the template with the smallest similarity of the data segments of each R point, and a new template is created according to the data segment corresponding to the template whose similarity is not less than the preset threshold.
  • the data segment corresponding to the template whose similarity is less than the preset threshold is marked as the same class as the template whose similarity is less than the preset threshold.
  • the ECG signal artifact recognition apparatus may cyclically perform steps 106 to 112 until all the data segments corresponding to the R points are classified.
  • the frequency of each category of the data segments corresponding to all the R points in the target ECG signal may be marked.
  • step 115 After obtaining the frequency of each category of the data segment table corresponding to all the R points in the target ECG signal, it can be determined whether the number of each category exceeds two categories, and if yes, step 115 is performed, and if not, then Go to step 116.
  • the category with the least frequency among the categories and the category with the second lowest frequency are marked as the artifacts in the target ECG signal.
  • the original ECG signal when detecting the human body ECG signal, the original ECG signal can be read first; the target ECG signal is determined according to the original ECG signal, and the target ECG signal is to remove baseline drift, myoelectric interference, and The power frequency interference ECG signal; determining all R points in the target ECG signal according to the target ECG signal; and identifying the artifact of the target ECG signal according to all R points of the target ECG signal.
  • an embodiment of a central electrical signal artifact recognition apparatus includes:
  • the reading module 201 is configured to read the original ECG signal
  • the first determining module 202 is configured to determine a target ECG signal according to the original ECG signal, where the target ECG signal is an ECG signal that removes baseline drift, myoelectric interference, and power frequency interference;
  • a second determining module 203 configured to determine all R points in the target ECG signal according to the target ECG signal
  • the identification module 204 is configured to identify an artifact of the target ECG signal according to all R points of the target ECG signal.
  • a central electrical signal artifact recognition apparatus includes:
  • the reading module 301 is configured to read the original ECG signal
  • the first determining module 302 is configured to determine a target ECG signal according to the original ECG signal, and the target ECG signal is an ECG signal that removes baseline drift, myoelectric interference, and power frequency interference;
  • a second determining module 303 configured to determine all R points in the target ECG signal according to the target ECG signal
  • the identification module 304 is configured to identify an artifact of the target ECG signal according to all R points of the target ECG signal.
  • the first determining module 302 may further include:
  • a first removing unit 3021 configured to remove the baseline drift by the method of secondary variation of the original ECG signal to determine the first ECG signal
  • the second removing unit 3022 is configured to remove the first electrocardiographic signal by the stationary wavelet transform method to remove the myoelectric interference of the first electrocardiographic signal and the power frequency interference to determine the target electrocardiographic signal.
  • the first removing unit 3021 is specifically configured to determine the first ECG signal by removing the baseline drift in the original ECG signal by using the following formula:
  • the restriction bar is:
  • Representing the estimated baseline drift signal Indicates the acquired signal.
  • is a non-negative parameter that controls the baseline drift to approach the true signal.
  • the second removing unit 3022 is specifically configured to:
  • the target electrocardiographic signal is determined by removing the myoelectric interference of the first electrocardiographic signal and the power frequency interference by the following formula:
  • is a free factor
  • the value is 2.5
  • is a threshold
  • Y is the target ECG signal
  • W i,j is the decomposed wavelet coefficient
  • N is the number of sampling points
  • u is the attenuation coefficient of each layer
  • i is the number of layers after decomposing the original signal.
  • the second determining module 303 may further include:
  • a first determining unit 3031 configured to perform a wavelet transform on the target ECG signal to determine a wavelet coefficient
  • the second determining unit 3032 is configured to process the wavelet coefficients to determine all R points in the target ECG signal.
  • the second determining unit 3032 is specifically configured to:
  • All R points in the target ECG signal are determined according to the maximum value minimum value pair.
  • the identification module 304 can further include:
  • a third determining unit 3041 configured to determine a data segment of each R point with all R point bit references in the target ECG signal
  • the first determining unit 3042 is configured to determine whether the data segment of each R point is the initial input
  • the second determining unit 3043 is configured to determine, according to the DTW algorithm, whether the similarity between the data segment of each R point and each template in the template library is less than a preset threshold according to the DTW algorithm.
  • the first marking unit 3044 is configured to: when the similarity of the template in the template library is less than a preset threshold in the data segment of each R point, the data segment corresponding to the template whose similarity is less than the preset threshold Marked as the same class as the template whose similarity is less than the preset threshold;
  • a looping unit 3045 configured to cyclically execute the actions of the first determining unit, the second determining unit, and the marking unit until the marking of the data segment corresponding to all the R points in the target electrocardiographic signal is classified;
  • the statistics unit 3046 is configured to count the frequency of each category of the data segment marker corresponding to all the R points in the target ECG signal;
  • the third determining unit 3047 is configured to determine whether the number of each category exceeds two categories;
  • a second marking unit 3048 configured to mark, when the number of each category exceeds two categories, a category with the least frequency among the categories and a category with the second lowest frequency among the categories as the artifact of the target ECG signal;
  • the first processing unit 3049 is configured to: when the data segment of each R point is the initial input, the data segment corresponding to the R point that is input for the first time is used as the first template;
  • the fourth determining unit 30410 is configured to determine whether the number of templates in the template library is less than a preset value when the similarity between the template of the template database and the template database is not greater than a preset threshold;
  • the second processing unit 30411 when the number of templates in the template library is less than a preset value, establish a new template according to the data segment corresponding to the template whose similarity is greater than the preset threshold;
  • the second processing unit 30411 is further configured to use, when the number of templates in the template library is not less than a preset value, The template in the template library is deleted from the template with the smallest similarity of the data segments of each R point, and a new template is created according to the data segment corresponding to the template whose similarity is greater than the preset threshold.
  • the ECG signal artifact recognition device can read the original ECG signal through the reading module 301, and the original determination module 302 is used according to the original.
  • the ECG signal determines the target ECG signal
  • the second determination module 303 determines all R points in the target ECG signal according to the target ECG signal
  • the recognition module 304 identifies the target ECG signal according to all R points in the target ECG signal.
  • the artifact It can be seen that when the artifact recognition needs to be performed, only the R point in the ECG signal needs to be determined, and no prior knowledge is needed.
  • FIG. 4 is a schematic structural diagram of an electrocardiogram signal artifact recognition apparatus according to an embodiment of the present invention.
  • the ECG signal artifact recognition apparatus 400 may generate a large difference due to different configurations or performances.
  • a central processing unit (CPU) 422 eg, one or more processors
  • memory 432 e.g, one or more storage media 430 (eg, one or one) storing application 442 or data 444 Take Shanghai storage device).
  • the memory 432 and the storage medium 430 may be short-term storage or persistent storage.
  • the program stored on storage medium 430 may include one or more modules (not shown), each of which may include a series of instruction operations in the server.
  • the central processing unit 422 can be configured to communicate with the storage medium 430 to perform a series of instruction operations in the storage medium 430 on the electrocardiographic signal artifact recognition apparatus 400.
  • the ECG signal artifact recognition device 400 may also include one or more power sources 426, one or more wired or wireless network interfaces 450, one or more input and output interfaces 458, and/or one or more operating systems 441, For example, Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.
  • the steps performed by the electrocardiographic signal artifact recognition apparatus in the above embodiment may be based on the ECG signal artifact recognition apparatus structure shown in FIG.
  • the disclosed system, apparatus, and method may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
  • the integrated unit if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium.
  • the technical solution of the present invention which is essential or contributes to the prior art, or all or part of the technical solution, may be embodied in the form of a software product stored in a storage medium.
  • a number of instructions are included to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Public Health (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Signal Processing (AREA)
  • Cardiology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Psychiatry (AREA)
  • Physiology (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

