WO2021143401A1 - 一种用于心电信号中p波和t波的检测方法和装置 - Google Patents

一种用于心电信号中p波和t波的检测方法和装置 Download PDF

Info

Publication number
WO2021143401A1
WO2021143401A1 PCT/CN2020/134754 CN2020134754W WO2021143401A1 WO 2021143401 A1 WO2021143401 A1 WO 2021143401A1 CN 2020134754 W CN2020134754 W CN 2020134754W WO 2021143401 A1 WO2021143401 A1 WO 2021143401A1
Authority
WO
WIPO (PCT)
Prior art keywords
wave
signal
point
data
peak
Prior art date
Application number
PCT/CN2020/134754
Other languages
English (en)
French (fr)
Inventor
孙洪岱
曹君
Original Assignee
上海优加利健康管理有限公司
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 上海优加利健康管理有限公司 filed Critical 上海优加利健康管理有限公司
Publication of WO2021143401A1 publication Critical patent/WO2021143401A1/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
    • 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
    • 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
    • A61B5/349Detecting specific parameters of the electrocardiograph cycle
    • 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
    • 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
    • A61B5/7235Details of waveform analysis
    • 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
    • A61B5/7235Details of waveform analysis
    • A61B5/725Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters

Definitions

  • the present invention relates to the technical field of signal processing, in particular to a method and device for detecting P waves and T waves in an electrocardiogram signal.
  • Electrocardiogram is a technology that uses an electrocardiograph to record the changes in the electrical activity patterns of the heart during each cardiac cycle from the body surface, that is, the potential changes in the bioelectric current generated when the heart beats.
  • T wave is a ventricular complex wave, an important feature in the ECG signal, and represents the potential change when the ventricle is repolarized.
  • there are many important indicators such as the measurement of QT interval, etc., which are based on the accurate positioning of T wave characteristic points.
  • the existence of u-waves and multiple frequency band interference it is extremely difficult to accurately locate T-waves. Therefore, the automatic detection of T-waves has always attracted attention.
  • P wave is a new atrial depolarization wave, which is also an important feature of the ECG signal. It represents the potential change of atrial depolarization.
  • the shape of P wave is useful for the diagnosis of ventricular escape, atrial flutter, atrial fibrillation, and blockade. important meaning.
  • the frequency component easily aliasing with other ECG components, and the sensitivity to interference, the detection accuracy of the P wave in the prior art still needs to be improved.
  • the purpose of the present invention is to provide a detection method and device for P wave and T wave in the ECG signal in view of the defects of the prior art. Based on wavelet transform and dynamic threshold method, the peak value of T wave can be located more accurately The position and width improve the accuracy of T-wave detection and at the same time realize the improvement of P-wave detection accuracy.
  • the present invention provides a method for detecting P waves and T waves in an ECG signal, including:
  • Receiving long-range ECG monitoring data to be detected preprocessing the long-range ECG monitoring data to obtain multiple data fragments in the long-range ECG monitoring data; the length of the data fragments is a preset duration;
  • the heartbeat classification information of the QRS complex signal included in each data segment in the long-range ECG monitoring data determine whether the heartbeat type of the QRS complex data in the data segment is a sinus heartbeat
  • the T wave detection includes: performing a first band-pass filtering process on the segment data to be processed using a 4-20 Hz band-pass filter; performing wavelet decomposition on the data after the first band-pass filtering and preserving low-frequency components; wherein the wavelet base Is db2, the number of decomposition layers is 2; the low-frequency components obtained by the wavelet decomposition and the 4-20Hz band-pass filtered data are dot-multiplied to obtain the T-wave characteristic signal; and the peak value of the T-wave characteristic signal is detected ,
  • the position of the peak point is the peak position of the T wave, and the inflection point positions before and after the peak point are the start and end points of the T wave;
  • the P wave detection includes: performing T wave data setting 0 processing on the segment data to be processed containing the P wave signal and the T wave signal according to the obtained starting and ending point positions of the T wave, to obtain the segment data to be processed containing the P wave signal; Use a 7-20 Hz band-pass filter to perform a second band-pass filtering process on the segment data to be processed containing the P-wave signal; combine the data after the second band-pass filtering with the to-be-processed segment data containing the P-wave signal and the T-wave signal
  • the processed segment data is point-multiplied to obtain the P wave characteristic signal; peak detection is performed on the P wave characteristic signal, the position of the peak point is the P wave peak position, and the inflection point positions before and after the peak point are the start and end points of the P wave.
  • the receiving the long-range ECG monitoring data to be detected, and preprocessing the long-range ECG monitoring data to obtain multiple data fragments in the long-range ECG monitoring data specifically include:
  • Receive long-range ECG monitoring data to be detected perform QRS signal detection processing on the long-range ECG monitoring data, and determine the position information of each QRS complex signal in the long-range ECG monitoring data; the position of the QRS complex signal
  • the information includes the starting position and ending position of each QRS wave;
  • the long-range ECG monitoring data after the QRS signal detection and processing is subjected to heartbeat classification processing to obtain the heartbeat classification information of each QRS complex signal, and according to the heartbeat classification information, the long-term ECG monitoring data Mark
  • the filtered long-range ECG monitoring data is intercepted by data fragments in a preset time sequence to obtain a plurality of data fragments with a preset duration; each data fragment includes a plurality of QRS complex signals.
  • the method further includes:
  • performing T wave data zeroing processing on the segment data to be processed containing the P wave signal and the T wave signal specifically includes:
  • the data segment obtained after the signal amplitude between the start and end positions of the T wave is set to 0 is subjected to edge data smoothing through an edge difference algorithm; the start and end positions of the T wave include a start position and an end position.
  • edge data smoothing processing is specifically:
  • the interception of data fragments of the filtered long-range ECG monitoring data according to a preset time length sequence specifically includes:
  • the filtered long-range ECG monitoring data starts from the data starting point, and sequentially intercepts data fragments of a preset duration according to a set step interval; the set step interval is less than the preset duration.
  • the peak detection is performed on the T wave characteristic signal
  • the position of the peak point is the T wave peak position
  • the inflection point positions before and after the peak point are the start and end points of the T wave are specifically:
  • the inflection point positions before and after the peak point are the starting and ending points of the T wave.
  • the determining the first end point and the second end point before and after the T wave according to the set rule specifically includes:
  • the position 0.2s before the start position of the first QRS wave after the T wave and the T wave The midpoint of the position 0.2s behind the wave is the second end point.
  • the peak detection of the P wave characteristic signal, the position of the peak point is the P wave peak position
  • the inflection point positions before and after the peak point are the start and end points of the P wave are specifically:
  • the inflection point positions before and after the peak point are the starting and ending points of the P wave.
  • the determining the third end point and the fourth end point before and after the P wave according to the set rule specifically includes:
  • the position 0.2s before the end position of the first QRS wave of the P wave front and the P wave front is the third end point.
  • the received long-range ECG monitoring data to be detected is dual-lead ECG monitoring data
  • the T-wave data is set to 0 for the segment data to be processed containing the P-wave signal and the T-wave signal .
  • the method further includes:
  • the detection method for P wave and T wave in ECG signal provided by the embodiment of the present invention is based on wavelet transform and dynamic threshold method, which locates the peak position and width of T wave more accurately, and improves the accuracy of T wave detection. At the same time, it realizes the improvement of P wave detection accuracy rate, which provides effective and accurate data guarantee for ECG analysis.
  • an embodiment of the present invention provides a device that includes a memory and a processor, the memory is used to store a program, and the processor is used to execute the first aspect and the methods in each implementation manner of the first aspect.
  • embodiments of the present invention provide a computer program product containing instructions, which when the computer program product runs on a computer, cause the computer to execute the first aspect and the methods in the implementation manners of the first aspect.
  • an embodiment of the present invention provides a computer-readable storage medium, and a computer program is stored on the computer-readable storage medium.
  • the computer program is executed by a processor, the first aspect and the implementation manners of the first aspect are implemented. method.
  • FIG. 1 is a flowchart of a method for detecting P waves and T waves in an ECG signal according to an embodiment of the present invention
  • Fig. 3 is the second schematic diagram of waveforms in the data processing process of the detection method provided by the embodiment of the present invention.
  • FIG. 4 is the third schematic diagram of waveforms in the data processing process of the detection method provided by the embodiment of the present invention.
  • FIG. 5 is a flowchart of a specific method for T wave detection provided by an embodiment of the present invention.
  • FIG. 6 is a schematic diagram of determining the inflection point for T wave detection according to an embodiment of the present invention.
  • FIG. 7 is a flowchart of a specific method for P wave detection provided by an embodiment of the present invention.
  • FIG. 8 is a schematic diagram of determining the inflection point for P wave detection according to an embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of a device provided by an embodiment of the present invention.
  • FIG. 1 is a flow chart of the detection method provided by an embodiment of the present invention.
  • the detection method for P waves and T waves in an ECG signal provided by an embodiment of the present invention will be described below with reference to FIG. 1.
  • Step 110 Receive the long-range ECG monitoring data to be detected, perform QRS signal detection processing on the long-range ECG monitoring data, and determine the position information of each QRS complex signal in the long-range ECG monitoring data;
  • the position information of the QRS complex signal includes the starting position and the ending position of each QRS wave.
  • the length of the ECG signal acquisition time of the long-range ECG monitoring data used is 12 hours or 24 hours.
  • a segment of long-range ECG monitoring data is shown in Figure 2.
  • step 120 The specific execution method of this step can be obtained by the method before step 120 and step 120 in the patent 201711203259.6 "automatic electrocardiogram analysis method and device based on artificial intelligence self-learning" previously applied by the applicant. Filtering is performed before step 120 of "Automatic ECG Analysis Method and Device Based on Artificial Intelligence Self-learning".
  • the long-range ECG monitoring data is subjected to QRS signal detection processing and then the filtering processing is performed.
  • filtering processing can also be performed before the QRS signal detection processing is performed on the long-range ECG monitoring data, which does not affect the implementation of the present invention.
  • Step 120 Perform heartbeat classification processing on the long-range ECG monitoring data after QRS signal detection and processing, to obtain heartbeat classification information of each QRS complex signal;
  • Step 130 Filtering the long-range ECG monitoring data after the heartbeat classification processing to remove the impurity signal of the ECG signal, and output the filtered long-range ECG monitoring data;
  • the impurity signal may include high-frequency impurity signals, such as high-frequency interference, and low-frequency impurity signals, such as baseline drift.
  • high-frequency impurity signals such as high-frequency interference
  • low-frequency impurity signals such as baseline drift.
  • Step 140 intercepting data fragments of the filtered long-range ECG monitoring data according to a preset time length sequence to obtain a plurality of data fragments of a preset time length;
  • each data segment includes multiple QRS complex signals.
  • the filtered long-range ECG monitoring data starts from the data starting point, and sequentially intercepts the data fragments of the preset duration according to the set step interval; wherein, the set step interval Less than the preset duration.
  • the preset duration is 10s
  • the step interval is 9s.
  • every 10s segment will have overlapping parts.
  • the overlap is 1 second. That is, the first fragment is 0-10s, the second fragment is 9-19s, the third fragment is 18-28s, and so on.
  • Step 150 according to the heartbeat classification information of the QRS complex signal included in each data fragment, determine whether the heartbeat type of the QRS complex data in the data fragment is all sinus heartbeat;
  • step 160 If yes, go to step 160, if not, go to step 190 to set all signal amplitudes in a data segment to zero.
  • Step 160 According to the position information of the QRS complex signal, the signal amplitude from the starting position to the ending position of the QRS complex data of each sinus heartbeat is set to 0, and the edge data is smoothed by the edge difference algorithm to obtain Fragment data to be processed containing P wave signal and T wave signal;
  • the edge difference algorithm is used to smooth the edge data in this example.
  • Step 170 Perform T wave detection on the segment data to be processed
  • T wave detection steps are shown in Figure 5, including:
  • Step 171 Perform a first band-pass filtering process on the segment data to be processed using a 4-20 Hz band-pass filter;
  • the components of the T-wave frequency band in the signal are retained, as shown in FIG. 3.
  • the arrow in the figure points to the T wave signal.
  • Step 172 Perform wavelet transformation on the first band-pass filtered data and retain low-frequency components
  • the wavelet base is db2, and the number of decomposition layers is 2;
  • Step 173 Do a dot multiplication of the low-frequency components obtained by wavelet decomposition and the data after 4-20 Hz band-pass filtering to obtain the T-wave characteristic signal;
  • Step 174 Perform peak detection on the T-wave characteristic signal, the location of the peak point is the T-wave peak position, and the inflection point positions before and after the peak point are the starting and ending points of the T-wave.
  • the starting and ending point positions are determined by dynamic threshold.
  • the specific starting and ending point positions that is, the calculation method of the inflection point, can be shown in Fig. 6, taking the T wave peak point position as the first initial point, and determining the before and after T wave according to the set rules The first end point and the second end point;
  • the corresponding assignments of each time point between the two points are fitted, and then the point on the line at the same time and the The amplitude difference of the T wave signal from the first end point to the first initial point, where the position corresponding to the maximum amplitude difference is the inflection point before the peak point;
  • the inflection point before and after the peak point is the starting and ending point of the T wave.
  • the position of the T wave front 0.2s is the first end point
  • the position 0.2s before the start position of the first QRS wave after the T wave is the second end point ;
  • the position 0.2s before the start position of the first QRS wave after the T wave and 0.2s after the T wave The midpoint of the position is the second end point.
  • Step 180 Perform P wave detection on the segment data to be processed
  • the P wave detection steps are shown in Figure 7 and include:
  • Step 181 According to the obtained starting and ending point positions of the T wave, perform T wave data setting 0 processing on the segment data to be processed containing the P wave signal and the T wave signal to obtain the segment data to be processed containing the P wave signal;
  • the signal amplitude between the start and end positions of the T wave is set to 0; the resulting signal waveform is shown in FIG. 4.
  • the edge difference algorithm is used to perform data smoothing on the data segment obtained after the signal amplitude between the start and end positions of the T wave is set to 0.
  • the data smoothing process can be:
  • Step 182 using a 7-20 Hz band pass filter to perform a second band pass filtering process on the segment data to be processed containing the P wave signal;
  • Step 183 Do a dot product of the second band-pass filtered data and the segment data to be processed containing the P wave signal and the T wave signal to obtain the P wave characteristic signal;
  • Step 184 Perform peak detection on the P wave characteristic signal, the position of the peak point is the P wave peak position, and the inflection point positions before and after the peak point are the start and end points of the P wave.
  • the starting and ending point positions are determined by a dynamic threshold method.
  • the specific starting and ending point positions that is, the calculation method of the inflection point, can be specifically shown in Figure 8.
  • the P wave peak point position is the second initial point, and the P wave before and after it is determined according to the set rules.
  • the inflection points before and after the peak point are the starting and ending points of the P wave.
  • the position 0.2s after the P wave is the fourth end point
  • the position 0.2s after the end position of the first QRS wave of the P wave front is regarded as the third end point
  • the position 0.2s before the end position of the first QRS wave of the P wave front and the position 0.2s before the P wave front is the third end point.
  • the dual-lead ECG monitoring data is the long-range ECG monitoring data input
  • a more optimized processing method can be used for the detection of the P wave, that is, when the P wave signal and the T wave signal are to be processed.
  • the processed segment data is processed with T wave data set to 0
  • the data of the two leads with T wave data set to 0 processing are merged and sorted to obtain a superimposed fragment containing P wave signal; then combined with long-range ECG monitoring data for superimposition
  • the fragments undergo P wave position fusion to obtain the to-be-processed fragment data containing the P wave signal, and then the method of steps 182-184 is executed.
  • the merge sort mentioned here is to put the data of the two leads with the T wave data set to 0, that is, the array of the P wave position of the P wave signal is placed in the same array for sorting, and this array contains two Lead P wave position information.
  • the P wave position fusion is to merge two or more P waves within 60ms between two QRS waves into one.
  • the detection method for P wave and T wave in ECG signal provided by the embodiment of the present invention is based on wavelet transform and dynamic threshold method, which locates the peak position and width of T wave more accurately, and improves the accuracy of T wave detection. At the same time, it realizes the improvement of P wave detection accuracy rate, which provides effective and accurate data guarantee for ECG analysis.
  • the embodiment of the present invention also provides a device for implementing the above detection method, which may specifically include a physical device and a virtual device, such as a device, a computer-readable storage medium, or a computer program product.
  • FIG. 9 is a schematic structural diagram of a device provided by an embodiment of the present invention.
  • the device includes a processor and a memory.
  • the memory can be connected to the processor through a bus.
  • the memory may be a non-volatile memory, such as a hard disk drive and a flash memory, and software programs and device drivers are stored in the memory.
  • the software program can execute various functions of the foregoing method provided by the embodiments of the present invention; the device driver may be a network and interface driver.
  • the processor is used to execute a software program, and when the software program is executed, the method provided in the embodiment of the present invention can be implemented.
  • the embodiment of the present invention also provides a computer-readable storage medium.
  • a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method provided in the embodiment of the present invention can be implemented.
  • the embodiment of the present invention also provides a computer program product containing instructions.
  • the processor is caused to execute the above method.
  • the steps of the method or algorithm described in combination with the embodiments disclosed in this document can be implemented by hardware, a software module executed by a processor, or a combination of the two.
  • the software module can be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or all areas in the technical field. Any other known storage media.