一种心电信号伪差识别方法及心电信号伪差识别装置,用于无需任何心电信号的先验知识来识别所采集到的心电信号中的伪差。所述方法包括:读取原始心电信号;根据所述原始心电信号确定目标心电信号,所述目标心电信号为去除基线漂移、肌电干扰以及工频干扰的心电信号;根据所述目标心电信号确定所述目标心电信号中的所有R点;根据所述目标心电信号的所有R点识别所述目标心电信号的伪差。

Description

一种心电信号伪差识别方法及心电信号伪差识别装置 技术领域
本发明涉及医学领域,具体涉及一种心电信号伪差识别方法及心电信号伪差识别装置。
背景技术
心电信号是人类最早研究并应用于医学临床的生物信号之一,它比其它生物电信号更易于检测,并且具有较直观的规律性,因而心电图分析技术促进了医学的发展。心电图检查是临床上诊断心血管疾病的重要方法。
在检测人体心电信号的过程中,最终获取的心电信号中会混有噪声。这些噪声中有一些具有固定的规律:比如基线漂移,肌电干扰,工频干扰等。而另外一种噪声,往往并不是人体心脏活动引起的,比如测量过程中,手的移动,手与电极接触面不光滑等,这些噪声,我们称之为伪差。其通常具有突变的特点。
心电信号中的伪差对心电信号特征提取以及之后的特征分析有很大影响。伪差即是心电信号中非凡不是心脏电激动而发生的心电图改变的部分。现有的去除心电信号中伪差的方法大都针对于动态心电图。动态心电图一般具有测量时间长的特点,这在一定程度上方便了检测。其次,这些方法大都基于先大致提取心电信号的非伪差部分,然后得到其RR间期,QRS波群的均值,方差等特征值,然后根据这些值对信号进行逐一检测。对于不符合条件的心电信号认定为伪差。
目前的心电信号中伪差识别大都面对于动态心电图,此外几乎所有的伪差识别算法都需要一定的先验知识,比如阈值的确定过程就需要对正常心电信号的学习。
发明内容
本发明实施例提供了一种心电信号的伪差识别方法及心电信号的伪差识别装置,用于识别所采集到的心电信号中的伪差。
本发明实施例第一方面提供了一种心电信号的伪差识别方法,具体包括:
读取原始心电信号;根据所述原始心电信号确定目标心电信号,所述目标心电信号为去除基线漂移、肌电干扰以及工频干扰的心电信号;根据所述 目标心电信号确定所述目标心电信号中的所有R点;根据所述目标心电信号的所有R点识别所述目标心电信号的伪差。
本发明实施例第二方面提供了一种心电信号伪差识别装置,包括:
读取模块,用于读取原始心电信号;
第一确定模块,用于根据所述原始心电信号确定目标心电信号,所述目标心电信号为去除基线漂移、肌电干扰以及工频干扰的心电信号;
第二确定模块,用于根据所述目标心电信号确定所述目标心电信号中的所有R点;
识别模块,用于根据所述目标心电信号的所有R点识别所述目标心电信号的伪差。
本发明实施例第三方面提供了一种心电信号伪差识别装置,包括:
中央处理器、存储器、存储介质、电源、无线网络接口以及输入输出接口;
通过调用所述存储器或存储介质上存储的操作指令,所述中央处理器,用于执行如权利要求1至9中任一项所执行的操作。
从以上技术方案可以看出,本发明实施例具有以下优点:读取原始心电信号;根据原始心电信号确定目标心电信号,目标心电信号为去除基线漂移、肌电干扰以及工频干扰的心电信号;根据目标心电信号确定所述目标心电信号中的所有R点;根据目标心电信号的所有R点识别目标心电信号的伪差。由此可以看出,只需要将原始心电信号去除基线漂移、几点干扰以及工频干扰,之后标定去除基线漂移、肌电干扰以及工频干扰的心电信号的所有R点,根据所有R点即可以识别心电信号中的伪差,无需一定的先验知识,即可以识别心电信号中的伪差。
附图说明
图1为本发明实施例中心电信号伪差识别方法的实施例示意图;
图2为本发明实施例中心电信号伪差识别装置的一个实施例示意图;
图3为本发明实施例中心电信号伪差识别装置的另一实施例示意图;
图4为本发明实施例中心电信号伪差识别装置的硬件结构示意图。
具体实施方式
本发明实施例提供了一种心电信号伪差识别方法及心电信号伪差识别装置,用于无需任何先验知识即可快速识别心电信号中的伪差。
本发明的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”、“第四”等(如果存在)是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换,以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。此外,术语“包括”和“具有”以及他们的任何变形,意图在于覆盖不排他的包含,例如,包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元,而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
请参阅图1,本发明实施例中心电信号伪差识别方法的一个实施例包括:
101、读取原始心电信号。
本实施例中,在检测人体心电信号的过程中,可以首先读取到原始心电信号。