Landscapes

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

Abstract

一种用于心电信号中P波和T波的检测方法和装置,方法包括:对接收到的待检测的长程心电监测数据进行QRS信号检测处理,确定长程心电监测数据中各QRS波群信号的位置信息(110);将长程心电监测数据进行心搏分类处理,得到每个QRS波群信号的心搏分类信息(120);对长程心电监测数据进行滤波处理,输出滤波后的长程心电监测数据(130);对滤波后的长程心电监测数据按照预设时长顺序进行数据片段截取,得到多个预设时长的数据片段(140);根据心搏分类信息识别窦性心搏的片段并根据QRS波群信号的位置信息,将每个窦性心搏中QRS波的起始位置到终止位置的信号幅值置为0,得到含有P波信号和T波信号的待处理片段数据,再对待处理片段数据进行T波检测和P波检测。

Description

一种用于心电信号中P波和T波的检测方法和装置
本申请要求于2020年1月17日提交中国专利局、申请号为202010052910.X、发明名称为“一种用于心电信号中P波和T波的检测方法和装置”的中国专利申请的优先权。
技术领域
本发明涉及信号处理技术领域,尤其涉及一种用于心电信号中P波和T波的检测方法和装置。
背景技术
心电图(ECG或者EKG)是利用心电图机从体表记录心脏每一心动周期所产生的电活动变化图形的技术,也就是将心脏搏动时产生的生物电流记录下来的电位变化图。
T波是心室复级波,是心电信号中的一项重要特征,代表了心室复极时的电位改变。在常规心电图分析中,有许多重要指标如QT间期等的测量,都是以T波特征点的准确定位为基础的。但是由于T波心态的多样性,u波以及多种频段干扰的存在,使得T波的准确定位极为困难,所以,一直以来T波的自动检测备受关注。
P波是新房除极波,同样是心电信号的一项重要特征,代表了心房除极的电位变化,P波的形态对于室性逸搏、心房扑动、心房颤动、阻滞的诊断有重要意义。但是由于P波的幅度低,频率成分易与其他心电成分混叠,对干扰敏感等因素,现有技术对P波的检测准确率还有待提高。
发明内容
本发明的目的是针对现有技术的缺陷,提供一种用于心电信号中P波和T波的检测方法和装置,基于小波变换以及动态阈值的方法,较为准确的定位了T波的峰值位置以及宽度,提高了T波检测的准确率,并同时实现了P波检测准确率的提高。
为实现上述目的,在第一方面,本发明提供了一种用于心电信号中P波和T波的检测方法,包括:
接收待检测的长程心电监测数据,对所述长程心电监测数据进行预处理,得到长程心电监测数据中的多个数据片段;所述数据片段的长度为预设时长;
根据长程心电监测数据中的每个数据片段包括的QRS波群信号的心搏分类信息,确定所述数据片段中QRS波群数据的心搏类型是否都是窦性心搏;
如果是,根据所述QRS波群信号的位置信息,将每个窦性心搏中QRS波的起始位置到终止位置的信号幅值置为0,并通过边缘差值算法进行边缘数据平滑处理,得到含有P波信号和T波信号的待处理片段数据;
对所述待处理片段数据进行T波检测和P波检测;
所述T波检测包括:使用4-20Hz带通滤波器对所述待处理片段数据进行第一带通滤波处理;对第一带通滤波后的数据进行小波分解并保留低频成分;其中小波基为db2,分解层数为2层;将所述小波分解的得到的低频成分与4-20Hz带通滤波后的数据进行点乘,得到T波特征信号;对所述T波特征信号进行峰值检测,峰值点的位置为T波波峰位置,峰值点前后的拐点位置为T波的起止点位置;
所述P波检测包括:根据所得的T波的起止点位置,对含有P波信号和T波信号的待处理片段数据进行T波数据置0处理,得到含有P波信号的待处理片段数据;使用7-20Hz带通滤波器对所述含有P波信号的待处理片段数据进行第二带通滤波处理;将第二带通滤波后的数据与所述含有P波信号和T波信号的待处理片段数据进行点乘,得到P波特征信号;对所述P波特征信 号进行峰值检测,峰值点的位置为P波波峰位置,峰值点前后的拐点位置为P波的起止点位置。
优选的,所述接收待检测的长程心电监测数据,对所述长程心电监测数据进行预处理,得到长程心电监测数据中的多个数据片段具体包括:
接收待检测的长程心电监测数据,对所述长程心电监测数据进行QRS信号检测处理,确定所述长程心电监测数据中各QRS波群信号的位置信息;所述QRS波群信号的位置信息包括每个QRS波的起始位置和终止位置;
将所述QRS信号检测处理后的长程心电监测数据进行心搏分类处理,得到每个QRS波群信号的心搏分类信息,并根据所述心搏分类信息,对所述长程心电监测数据进行标注;
对所述心搏分类处理及标注后的长程心电监测数据进行滤波处理,以去除心电信号的杂质信号,输出滤波后的长程心电监测数据;
对所述滤波后的长程心电监测数据按照预设时长顺序进行数据片段截取,得到多个预设时长的所述数据片段;每个数据片段包括多个QRS波群信号。
优选的,在所述确定所述数据片段中QRS波群数据的心搏类型是否都是窦性心搏之后,所述方法还包括:
如果不是,则将所述一个数据片段中全部信号幅值都置为0。
优选的,所述根据所得的T波的起止点位置,对含有P波信号和T波信号的待处理片段数据进行T波数据置0处理具体包括:
将含有P波信号和T波信号的待处理数据片段中,T波的起止点位置之间的信号幅值置为0;
对所述T波的起止点位置之间的信号幅值置为0后得到的数据片段通过边缘差值算法进行边缘数据平滑处理;所述T波的起止点位置包括起始位置和终止位置。
进一步优选的,所述边缘数据平滑处理具体为:
以波的起始位置作为第一边界位置的终点,以第一边界位置向前0.03s的位置为起点,对所述第一边界位置的起点到终点间的信号幅值进行线性差值;以及
以波的终止位置作为第二边界位置的起点,以第二边界位置向后0.03s的位置为终点,对所述第二边界位置的起点到终点间的信号幅值进行线性差值。
优选的,所述对所述滤波后的长程心电监测数据按照预设时长顺序进行数据片段截取具体为:
对所述滤波后的长程心电监测数据从数据起始点开始,按照设定步进间隔,依次顺序截取预设时长的数据片段;所述设定步进间隔小于所述预设时长。
优选的,所述对所述T波特征信号进行峰值检测,峰值点的位置为T波波峰位置,峰值点前后的拐点位置为T波的起止点位置具体为:
以T波峰值点位置为第一初始点,并根据设定规则确定T波前后的第一末端点和第二末端点;
以第一末端点和第一初始点为起止点进行线性拟合处理,得到第一线性拟合信号,计算每一时刻第一线性拟合信号与所述第一初始点到第一末端点间的T波信号的幅值差,确定其中最大幅值差对应的位置为峰值点前的拐点;
以第一初始点和第二末端点为起止点进行线性拟合处理,得到第二线性拟合信号,计算每一时刻第二线性拟合信号与所述第一初始点到第二末端点间的T波信号的幅值差,确定其中最大幅值差对应的位置为峰值点后的拐点;
所述峰值点前后的拐点位置即为T波的起止点位置。
进一步优选的,所述根据设定规则确定T波前后的第一末端点和第二末端点具体包括:
以所述T波前0.2s的位置为所述第一末端点;
如果所述T波峰值点至T波后第一个QRS波的起始位置对应的时间间隔 不小于0.2s,以所述T波后第一个QRS波的起始位置前0.2s的位置为所述第二末端点;
如果所述T波峰值点至T波后第一个QRS波的起始位置对应的时间间隔小于0.2s,以所述T波后第一个QRS波的起始位置前0.2s的位置和T波后0.2s的位置的中点为所述第二末端点。
优选的,所述对所述P波特征信号进行峰值检测,峰值点的位置为P波波峰位置,峰值点前后的拐点位置为P波的起止点位置具体为:
以P波峰值点位置为第二初始点,并根据设定规则确定P波前后的第三末端点和第四末端点;