102、将原始心电信号通过二次变分的方法去除基线漂移确定第一心电信号。
本实施例中,当读取到原始心电信号之后,可以在限制条件下,通过入校公式去除原始心电信号中的基线漂移确定第一心电信号:
Figure PCTCN2017079426-appb-000001
限制条为:
Figure PCTCN2017079426-appb-000002
其中,
Figure PCTCN2017079426-appb-000003
代表估计的基线漂移信号,
Figure PCTCN2017079426-appb-000004
表示采集到的信号,ρ为控制基线漂移趋近真实信号的非负参数,
Figure PCTCN2017079426-appb-000005
表示信号的二次变分,z表示去除基线漂移后的心电信号即第一心电信号,D表示二次变分矩阵,I表示对大小的单位矩阵。
103、将第一心电信号通过平稳小波变换的方法去除第一心电信号的肌电干扰以及工频干扰确定所述目标心电信号。
本实施例中,可以采取平稳小波变换的方法去除肌电干扰以及工频干扰,将第一心电信号进行平稳小波变换,然后选取预设的阈值以及衰减系数,将 肌电干扰以及工频干扰从第一心电信号中分离出来,从而将肌电干扰以及工频干扰去除。
可以通过如下公式去除第一心电信号中的肌电干扰以及工频干扰确定目标心电信号:
Figure PCTCN2017079426-appb-000006
其中,α为自由因子,其值为2.5。γ为阈值,Y为所述目标心电信号;
通过如下公式计算:
Figure PCTCN2017079426-appb-000007
其中,Wi,j为分解后的小波系数,N为采样点数,u为每一层的衰减系数,i为将所述原始信号分解后的层数。
104、根据目标心电信号确定目标心电信号中的所有R点。
本实施例中,可以将目标心电信号进行样条小波变换确定小波系数,确定所述小波系数中的极大值极小值对,根据极大值极小值对确定目标心电信号中的所有R点对小波系数进行处理确定所述目标心电信号中的所有R点。例如可以对目标心电信号进行3次样条小波变换,选取第3层的小波系数Mj3进行分析,并选取合适的阈值去掉相对较小的极值点,在剩下的数据值中,找出极大值和极小值对,并找出极大值和极小值对应的原始信号的位置,在极大值以及极小值对应的原始信号X的位置范围内幅值最大的点即为R点。一个遍历之后,对Mj3层进行捡漏和删除多检点,当检出的相邻R点的距离过小时,此时可以认为是多检了R点,即将不是R点的点确定成R点,删除幅值更小的R点;当检出的相邻R点的距离过大时,则认为检漏了,则调整预置,重新检测R点,在这一段信号中调整阈值,按照上述步骤再次检测R点。
105、以目标心电信号中的所有R点位置基准确定每个R点的数据段。
本实施例,当得到目标心电信号中的所有R点之后,可以以每一个R点的位置为基准形成各自的数据段,可以通过编程实现:
x(Rpeak(i)-floor((rpeak(i)-Rpeak(i-1))/3):Rpeak(i)+floor((RPeak(i+1)-Rpeak( i))/2)),其中Rpeak为标识心电信号R波点的横坐标的数组。
106、判断每个R点的数据段是否是初次输入,若是,则执行步骤107,若否,则执行步骤108。
本实施例中,在以每个R点的位置为基准确定每个R点的数据段之后,可以分别在每个R点对应的数据段输入时进行判断,判断每个R点的数据段是否是初次输入,若是,则执行步骤107,若否,则执行步骤108。
107、将初次输入的R点对应的数据段作为第一个模板。
本实施例中,当确定输入的R点对应的数据段是初次输入时,可以将初次输入的R点对应的数据段作为第一个模板。
108、根据DTW算法分别判断每个R点的数据段与模板库中的各个模板的相似度是否小于预设的阈值,若是,则执行步骤112,若否,则执行步骤109至步骤111。
本实施例中,当确定输入的R点对应的数据段不是初次输入时,可以根据DTW算法分别判断每个R点的数据段与模板库中的各个模板的相似度是否小于预设的阈值,若是,则执行步骤112,若否,则执行步骤109至步骤111。
109、判断模板库中的模板的数量是否小于预设值,若是,则执行步骤110,若否,则执行步骤111。
本实施例中,当根据DTW算法分别判断每个R点的数据段与模板库中的各个模板的相似度,当确定存在有与模板库中的模板的相似度不小于预设的阈值时,此时,可以判断模板库中的模板的数量是否小于预设值,若是,则执行步骤110,若否,则执行步骤111。
110、根据相似度不小于预设的阈值的模板对应的数据段建立新的模板。
本实施例中,当确定模板库中的模板数量小于预设值时,可以将与每个R点对应的数据段中的相似度不小于预设的阈值的模板对应的数据段建立新的模板。
111、将模板库中的模板与每个R点的数据段的相似度最小的模板删除,且根据相似度不小于预设的阈值的模板对应的数据段建立新的模板。
本实施例中,当确定模板库中的模板数量不小于预设值时,可以将模板 库中的模板与每个R点的数据段的相似度最小的模板删除,且根据相似度不小于预设的阈值的模板对应的数据段建立新的模板。