以第三末端点和第二初始点为起止点进行线性拟合处理,得到第三线性拟合信号,计算每一时刻第三线性拟合信号与所述第二初始点到第三末端点间的T波信号的幅值差,确定其中最大幅值差对应的位置为峰值点前的拐点;
以第二初始点和第四末端点为起止点进行线性拟合处理,得到第四线性拟合信号,计算每一时刻第四线性拟合信号与所述第二初始点到第四末端点间的P波信号的幅值差,确定其中最大幅值差对应的位置为峰值点后的拐点;
所述峰值点前后的拐点位置即为P波的起止点位置。
进一步优选的,所述根据设定规则确定P波前后的第三末端点和第四末端点具体包括:
以所述P波后0.2s的位置为所述第四末端点;
如果所述P波峰值点至P波前第一个QRS波的终止位置对应的时间间隔不小于0.2s,以所述P波前第一个QRS波的终止位置后0.2s的位置为所述第三末端点;
如果所述P波峰值点至P波前第一个QRS波的终止位置对应的时间间隔小于0.2s,以所述P波前第一个QRS波的终止位置前0.2s的位置和P波前0.2s的位置的中点为所述第三末端点。
优选的,当接收到的待检测的长程心电监测数据为双导联心电监测数据 时,在所述对含有P波信号和T波信号的待处理片段数据进行T波数据置0处理之后,所述方法还包括:
将两个导联各自的进行T波数据置0处理的数据进行归并排序得到一个含有P波信号的叠加片段;
结合所述长程心电监测数据对于所述叠加片段进行P波位置融合,将处在两个QRS波之间的处于60ms以内的两个或多个P波融合为一个,得到所述含有P波信号的待处理片段数据。本发明实施例提供的用于心电信号中P波和T波的检测方法,基于小波变换以及动态阈值的方法,较为准确的定位了T波的峰值位置以及宽度,提高了T波检测的准确率,并同时实现了P波检测准确率的提高,对于心电分析提供了有效和准确的数据保障。
第二方面,本发明实施例提供了一种设备,该设备包括存储器和处理器,存储器用于存储程序,处理器用于执行第一方面及第一方面的各实现方式中的方法。
第三方面,本发明实施例提供了一种包含指令的计算机程序产品,当计算机程序产品在计算机上运行时,使得计算机执行第一方面及第一方面的各实现方式中的方法。
第四方面,本发明实施例提供了一种计算机可读存储介质,计算机可读存储介质上存储有计算机程序,计算机程序被处理器执行时实现第一方面及第一方面的各实现方式中的方法。
附图说明
图1为本发明实施例提供的用于心电信号中P波和T波的检测方法流程图;
图2为本发明实施例提供的检测方法的数据处理过程中的波形示意图之一;
图3为本发明实施例提供的检测方法的数据处理过程中的波形示意图之 二;
图4为本发明实施例提供的检测方法的数据处理过程中的波形示意图之三;
图5为本发明实施例提供的针对T波检测的具体方法流程图;
图6为本发明实施例提供的针对T波检测的拐点确定示意图;
图7为本发明实施例提供的针对P波检测的具体方法流程图;
图8为本发明实施例提供的针对P波检测的拐点确定示意图;
图9为本发明实施例提供的一种设备结构示意图。
具体实施方式
下面通过附图和实施例,对本发明的技术方案做进一步的详细描述。
本发明实施例提供的用于心电信号中P波和T波的检测方法流程图,可以用于心电信号的自动分析。图1为本发明实施例提供的检测方法流程图,下面结合图1所述,对本发明实施例提供的用于心电信号中P波和T波的检测方法进行说明。
在本例中,其实施方法步骤如图1所示:
步骤110,接收待检测的长程心电监测数据,对长程心电监测数据进行QRS信号检测处理,确定长程心电监测数据中各QRS波群信号的位置信息;
具体的,QRS波群信号的位置信息包括每个QRS波的起始位置和终止位置。所用长程心电监测数据的心电信号采集时间长度为12小时或者24小时。长程心电监测数据的一段数据如图2所示。
本步骤的具体执行方法可以采用申请人在先申请的专利201711203259.6《基于人工智能自学习的心电图自动分析方法和装置》中步骤120及步骤120之前的方法获得。在《基于人工智能自学习的心电图自动分析方法和装置》的步骤120之前先进行了滤波处理,在本例中采用对长程心电监测数据进行QRS信号检测处理之后再进行滤波处理的方式。当然也可以在对长程心电监测 数据进行QRS信号检测处理之前进行滤波处理,这都不影响本发明的实现。
步骤120,将QRS信号检测处理后的长程心电监测数据进行心搏分类处理,得到每个QRS波群信号的心搏分类信息;
具体的,心搏分类处理的过程,可以参考申请人在先专利201711203546.7《基于人工智能的心电图心搏自动识别分类方法》,通过该专利记载的方法实现,在此不再着重说明。
步骤130,对心搏分类处理后的长程心电监测数据进行滤波处理,以去除心电信号的杂质信号,输出滤波后的长程心电监测数据;
具体的,杂质信号可以包括高频杂质信号—例如高频干扰,以及低频杂质信号—例如基线漂移。可以先通过0.5Hz高通滤波器去除心电信号的基线漂移,然后再通过30Hz低通滤波器去除心电信号频率之外的高频干扰,输出用于后续处理的滤波后的长程心电监测数据。
步骤140,对滤波后的长程心电监测数据按照预设时长顺序进行数据片段截取,得到多个预设时长的数据片段;
其中,每个数据片段包括多个QRS波群信号。
优选的,为了更加合理的处理边界情况,对滤波后的长程心电监测数据从数据起始点开始,按照设定步进间隔,依次顺序截取预设时长的数据片段;其中,设定步进间隔小于预设时长。
例如预设时长选取10s,设定步进间隔为9s。也就是说,每个10s的片段都会有重叠部分。比如重叠为1秒。即第一个片段为0-10s,第二个片段为9-19s,第三个片段为18-28s,以此类推。
步骤150,根据每个数据片段包括的QRS波群信号的心搏分类信息,确定数据片段中QRS波群数据的心搏类型是否都是窦性心搏;
如果是,执行步骤160,如果不是,则执行步骤190,将一个数据片段中全部信号幅值都置为0。
步骤160,根据QRS波群信号的位置信息,将每个窦性心搏的QRS波群数 据的起始位置到终止位置的信号幅值置为0,并通过边缘差值算法进行边缘数据平滑处理,得到含有P波信号和T波信号的待处理片段数据;
也就是说,判断每个10s片段中的QRS波的类型,将心搏类型不是窦性心搏的心电信号片段全部置零,也就是只要10s片段中存在任一不是窦性的心搏,就将整个将片段的心搏信号的幅值置为0,只保留全部为窦性心搏的心电信号片段。
为了避免出现阶跃信号造成后续滤波出现较大的数据扰动,在本例中通过边缘差值算法进行边缘数据平滑处理。
以QRS波的起始位置作为一个边界位置的终点,以该边界位置向前0.03s的位置为起点,对该边界位置的起点到终点间的信号幅值进行线性差值;以及
以QRS波的终止位置作为另一个边界位置的起点,以该边界位置向后0.03s的位置为终点,对该边界位置的起点到终点间的信号幅值进行线性差值。
这样就避免了在边界点处出现前后两个点的斜率发生突变(阶跃)引起后续滤波后的信号不稳定。
步骤170,对待处理片段数据进行T波检测;
具体的,T波检测步骤如图5所示,包括:
步骤171,使用4-20Hz带通滤波器对待处理片段数据进行第一带通滤波处理;
具体的,第一带通滤波处理后,保留了信号中T波频段的成分,如图3所示。图中箭头所指为T波信号。
步骤172,对第一带通滤波后的数据进行小波分解(wavelet transform)并保留低频成分;
其中,小波基为db2,分解层数为2层;
步骤173,将小波分解的得到的低频成分与4-20Hz带通滤波后的数据进 行点乘,得到T波特征信号;
步骤174,对T波特征信号进行峰值检测,峰值点的位置为T波波峰位置,峰值点前后的拐点位置为T波的起止点位置。