112、将相似度小于预设的阈值的模板对应的数据段标记为与相似度小于预设的阈值的模板为同一类。
本实施例中,当根据DTW算法分别判断每个R点的数据段与模板库中的各个模板的相似度之后,确定存在有与模板库中的模板的相似度小于预设的阈值时,可以将相似度小于预设的阈值的模板对应的数据段标记为与相似度小于预设的阈值的模板为同一类。
需要说明的是,心电信号伪差识别装置可以循环执行步骤106至步骤112直至所有的R点对应的数据段分类完毕。
113、统计目标心电信号中的所有R点对应的数据段标记的各个类别的频数。
本实施例中,当目标心电信号中的所有R点对应的数据段分类完毕之后,可以通过目标心电信号中的所有R点对应的数据段标记的各个类别的频数。
114、判断各个类别的数量是否超过两类,若是,则执行步骤115,若否,则执行步骤116。
本实施例中,当得到目标心电信号中的所有R点对应的数据段表的各个类别的频数之后,可以判断各个类别的数量是否超过两类,若是,则执行步骤115,若否,则执行步骤116。
115、将各个类别中频数最少的类别以及各个类别中频数第二少的类别标记为目标心电信号的伪差。
本实施例中,当各个类别的数量超过两类,则将各个类别中频数最少的类别以及频数第二少的类别标记为目标心电信号中的伪差。
116、确定目标心电信号中没有伪差。
本实施例中,当各个类别的数量没有超过两类,则确定目标心电信号中没有伪差。
需要说明的是,由于早搏信号与长长心电信号的相似度所在的数量级在e-4和e-5之间,而伪差与正常心电信号的相似度数量级在e-3以上,因此通过DTW计算出来的数值越大,说明差异越大,因此可以很容易的将早搏信号 与伪差分别出来,防止将早搏信号误认为是伪差。
综上所述,可以看出,当检测人体心电信号时,可以先读取原始心电信号;根据原始心电信号确定目标心电信号,目标心电信号为去除基线漂移、肌电干扰以及工频干扰的心电信号;根据目标心电信号确定所述目标心电信号中的所有R点;根据目标心电信号的所有R点识别目标心电信号的伪差。由此可以看出,只需要将原始心电信号去除基线漂移、几点干扰以及工频干扰,之后标定去除基线漂移、肌电干扰以及工频干扰的心电信号的所有R点,根据所有R点即可以识别心电信号中的伪差,无需一定的先验知识,即可以识别心电信号中的伪差。
上面从心电信号伪差识别的方法的角度对本发明实施例进行描述,下面从心电信号伪差识别装置的角度对本发明实施例进行描述。
请参阅图2,本发明实施例中心电信号伪差识别装置的一个实施例包括:
读取模块201,用于读取原始心电信号;
第一确定模块202,用于根据原始心电信号确定目标心电信号,目标心电信号为去除基线漂移、肌电干扰以及工频干扰的心电信号;
第二确定模块203,用于根据目标心电信号确定目标心电信号中的所有R点;
识别模块204,用于根据目标心电信号的所有R点识别目标心电信号的伪差。
为了便于理解,下面结合图3进行详细说明。
请参阅图3,本发明实施例中心电信号伪差识别装置包括:
读取模块301,用于读取原始心电信号;
第一确定模块302,用于根据原始心电信号确定目标心电信号,目标心电信号为去除基线漂移、肌电干扰以及工频干扰的心电信号;
第二确定模块303,用于根据目标心电信号确定目标心电信号中的所有R点;
识别模块304,用于根据目标心电信号的所有R点识别目标心电信号的伪差。
其中,第一确定模块302可以进一步包括:
第一去除单元3021,用于将原始心电信号通过二次变分的方法去除基线漂移确定第一心电信号;
第二去除单元3022,用于将第一心电信号通过平稳小波变换的方法去除第一心电信号的肌电干扰以及工频干扰确定所述目标心电信号。
其中,第一去除单元3021具体用于:在限制条件下,通过如下公式去除原始心电信号中的基线漂移确定第一心电信号:
Figure PCTCN2017079426-appb-000008
限制条为:
Figure PCTCN2017079426-appb-000009
其中,其中
Figure PCTCN2017079426-appb-000010
代表估计的基线漂移信号,
Figure PCTCN2017079426-appb-000011
表示采集到的信号。ρ为控制基线漂移趋近真实信号的非负参数。
Figure PCTCN2017079426-appb-000012
表示信号的二次变分;z表示去除基线漂移后的心电信号即所述第一心电信号,D表示二次变分矩阵,I表示对大小的单位矩阵。
其中,第二去除单元3022具体用于:
通过如下公式去除第一心电信号的肌电干扰以及工频干扰确定目标心电信号:
Figure PCTCN2017079426-appb-000013
其中,α为自由因子,其值为2.5,γ为阈值,Y为所述目标心电信号;
通过如下公式计算:
Figure PCTCN2017079426-appb-000014
其中,Wi,j为分解后的小波系数,N为采样点数,u为每一层的衰减系数,i为将所述原始信号分解后的层数。
其中,第二确定模块303可以进一步包括:
第一确定单元3031,用于将目标心电信号进行样条小波变换确定小波系数;
第二确定单元3032,用于对小波系数进行处理确定目标心电信号中的所有R点。
其中,第二确定单元3032具体用于:
确定小波系数中的极大值极小值对;
根据极大值极小值对确定目标心电信号中的所有R点。
其中,识别模块304可以进一步包括:
第三确定单元3041,用于以目标心电信号中的所有R点位基准确定每个R点的数据段;
第一判断单元3042,用于判断每个R点的数据段是否是初次输入;
第二判断单元3043,用于在每个R点的数据段不是初次输入,则根据DTW算法分别判断每个R点的数据段与模板库中的各个模板的相似度是否小于预设的阈值;
第一标记单元3044,用于在每个R点的数据段中存在有与模板库中的模板的相似度小于预设的阈值时,则将相似度小于预设的阈值的模板对应的数据段标记为与相似度小于预设的阈值的模板为同一类;
循环单元3045,用于循环执行第一判断单元、第二判断单元以及标记单元的动作直至目标心电信号中的所有R点对应的数据段的标记分类完毕;
统计单元3046,用于统计目标心电信号中的所有R点对应的数据段标记的各个类别的频数;
第三判断单元3047,用于判断各个类别的数目是否超过两类;
第二标记单元3048,用于在各个类别的数目超过两类时,将各个类别中频数最少的类别以及各个类别中频数第二少的类别标记为目标心电信号的伪差;
第一处理单元3049,用于在每个R点的数据段是初次输入时,将初次输入的R点对应的数据段作为第一个模板;
第四判断单元30410,用于当每个R点的数据段中不存在有与模板数据库的模板的相似度大于预设的阈值时,判断模板库中的模板的数量是否小于预设值;
第二处理单元30411,用于模板库中的模板的数量小于预设值时,根据相似度大于预设的阈值的模板对应的数据段建立新的模板;
第二处理单元30411还用于模板库中的模板的数量不小于预设值时,将 模板库中的模板与每个R点的数据段的相似度最小的模板删除,且根据相似度大于预设的阈值的模板对应的数据段建立新的模板。
本实施例中的心电信号伪差识别装置中的各模块与单元之间的交互方式如前述图1所示实施例中的描述,具体此处不再赘述。
综上所述可以看出,当接收到的心电信号需要进行伪差识别时,心电信号伪差识别装置可以通过读取模块301读取原始心电信号,通过第一确定模块302根据原始心电信号确定目标心电信号,通过第二确定模块303根据目标心电信号确定目标心电信号中的所有R点,通过识别模块304根据目标心电信号中的所有R点识别目标心电信号的伪差。由此可以看出,当需要进行伪差识别时,只需要确定心电信号中的R点即可,不需要任何的先验知识。
请参阅图4,图4是本发明实施例提供的一种心电信号伪差识别装置的结构示意图,该心电信号伪差识别装置400可因配置或性能不同而产生比较大的差异,可以包括一个或一个以上中央处理器(central processing units,CPU)422(例如,一个或一个以上处理器)和存储器432,一个或一个以上存储应用程序442或数据444的存储介质430(例如一个或一个以上海量存储设备)。其中,存储器432和存储介质430可以是短暂存储或持久存储。存储在存储介质430的程序可以包括一个或一个以上模块(图示没标出),每个模块可以包括对服务器中的一系列指令操作。更进一步地,中央处理器422可以设置为与存储介质430通信,在心电信号伪差识别装置400上执行存储介质430中的一系列指令操作。
心电信号伪差识别装置400还可以包括一个或一个以上电源426,一个或一个以上有线或无线网络接口450,一个或一个以上输入输出接口458,和/或,一个或一个以上操作系统441,例如Windows ServerTM,Mac OS XTM,UnixTM,LinuxTM,FreeBSDTM等等。
上述实施例中由心电信号伪差识别装置所执行的步骤可以基于该图4所示的心电信号伪差识别装置结构。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统,装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本申请所提供的几个实施例中,应该理解到,所揭露的系统,装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本发明各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(ROM,Read-Only Memory)、随机存取存储器(RAM,Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述,以上实施例仅用以说明本发明的技术方案,而非对其限制;尽管参照前述实施例对本发明进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。