起止点位置的确定采用动态阈值的方式,具体的起止点位置即拐点计算的方式具体可以如图6所示,以T波峰值点位置为第一初始点,并根据设定规则确定T波前后的第一末端点和第二末端点;
以第一末端点和第一初始点为起止点进行线性拟合处理,得到线性拟合信号,计算每一时刻线性拟合信号与第一初始点到第一末端点间的T波信号的幅值差,确定其中最大幅值差对应的位置为峰值点前的拐点;其中,进行线性拟合处理的过程可以看作是形成图中第一末端点到第一初始点的连线的过程,根据第一末端点到第一初始点间的时间间隔以及两点的幅值按照一维线性关系拟合成两点间的各个时间点对应的赋值,然后计算同一时刻该连线上一个点与第一末端点到第一初始点间的T波信号的幅值差,其中最大幅值差对应的位置为峰值点前的拐点;
以第一初始点和第二末端点为起止点进行线性拟合处理,得到第二线性拟合信号,计算每一时刻第二线性拟合信号与第一初始点到第二末端点间的T波信号的幅值差,确定其中最大幅值差对应的位置为峰值点后的拐点;其中,进行线性拟合处理的过程可以看作是形成图中第一初始点到第二末端点的连线的过程,方法同上所述,然后计算同一时刻该连线上一个点与第一初始点到第二末端点间的T波信号的幅值差,其中最大幅值差对应的位置为峰值点后的拐点;
峰值点前后的拐点位置即为T波的起止点位置。
优选的,以T波前0.2s的位置为第一末端点;
如果T波峰值点至T波后第一个QRS波的起始位置对应的时间间隔不小于0.2s,以T波后第一个QRS波的起始位置前0.2s的位置为第二末端点;
如果T波峰值点至T波后第一个QRS波的起始位置对应的时间间隔小于 0.2s,以T波后第一个QRS波的起始位置前0.2s的位置和T波后0.2s的位置的中点为所述第二末端点。
步骤180,对待处理片段数据进行P波检测;
具体的,P波检测步骤如图7所示,包括:
步骤181,根据所得的T波的起止点位置,对含有P波信号和T波信号的待处理片段数据进行T波数据置0处理,得到含有P波信号的待处理片段数据;
具体的,将含有P波信号和T波信号的待处理数据片段中,T波的起止点位置之间的信号幅值置为0;所得信号波形如图4所示。
采用边缘差值算法对T波的起止点位置之间的信号幅值置为0后得到的数据片段进行数据平滑处理。
具体的,数据平滑处理可以是:
以T波的起始位置作为第一边界位置的终点,以第一边界位置向前0.03s的位置为起点,对第一边界位置的起点到终点间的信号幅值进行线性差值;以及
以T波的终止位置作为第二边界位置的起点,以第二边界位置向后0.03s的位置为终点,对第二边界位置的起点到终点间的信号幅值进行线性差值。
步骤182,使用7-20Hz带通滤波器对含有P波信号的待处理片段数据进行第二带通滤波处理;
步骤183,将第二带通滤波后的数据与含有P波信号和T波信号的待处理片段数据进行点乘,得到P波特征信号;
步骤184,对P波特征信号进行峰值检测,峰值点的位置为P波波峰位置,峰值点前后的拐点位置为P波的起止点位置。
起止点位置的确定采用动态阈值的方式,具体的起止点位置即拐点计算的方式具体可以如图8所示,以P波峰值点位置为第二初始点,并根据设定规则确定P波前后的第三末端点和第四末端点;
以第三末端点和第二初始点为起止点进行线性拟合处理,得到第三线性拟合信号,计算每一时刻第三线性拟合信号与第三末端点到第二初始点间的T波信号的幅值差,确定其中最大幅值差对应的位置为峰值点后的拐点;其中,进行线性拟合处理的过程可以看作是形成图中第三末端点到第二初始点的连线的过程,方法同上所述,然后计算同一时刻该连线上一个点与第二初始点到第三末端点间的T波信号的幅值差,其中最大幅值差对应的位置为峰值点前的拐点;
以第二初始点和第四末端点为起止点进行线性拟合处理,得到第四线性拟合信号,计算每一时刻第四线性拟合信号与第二初始点到第四末端点间的T波信号的幅值差,确定其中最大幅值差对应的位置为峰值点后的拐点;其中,进行线性拟合处理的过程可以看作是形成图中第二初始点到第四末端点的连线的过程,方法同上所述,然后计算同一时刻该连线上一个点到第四末端点间的P波信号的幅值差,其中最大幅值差对应的位置为峰值点后的拐点;
峰值点前后的拐点位置即为P波的起止点位置。
优选的,以P波后0.2s的位置为第四末端点;
如果P波峰值点至P波前第一个QRS波的终止位置对应的时间间隔不小于0.2s,以P波前第一个QRS波的终止位置后0.2s的位置为第三末端点;
如果P波峰值点至P波前第一个QRS波的终止位置对应的时间间隔小于0.2s,以P波前第一个QRS波的终止位置前0.2s的位置和P波前0.2s的位置的中点为第三末端点。
此外,对于双导联的心电监测数据为输入的长程心电监测数据的情况,对于P波的检测可以进一步的采用更加优化的处理方式,即在对含有P波信号和T波信号的待处理片段数据进行T波数据置0处理之后,将两个导联各自的进行T波数据置0处理的数据进行归并排序得到一个含有P波信号的叠加片段;然后结合长程心电监测数据对于叠加片段进行P波位置融合,得到所述含有P波信号的待处理片段数据,再执行步骤182-184的方法。
这里所说的归并排序,就是将两个导联各自的进行T波数据置0处理的数据,即P波信号的P波位置的数组放到同一个数组中进行排序,这个数组中包含两个导联的P波位置信息。
而P波位置融合,是将处在两个QRS波之间的处于60ms以内的两个或多个P波融合为一个,
通过这个过程,实现了双导联情况下P波的优化定位,使得P波的检测更加准确。
本发明实施例提供的用于心电信号中P波和T波的检测方法,基于小波变换以及动态阈值的方法,较为准确的定位了T波的峰值位置以及宽度,提高了T波检测的准确率,并同时实现了P波检测准确率的提高,对于心电分析提供了有效和准确的数据保障。
本发明实施例还提供了用以实现以上检测方法的装置,具体可以包括实体装置和虚拟装置,如设备、计算机可读存储介质或计算机程序产品。
图9为本发明实施例提供的一种设备结构示意图,该设备包括:处理器和存储器。存储器可通过总线与处理器连接。存储器可以是非易失存储器,例如硬盘驱动器和闪存,存储器中存储有软件程序和设备驱动程序。软件程序能够执行本发明实施例提供的上述方法的各种功能;设备驱动程序可以是网络和接口驱动程序。处理器用于执行软件程序,该软件程序被执行时,能够实现本发明实施例提供的方法。
需要说明的是,本发明实施例还提供了一种计算机可读存储介质。该计算机可读存储介质上存储有计算机程序,该计算机程序被处理器执行时,能够实现本发明实施例提供的方法。
本发明实施例还提供了一种包含指令的计算机程序产品。当该计算机程序产品在计算机上运行时,使得处理器执行上述方法。
专业人员应该还可以进一步意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、计算机软件或者二者的结合来 实现,为了清楚地说明硬件和软件的可互换性,在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本发明的范围。
结合本文中所公开的实施例描述的方法或算法的步骤可以用硬件、处理器执行的软件模块,或者二者的结合来实施。软件模块可以置于随机存储器(RAM)、内存、只读存储器(ROM)、电可编程ROM、电可擦除可编程ROM、寄存器、硬盘、可移动磁盘、CD-ROM、或技术领域内所公知的任意其它形式的存储介质中。
以上所述的具体实施方式,对本发明的目的、技术方案和有益效果进行了进一步详细说明,所应理解的是,以上所述仅为本发明的具体实施方式而已,并不用于限定本发明的保护范围,凡在本发明的精神和原则之内,所做的任何修改、等同替换、改进等,均应包含在本发明的保护范围之内。

Claims (14)

  1. 一种用于心电信号中P波和T波的检测方法,其特征在于,所述检测方法包括:
    接收待检测的长程心电监测数据,对所述长程心电监测数据进行预处理,得到长程心电监测数据中的多个数据片段;所述数据片段的长度为预设时长;
    根据长程心电监测数据中的每个数据片段包括的QRS波群信号的心搏分类信息,确定所述数据片段中QRS波群数据的心搏类型是否都是窦性心搏;
    如果是,根据所述QRS波群信号的位置信息,将每个窦性心搏中QRS波的起始位置到终止位置的信号幅值置为0,并通过边缘差值算法进行边缘数据平滑处理,得到含有P波信号和T波信号的待处理片段数据;
    对所述待处理片段数据进行T波检测和P波检测;
    所述T波检测包括:使用4-20Hz带通滤波器对所述待处理片段数据进行第一带通滤波处理;对第一带通滤波后的数据进行小波分解并保留低频成分;其中小波基为db2,分解层数为2层;将所述小波分解的得到的低频成分与4-20Hz带通滤波后的数据进行点乘,得到T波特征信号;对所述T波特征信号进行峰值检测,峰值点的位置为T波波峰位置,峰值点前后的拐点位置为T波的起止点位置;
    所述P波检测包括:根据所得的T波的起止点位置,对含有P波信号和T波信号的待处理片段数据进行T波数据置0处理,得到含有P波信号的待处理片段数据;使用7-20Hz带通滤波器对所述含有P波信号的待处理片段数据进行第二带通滤波处理;将第二带通滤波后的数据与所述含有P波信号和T波信号的待处理片段数据进行点乘,得到P波特征信号;对所述P波特征信号进行峰值检测,峰值点的位置为P波波峰位置,峰值点前后的拐点位置为P波的起止点位置。
  2. 根据权利要求1所述的检测方法,其特征在于,所述接收待检测的长程心电监测数据,对所述长程心电监测数据进行预处理,得到长程心电监测 数据中的多个数据片段具体包括:
    接收待检测的长程心电监测数据,对所述长程心电监测数据进行QRS信号检测处理,确定所述长程心电监测数据中各QRS波群信号的位置信息;所述QRS波群信号的位置信息包括每个QRS波的起始位置和终止位置;
    将所述QRS信号检测处理后的长程心电监测数据进行心搏分类处理,得到每个QRS波群信号的心搏分类信息,并根据所述心搏分类信息,对所述长程心电监测数据进行标注;
    对所述心搏分类处理及标注后的长程心电监测数据进行滤波处理,以去除心电信号的杂质信号,输出滤波后的长程心电监测数据;
    对所述滤波后的长程心电监测数据按照预设时长顺序进行数据片段截取,得到多个预设时长的所述数据片段;每个数据片段包括多个QRS波群信号。
  3. 根据权利要求1所述的检测方法,其特征在于,在所述确定所述数据片段中QRS波群数据的心搏类型是否都是窦性心搏之后,所述方法还包括:
    如果不是,则将所述一个数据片段中全部信号幅值都置为0。
  4. 根据权利要求1所述的检测方法,其特征在于,所述根据所得的T波的起止点位置,对含有P波信号和T波信号的待处理片段数据进行T波数据置0处理具体包括:
    将含有P波信号和T波信号的待处理数据片段中,T波的起止点位置之间的信号幅值置为0;
    对所述T波的起止点位置之间的信号幅值置为0后得到的数据片段通过边缘差值算法进行边缘数据平滑处理;所述T波的起止点位置包括起始位置和终止位置。
  5. 根据权利要求4所述的检测方法,其特征在于,所述边缘数据平滑处理具体为:
    以波的起始位置作为第一边界位置的终点,以第一边界位置向前0.03s 的位置为起点,对所述第一边界位置的起点到终点间的信号幅值进行线性差值;以及
    以波的终止位置作为第二边界位置的起点,以第二边界位置向后0.03s的位置为终点,对所述第二边界位置的起点到终点间的信号幅值进行线性差值。
  6. 根据权利要求1所述的检测方法,其特征在于,所述对所述滤波后的长程心电监测数据按照预设时长顺序进行数据片段截取具体为:
    对所述滤波后的长程心电监测数据从数据起始点开始,按照设定步进间隔,依次顺序截取预设时长的数据片段;所述设定步进间隔小于所述预设时长。
  7. 根据权利要求1所述的检测方法,其特征在于,所述对所述T波特征信号进行峰值检测,峰值点的位置为T波波峰位置,峰值点前后的拐点位置为T波的起止点位置具体为:
    以T波峰值点位置为第一初始点,并根据设定规则确定T波前后的第一末端点和第二末端点;
    以第一末端点和第一初始点为起止点进行线性拟合处理,得到第一线性拟合信号,计算每一时刻第一线性拟合信号与所述第一初始点到第一末端点间的T波信号的幅值差,确定其中最大幅值差对应的位置为峰值点前的拐点;
    以第一初始点和第二末端点为起止点进行线性拟合处理,得到第二线性拟合信号,计算每一时刻第二线性拟合信号与所述第一初始点到第二末端点间的T波信号的幅值差,确定其中最大幅值差对应的位置为峰值点后的拐点;
    所述峰值点前后的拐点位置即为T波的起止点位置。
  8. 根据权利要求7所述的检测方法,其特征在于,所述根据设定规则确定T波前后的第一末端点和第二末端点具体包括:
    以所述T波前0.2s的位置为所述第一末端点;
    如果所述T波峰值点至T波后第一个QRS波的起始位置对应的时间间隔 不小于0.2s,以所述T波后第一个QRS波的起始位置前0.2s的位置为所述第二末端点;
    如果所述T波峰值点至T波后第一个QRS波的起始位置对应的时间间隔小于0.2s,以所述T波后第一个QRS波的起始位置前0.2s的位置和T波后0.2s的位置的中点为所述第二末端点。
  9. 根据权利要求1所述的检测方法,其特征在于,所述对所述P波特征信号进行峰值检测,峰值点的位置为P波波峰位置,峰值点前后的拐点位置为P波的起止点位置具体为:
    以P波峰值点位置为第二初始点,并根据设定规则确定P波前后的第三末端点和第四末端点;
    以第三末端点和第二初始点为起止点进行线性拟合处理,得到第三线性拟合信号,计算每一时刻第三线性拟合信号与所述第二初始点到第三末端点间的T波信号的幅值差,确定其中最大幅值差对应的位置为峰值点前的拐点;
    以第二初始点和第四末端点为起止点进行线性拟合处理,得到第四线性拟合信号,计算每一时刻第四线性拟合信号与所述第二初始点到第四末端点间的P波信号的幅值差,确定其中最大幅值差对应的位置为峰值点后的拐点;
    所述峰值点前后的拐点位置即为P波的起止点位置。
  10. 根据权利要求9所述的检测方法,其特征在于,所述根据设定规则确定P波前后的第三末端点和第四末端点具体包括:
    以所述P波后0.2s的位置为所述第四末端点;
    如果所述P波峰值点至P波前第一个QRS波的终止位置对应的时间间隔不小于0.2s,以所述P波前第一个QRS波的终止位置后0.2s的位置为所述第三末端点;
    如果所述P波峰值点至P波前第一个QRS波的终止位置对应的时间间隔小于0.2s,以所述P波前第一个QRS波的终止位置前0.2s的位置和P波前0.2s的位置的中点为所述第三末端点。
  11. 根据权利要求1所述的检测方法,其特征在于,当接收到的待检测的长程心电监测数据为双导联心电监测数据时,在所述对含有P波信号和T波信号的待处理片段数据进行T波数据置0处理之后,所述方法还包括:
    将两个导联各自的进行T波数据置0处理的数据进行归并排序得到一个含有P波信号的叠加片段;
    结合所述长程心电监测数据对于所述叠加片段进行P波位置融合,将处在两个QRS波之间的处于60ms以内的两个或多个P波融合为一个,得到所述含有P波信号的待处理片段数据。
  12. 一种设备,包括存储器和处理器,其特征在于,所述存储器用于存储程序,所述处理器用于执行权利要求1至11任一项所述的方法。
  13. 一种包含指令的计算机程序产品,当其在计算机上运行时,使得计算机执行权利要求1至11任一项所述的方法。
  14. 一种计算机可读存储介质,包括指令,当所述指令在计算机上运行时,使所述计算机执行权利要求1至11任一项所述的方法。
PCT/CN2020/134754 2020-01-17 2020-12-09 一种用于心电信号中p波和t波的检测方法和装置 WO2021143401A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010052910.X 2020-01-17
CN202010052910.XA CN111134659B (zh) 2020-01-17 2020-01-17 一种用于心电信号中p波和t波的检测方法和装置