Claims (19)

  1. 一种心电信号伪差识别方法,其特征在于,包括:
    读取原始心电信号;
    根据所述原始心电信号确定目标心电信号,所述目标心电信号为去除基线漂移、肌电干扰以及工频干扰的心电信号;
    根据所述目标心电信号确定所述目标心电信号中的所有R点;
    根据所述目标心电信号的所有R点识别所述目标心电信号的伪差。
  2. 根据权利要求1所述的方法,其特征在于,所述根据所述原始心电信号确定目标心电信号包括:
    将所述原始心电信号通过二次变分的方法去除基线漂移确定第一心电信号;
    将所述第一心电信号通过平稳小波变换的方法去除所述第一心电信号的肌电干扰以及工频干扰确定所述目标心电信号。
  3. 根据权利要求2所述的方法,其特征在于,所述将所述原始心电信号通过二次变分的方法去除基线漂移确定第一心电信号包括:
    在限制条件下,通过如下公式去除所述原始心电信号中的基线漂移确定第一心电信号:
    Figure PCTCN2017079426-appb-100001
    所述限制条为:
    Figure PCTCN2017079426-appb-100002
    其中,
    Figure PCTCN2017079426-appb-100003
    代表估计的基线漂移信号,
    Figure PCTCN2017079426-appb-100004
    表示采集到的信号,ρ为控制基线漂移趋近真实信号的非负参数,
    Figure PCTCN2017079426-appb-100005
    表示信号的二次变分,z表示去除基线漂移后的心电信号即所述第一心电信号,D表示二次变分矩阵,I表示对大小的单位矩阵。
  4. 根据权利要求2或3所述的方法,其特征在于,所述将所述第一心电信号通过平稳小波变换的方法去除所述第一心电信号的肌电干扰以及工频干扰确定所述目标心电信号包括:
    通过如下公式去除所述第一心电信号的肌电干扰以及工频干扰确定所述目标心电信号:
    Figure PCTCN2017079426-appb-100006
    其中,α为自由因子,其值为2.5。γ为阈值,Y为所述目标心电信号;
    通过如下公式计算:
    Figure PCTCN2017079426-appb-100007
    其中,Wi,j为分解后的小波系数,N为采样点数,u为每一层的衰减系数,i为将所述原始信号分解后的层数。
  5. 根据权利要求1所述的方法,其特征在于,所述根据所述目标心电信号确定所述目标心电信号中的所有R点包括:
    将所述目标心电信号进行样条小波变换确定小波系数;
    对所述小波系数进行处理确定所述目标心电信号中的所有R点。
  6. 根据权利要求5所述的方法,其特征在于,所述对所述小波系数进行处理确定所述目标心电信号中的所有R点包括:
    确定所述小波系数中的极大值极小值对;
    根据所述极大值极小值对确定所述目标心电信号中的所有R点。
  7. 根据权利要求1所述的方法,其特征在于,所述根据所述目标心电信号的所有R点识别所述目标心电信号的伪差包括:
    步骤1:以所述目标心电信号中的所有R点位基准确定每个R点的数据段;
    步骤2:判断每个R点的数据段是否是初次输入;
    步骤3:若所述每个R点的数据段不是初次输入,则根据DTW算法分别判断所述每个R点的数据段与模板库中的各个模板的相似度是否小于预设的阈值;
    步骤4:若所述每个R点的数据段中存在有与所述模板库中的模板的相似度小于预设的阈值时,则将所述相似度小于预设的阈值的模板对应的数据段标记为与所述相似度小于预设的阈值的模板为同一类;
    循环执行步骤2至步骤4直至所述目标心电信号中的所有R点对应的数据段的标记分类完毕;
    统计所述目标心电信号中的所有R点对应的数据段标记的各个类别的频数;
    判断所述各个类别的数目是否超过两类;
    若否,则将所述各个类别中频数最少的类别以及所述各个类别中频数第二少的类别标记为所述目标心电信号的伪差。
  8. 根据权利要求7所述的方法,其特征在于,当所述每个R点的数据段是初次输入时,所述方法还包括:
    将所述初次输入的R点对应的数据段作为第一个模板。
  9. 根据权利要求7所述的方法,其特征在于,当所述每个R点的数据段中不存在有与所述模板数据库的模板的相似度大于预设的阈值时,所述方法还包括:
    判断所述模板库中的模板数据是否小于预设值;
    若是,则根据所述相似度大于预设的阈值的模板对应的数据段建立新的模板;
    若否,则将所述模板库中的模板与所述每个R点的数据段的相似度最小的模板删除,且根据所述相似度大于预设的阈值的模板对应的数据段建立新的模板。
  10. 一种心电信号伪差识别装置,其特征在于,包括:
    读取模块,用于读取原始心电信号;
    第一确定模块,用于根据所述原始心电信号确定目标心电信号,所述目标心电信号为去除基线漂移、肌电干扰以及工频干扰的心电信号;
    第二确定模块,用于根据所述目标心电信号确定所述目标心电信号中的所有R点;
    识别模块,用于根据所述目标心电信号的所有R点识别所述目标心电信号的伪差。
  11. 根据权利要求1所述的心电信号伪差识别装置,其特征在于,所述第一确定模块包括:
    第一去除单元,用于将所述原始心电信号通过二次变分的方法去除基线漂移确定第一心电信号;
    第二去除单元,用于将所述第一心电信号通过平稳小波变换的方法去除所述第一心电信号的肌电干扰以及工频干扰确定所述目标心电信号。
  12. 根据权利要求11所述的心电信号伪差识别装置,其特征在于,所述第一去除单元具体用于:
    在限制条件下,通过如下公式去除所述原始心电信号中的基线漂移确定第一心电信号:
    Figure PCTCN2017079426-appb-100008
    所述限制条为:
    Figure PCTCN2017079426-appb-100009
    其中,其中
    Figure PCTCN2017079426-appb-100010
    代表估计的基线漂移信号,
    Figure PCTCN2017079426-appb-100011
    表示采集到的信号。ρ为控制基线漂移趋近真实信号的非负参数。
    Figure PCTCN2017079426-appb-100012
    表示信号的二次变分;z表示去除基线漂移后的心电信号即所述第一心电信号,D表示二次变分矩阵,I表示对大小的单位矩阵。
  13. 根据权利要求11或12所述的心电信号伪差识别装置,其特征在于,所述第二去除单元具体用于:
    通过如下公式去除所述第一心电信号的肌电干扰以及工频干扰确定所述目标心电信号:
    Figure PCTCN2017079426-appb-100013
    其中,α为自由因子,其值为2.5。γ为阈值,Y为所述目标心电信号;
    通过如下公式计算:
    Figure PCTCN2017079426-appb-100014
    其中,Wi,j为分解后的小波系数,N为采样点数,u为每一层的衰减系数,i为将所述原始信号分解后的层数。
  14. 根据权利要求10所述的心电信号伪差识别装置,其特征在于,所述第二确定模块包括:
    第一确定单元,用于将所述目标心电信号进行样条小波变换确定小波系数;
    第二确定单元,用于对所述小波系数进行处理确定所述目标心电信号中的所有R点。
  15. 根据权利要求14所述的心电信号伪差识别装置,其特征在于,所述第二确定单元具体用于:
    确定所述小波系数中的极大值极小值对;
    根据所述极大值极小值对确定所述目标心电信号中的所有R点。
  16. 根据权利要求10所述的心电信号伪差识别装置,其特征在于,所述识别模块包括:
    第三确定单元,用于以所述目标心电信号中的所有R点位基准确定每个R点的数据段;
    第一判断单元,用于判断每个R点的数据段是否是初次输入;
    第二判断单元,用于在所述每个R点的数据段不是初次输入,则根据DTW算法分别判断所述每个R点的数据段与模板库中的各个模板的相似度是否小于预设的阈值;
    第一标记单元,用于在所述每个R点的数据段中存在有与所述模板库中的模板的相似度小于预设的阈值时,则将所述相似度小于预设的阈值的模板对应的数据段标记为与所述相似度小于预设的阈值的模板为同一类;
    循环单元,用于循环执行所述第一判断单元、所述第二判断单元以及所述标记单元的动作直至所述目标心电信号中的所有R点对应的数据段的标记分类完毕;
    统计单元,用于统计所述目标心电信号中的所有R点对应的数据段标记的各个类别的频数;
    第三判断单元,用于判断所述各个类别的数目是否超过两类;
    第二标记单元,用于在所述各个类别的数目超过两类时,将所述各个类别中频数最少的类别以及所述各个类别中频数第二少的类别标记为所述目标心电信号的伪差。
  17. 根据权利要求16所述的心电信号伪差识别装置,其特征在于,所述识别模块还包括:
    第一处理单元,用于在所述每个R点的数据段是初次输入时,将所述初 次输入的R点对应的数据段作为第一个模板。
  18. 根据权利要求16所述的心电信号伪差识别装置,其特征在于,所述识别模块还包括:
    第四判断单元,用于当所述每个R点的数据段中不存在有与所述模板数据库的模板的相似度大于预设的阈值时,判断所述模板库中的模板的数量是否小于预设值;
    第二处理单元,用于所述模板库中的模板的数量小于预设值时,根据所述相似度大于预设的阈值的模板对应的数据段建立新的模板;
    所述第二处理单元,还用于所述模板库中的模板的数量不小于预设值时,将所述模板库中的模板与所述每个R点的数据段的相似度最小的模板删除,且根据所述相似度大于预设的阈值的模板对应的数据段建立新的模板。
  19. 一种心电信号伪差识别装置,其特征在于,包括:
    中央处理器、存储器、存储介质、电源、无线网络接口以及输入输出接口;
    通过调用所述存储器或存储介质上存储的操作指令,所述中央处理器,用于执行如权利要求1至9中任一项所述的操作。
PCT/CN2017/079426 2017-03-10 2017-04-05 一种心电信号伪差识别方法及心电信号伪差识别装置 WO2018161391A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710144783.4 2017-03-10
CN201710144783.4A CN106923820B (zh) 2017-03-10 2017-03-10 一种心电信号伪差识别方法及心电信号伪差识别装置