Publications (1)

Publication Number Publication Date
WO2021143401A1 true WO2021143401A1 (zh) 2021-07-22

Family

ID=70525704

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/134754 WO2021143401A1 (zh) 2020-01-17 2020-12-09 一种用于心电信号中p波和t波的检测方法和装置

Country Status (2)

Country Link
CN (1) CN111134659B (zh)
WO (1) WO2021143401A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113749665A (zh) * 2021-08-19 2021-12-07 深圳邦健生物医疗设备股份有限公司 异常指标的捕捉方法、装置、设备和介质
CN113974648A (zh) * 2021-11-01 2022-01-28 山东众阳健康科技集团有限公司 一种基于PQRST波位置识别的RonT信号检测方法
CN114073534A (zh) * 2021-11-25 2022-02-22 成都泰盟软件有限公司 一种心功能分析算法

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111134659B (zh) * 2020-01-17 2022-04-26 上海乐普云智科技股份有限公司 一种用于心电信号中p波和t波的检测方法和装置
CN114533082B (zh) * 2020-11-26 2023-07-14 深圳市科瑞康实业有限公司 一种基于心搏间期数据对qrs波类型进行标记的方法
CN113288169B (zh) * 2021-05-26 2024-05-17 东软集团股份有限公司 一种心电波形信号的波形识别方法、装置及设备
CN113440145B (zh) * 2021-06-28 2022-04-08 西北大学 一种心电信号t波及其终点自动检测方法
CN116570295B (zh) * 2023-07-14 2024-04-30 浙江好络维医疗技术有限公司 一种心电图低电压t波终点定位方法

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5758654A (en) * 1996-08-29 1998-06-02 Harley Street Software Ltd. ECG P QRS T onset and peak detection method
US5778881A (en) * 1996-12-04 1998-07-14 Medtronic, Inc. Method and apparatus for discriminating P and R waves
KR20060117545A (ko) * 2005-05-11 2006-11-17 인하대학교 산학협력단 심전도 신호의 특징추출방법
CN104323771A (zh) * 2014-11-11 2015-02-04 北京海思敏医疗技术有限公司 检测心电图信号中p波、t波的方法和装置
CN105030228A (zh) * 2015-06-29 2015-11-11 深圳市理邦精密仪器股份有限公司 在心电信号中确定其p波位置的方法及装置
CN107303177A (zh) * 2016-04-15 2017-10-31 中国科学院微电子研究所 一种心电图t波、p波的检测方法及系统
CN108294745A (zh) * 2018-03-07 2018-07-20 武汉大学 多导联心电图信号中p波、t波起止点检测方法及系统
CN110013247A (zh) * 2019-05-24 2019-07-16 东北大学 一种心电图p波和t波的检测、区分与定位方法
CN111134659A (zh) * 2020-01-17 2020-05-12 乐普(北京)医疗器械股份有限公司 一种用于心电信号中p波和t波的检测方法和装置

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1084044A (zh) * 1992-09-11 1994-03-23 中日友好医院 三维高频心电信号分析系统及方法
US7174204B2 (en) * 2004-03-30 2007-02-06 Cardiac Science Corporation Methods for quantifying the morphology and amplitude of cardiac action potential alternans
CN1582844A (zh) * 2004-06-03 2005-02-23 上海交通大学 缺血缺氧心率变异信号处理方法
CN100536765C (zh) * 2007-07-13 2009-09-09 天津大学 基于进化型小波域维纳去卷积的心电特征提取方法
CN102485172B (zh) * 2010-12-01 2015-02-25 通用电气公司 T波的峰值点检测方法和系统
EP2499971B1 (fr) * 2011-03-18 2015-12-30 Sorin CRM SAS Dispositif de traitement adaptatif d'un signal d'accélération endocardiaque
CN102579039B (zh) * 2012-03-13 2014-09-03 广东工业大学 一种心电图中的t波交替检测的方法
CN103860162B (zh) * 2012-12-11 2018-02-23 深圳迈瑞生物医疗电子股份有限公司 一种起搏信号检测方法、系统和心电检测设备
US9597513B2 (en) * 2014-10-24 2017-03-21 Medtronic, Inc. Sensing and atrial-synchronized ventricular pacing in an intracardiac pacemaker
EP3114991A1 (en) * 2015-07-10 2017-01-11 Alekseev Alekseev Siegle Röder GbR Data processing apparatus for myocardium condition assessment
CN109381181B (zh) * 2017-08-14 2021-03-30 深圳大学 心电信号特征波形的端点检测方法
CN109431488B (zh) * 2018-09-28 2021-03-19 上海优加利健康管理有限公司 一种心电图动态实时分析数据中典型数据的识别方法
CN109350037A (zh) * 2018-10-26 2019-02-19 安徽智云医疗科技有限公司 一种从10秒心电图中识别房颤和房性早搏的方法
CN109770920A (zh) * 2019-01-31 2019-05-21 东南大学 基于穿戴式心电信号的紧张情绪判别方法及其系统
CN109998529B (zh) * 2019-04-15 2021-11-05 北京工业大学 一种基于高斯函数拟合的ecg信号中p波、t波检测方法
CN110477906B (zh) * 2019-08-26 2021-11-26 电子科技大学 一种心电信号qrs波起止点定位方法

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5758654A (en) * 1996-08-29 1998-06-02 Harley Street Software Ltd. ECG P QRS T onset and peak detection method
US5778881A (en) * 1996-12-04 1998-07-14 Medtronic, Inc. Method and apparatus for discriminating P and R waves
KR20060117545A (ko) * 2005-05-11 2006-11-17 인하대학교 산학협력단 심전도 신호의 특징추출방법
CN104323771A (zh) * 2014-11-11 2015-02-04 北京海思敏医疗技术有限公司 检测心电图信号中p波、t波的方法和装置
CN105030228A (zh) * 2015-06-29 2015-11-11 深圳市理邦精密仪器股份有限公司 在心电信号中确定其p波位置的方法及装置
CN107303177A (zh) * 2016-04-15 2017-10-31 中国科学院微电子研究所 一种心电图t波、p波的检测方法及系统
CN108294745A (zh) * 2018-03-07 2018-07-20 武汉大学 多导联心电图信号中p波、t波起止点检测方法及系统
CN110013247A (zh) * 2019-05-24 2019-07-16 东北大学 一种心电图p波和t波的检测、区分与定位方法
CN111134659A (zh) * 2020-01-17 2020-05-12 乐普(北京)医疗器械股份有限公司 一种用于心电信号中p波和t波的检测方法和装置

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113749665A (zh) * 2021-08-19 2021-12-07 深圳邦健生物医疗设备股份有限公司 异常指标的捕捉方法、装置、设备和介质
CN113749665B (zh) * 2021-08-19 2024-02-02 深圳邦健生物医疗设备股份有限公司 异常指标的捕捉方法、装置、设备和介质
CN113974648A (zh) * 2021-11-01 2022-01-28 山东众阳健康科技集团有限公司 一种基于PQRST波位置识别的RonT信号检测方法
CN113974648B (zh) * 2021-11-01 2023-08-29 众阳健康科技集团有限公司 一种基于PQRST波位置识别的RonT信号检测方法
CN114073534A (zh) * 2021-11-25 2022-02-22 成都泰盟软件有限公司 一种心功能分析算法
CN114073534B (zh) * 2021-11-25 2023-05-26 成都泰盟软件有限公司 一种心功能分析算法

Also Published As

Publication number Publication date
CN111134659B (zh) 2022-04-26
CN111134659A (zh) 2020-05-12

Similar Documents

Publication Publication Date Title
WO2021143401A1 (zh) 一种用于心电信号中p波和t波的检测方法和装置
CA2979135C (en) Systems, apparatus and methods for sensing fetal activity
US7412282B2 (en) Algorithms for detecting cardiac arrhythmia and methods and apparatuses utilizing the algorithms
JP2019055173A (ja) 単極誘導心電図(ecg)信号中の律動を識別するためのカスケード2値クラシファイア
EP1808125A1 (en) Electrophysiological system for analysing an intracardiac electrocardiogram
WO2021169296A1 (zh) 心电图数据处理的方法、装置、计算机设备及存储介质
Houben et al. Analysis of fractionated atrial fibrillation electrograms by wavelet decomposition
EP3641643B1 (en) System for detecting atrial fibrillation
CN111134658B (zh) 一种心电信号中RonT类型心搏的检测方法和装置
Vijaya et al. Arrhythmia detection through ECG feature extraction using wavelet analysis
CN107622259B (zh) 一种t波检测方法、心电数据分析方法及装置
CN110448289B (zh) 一种心率变异性(hrv)分析方法、装置、存储介质及设备
Sanamdikar et al. A literature review on arrhythmia analysis of ECG signal
Chouhan et al. Delineation of QRS-complex, P and T-wave in 12-lead ECG
Lai et al. A real-time QRS complex detection algorithm based on differential threshold method
Mykoliuk et al. Machine learning methods in ECG classification
CN111528833B (zh) 一种心电信号的快速识别与处理方法及系统
Gajare et al. MATLAB-based ECG R-peak Detection and Signal Classification using Deep Learning Approach
Turnip et al. Extraction of P and T waves from electrocardiogram signals with modified Hamilton algorithm
EP3395244A1 (en) Ecg machine including filter for feature detection
Jindal et al. MATLAB based GUI for ECG arrhythmia detection using Pan-Tompkin algorithm
Gliner et al. Identification of features for machine learning analysis for automatic arrhythmogenic event classification
Borodin et al. Overview of algorithms for electrocardiograms analysis
Govrin et al. Cross-correlation technique for arrythmia detection using PR and PP intervals
Zhao et al. A multilead fusion based QRS complex detection method on 12-lead electrocardiogram signals

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: 20913172

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20913172

Country of ref document: EP

Kind code of ref document: A1