Publications (1)

Publication Number Publication Date
WO2018161391A1 true WO2018161391A1 (zh) 2018-09-13

Family

ID=59431954

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/079426 WO2018161391A1 (zh) 2017-03-10 2017-04-05 一种心电信号伪差识别方法及心电信号伪差识别装置

Country Status (2)

Country Link
CN (1) CN106923820B (zh)
WO (1) WO2018161391A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109948396B (zh) * 2017-12-20 2021-07-23 深圳市理邦精密仪器股份有限公司 一种心拍分类方法、心拍分类装置及电子设备
CN110547786B (zh) * 2018-06-01 2020-05-05 山东省计算中心(国家超级计算济南中心) 基于多尺度小波变换的心电图波形的形态识别方法
CN110731762B (zh) * 2019-09-18 2022-02-08 平安科技(深圳)有限公司 基于相似度预处理脉搏波的方法、装置、计算机系统及可读存储介质
CN111685759B (zh) * 2020-05-12 2022-09-09 河北大学 一种心电信号的p、t特征波检测方法
CN113712525A (zh) * 2020-05-21 2021-11-30 深圳市理邦精密仪器股份有限公司 一种生理参数处理方法、装置及医疗设备
US20230017546A1 (en) * 2021-07-19 2023-01-19 GE Precision Healthcare LLC Methods and systems for real-time cycle length determination in electrocardiogram signals

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130190638A1 (en) * 2010-10-12 2013-07-25 Ki H. Chon Motion and noise artifact detection for ecg data
CN103720468A (zh) * 2013-12-05 2014-04-16 深圳先进技术研究院 应用于动态心电数据的伪差识别方法和装置
CN104545887A (zh) * 2014-12-24 2015-04-29 深圳先进技术研究院 伪差心电波形识别方法和装置
CN105741305A (zh) * 2016-03-02 2016-07-06 深圳竹信科技有限公司 一种基于平稳小波变换滤除肌电干扰的方法和系统
CN105748066A (zh) * 2016-03-03 2016-07-13 深圳竹信科技有限公司 一种心电信号波形特征点的提取方法及装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103705234B (zh) * 2013-12-05 2015-08-26 深圳先进技术研究院 动态心电信号数据中的检波方法和装置
CN105726015A (zh) * 2016-01-29 2016-07-06 武汉朗迪远程医疗科技有限公司 一种心电数据智能分析学习方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130190638A1 (en) * 2010-10-12 2013-07-25 Ki H. Chon Motion and noise artifact detection for ecg data
CN103720468A (zh) * 2013-12-05 2014-04-16 深圳先进技术研究院 应用于动态心电数据的伪差识别方法和装置
CN104545887A (zh) * 2014-12-24 2015-04-29 深圳先进技术研究院 伪差心电波形识别方法和装置
CN105741305A (zh) * 2016-03-02 2016-07-06 深圳竹信科技有限公司 一种基于平稳小波变换滤除肌电干扰的方法和系统
CN105748066A (zh) * 2016-03-03 2016-07-13 深圳竹信科技有限公司 一种心电信号波形特征点的提取方法及装置

Also Published As

Publication number Publication date
CN106923820A (zh) 2017-07-07
CN106923820B (zh) 2020-01-17

Similar Documents

Publication Publication Date Title
WO2018161391A1 (zh) 一种心电信号伪差识别方法及心电信号伪差识别装置
US20200335217A1 (en) Methods and systems using mathematical analysis and machine learning to diagnose disease
CN109171712B (zh) 心房颤动识别方法、装置、设备及计算机可读存储介质
CN108186011B (zh) 房颤检测方法、装置及可读存储介质
Safara et al. Multi-level basis selection of wavelet packet decomposition tree for heart sound classification
Elgendi et al. Revisiting QRS detection methodologies for portable, wearable, battery-operated, and wireless ECG systems
JP5271718B2 (ja) 複数の時間区分にわたり胎児及び産婦のecgを特定する方法
CN110141218B (zh) 一种心电信号分类方法、装置及程序产品、存储介质
US20220142583A1 (en) Methods and systems to quantify and remove asynchronous noise in biophysical signals
US11051741B2 (en) Non-invasive detection of coronary heart disease from short single-lead ECG
WO2021253709A1 (zh) 心电图心搏分类方法、装置、电子设备和介质
CN109948396B (zh) 一种心拍分类方法、心拍分类装置及电子设备
US11344246B2 (en) Long QT syndrome diagnosis and classification
Zhu et al. Feature extraction from a novel ECG model for arrhythmia diagnosis
Lee et al. A real-time abnormal beat detection method using a template cluster for the ECG diagnosis of IoT devices
Anto et al. Segmentation of ultrasound images of fetal anatomic structures using random forest for low-cost settings
Keenan et al. Detection of fetal arrhythmias in non-invasive fetal ECG recordings using data-driven entropy profiling
Chin et al. A fast critical arrhythmic ECG waveform identification method using cross-correlation and multiple template matching
JP6435176B2 (ja) 身体情報取得装置、身体情報取得方法、およびプログラム
CN115486849A (zh) 心电信号质量评估方法、装置及设备
Chen et al. An automatic R-peak detection method based on hierarchical clustering
Ródenas et al. Combined nonlinear analysis of atrial and ventricular series for automated screening of atrial fibrillation
JP6103591B2 (ja) 聴診心音信号の処理方法、聴診心音信号の処理装置及び聴診心音信号を処理するためのプログラム
CN111345815B (zh) 一种心电信号中qrs波的检测方法、装置、设备和存储介质
Amiri et al. PhonoSys: mobile phonocardiography diagnostic system for newborns

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17899357

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 22.01.20)

122 Ep: pct application non-entry in european phase

Ref document number: 17899357

Country of ref document: EP

Kind code of ref document: A1