A fetal electrocardiogram monitoring method and system based on marker interference signal

Abstract

The present invention discloses a fetal electrocardiogram monitoring method and system based on a marked interference signal. After a mixed maternal-fetal electrocardiogram signal is collected, the mixed maternal-fetal electrocardiogram signal is filtered and segmented, and then it is detected whether there are first to fourth category abnormalities. R wave recognition is performed on the segmented mixed maternal-fetal electrocardiogram signal to detect whether there is a fifth category abnormality. After a target signal containing a fetal heart rate signal is extracted, a fetal heart rate curve is drawn according to the R wave recognition result of the target signal. According to the first to fifth category abnormality detection results, the data segment marked as abnormal is not displayed in the fetal heart rate curve, or the data segment marked as abnormal is displayed in the fetal heart rate curve in a visual manner of different colors or different transparency color blocks, so that users can be prevented from clinically interpreting unreliable monitoring results, thereby improving the accuracy and reliability of fetal electrocardiogram monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of fetal electrocardiogram monitoring, and in particular to a fetal electrocardiogram monitoring method and system based on a marker interference signal. Background Art

[0002] The non-invasive fetal electrocardiogram (NI-FECG) device can collect maternal and fetal electrocardiogram signals and uterine myoelectric signals through electrodes placed on the pregnant woman's abdomen, and use them to calculate the fetal heart rate and uterine contractions. The device can be used for fetal monitoring and timely detection of fetal intrauterine distress.

[0003] In non-invasive fetal ECG monitoring equipment, pregnant women need to wear the equipment for a long time for long-term monitoring. During the monitoring process, the normal behavior of pregnant women cannot be restricted. The behavior of pregnant women will make the monitoring signal abnormal, and there may be other external interference factors during the monitoring process that cause signal abnormality. In the case where the monitored signal is abnormal due to interference, no matter what separation and detection algorithm is used, the final result will deviate from the actual situation, affecting the reliability of the monitoring result. Therefore, it is a technical problem that technicians in this field need to solve urgently to identify the signal under interference and prevent the unreliable monitoring results from being clinically interpreted to improve the accuracy of fetal ECG monitoring results. Summary of the invention

[0004] The present invention provides a fetal electrocardiogram monitoring method and system based on a marker interference signal, which is used to solve the technical problem that during the fetal electrocardiogram monitoring process, the monitoring signal is abnormal due to interference, resulting in unreliable fetal electrocardiogram monitoring results.

[0005] In view of this, the first aspect of the present invention provides a fetal electrocardiogram monitoring method based on a marker interference signal, comprising:

[0006] The mixed maternal and fetal ECG signals of pregnant women are collected by non-invasive fetal ECG monitoring equipment;

[0007] Performing filtering and segmentation processing on the mixed maternal-fetal electrocardiogram signal to obtain a segmented mixed maternal-fetal electrocardiogram signal;

[0008] Detect whether each segment of the mixed maternal-fetal electrocardiogram signal has the first type of abnormality, the second type of abnormality, the third type of abnormality and the fourth type of abnormality, and mark the mixed maternal-fetal electrocardiogram signal data segment with abnormalities, wherein the first type of abnormality is the interface state abnormality, the second type of abnormality is the abdominal muscle electrical signal abnormality, the third type of abnormality is the uterine muscle electrical signal abnormality, and the fourth type of abnormality is the strong electromagnetic interference abnormality;

[0009] Performing signal separation on the segmented mixed maternal-fetal electrocardiogram signal to obtain a target signal after removing the maternal electrocardiogram signal from the mixed maternal-fetal electrocardiogram signal;

[0010] Perform R wave recognition on the target signal to obtain the R wave recognition result of the target signal;

[0011] According to the R wave recognition result of the target signal, it is judged whether the target signal has the fifth type of abnormality. If the fifth type of abnormality exists, the target signal is marked as abnormal, wherein the fifth type of abnormality is the absence of fetal ECG signal abnormality;

[0012] According to the R wave recognition result of the target signal, the fetal heart rate curve is drawn;

[0013] The data segment marked as abnormal is not displayed or the fetal heart rate curve is covered with blocks of different colors or transparency, wherein the same type of abnormality corresponds to the same color or transparency block.

[0014] Optionally, filtering and segmenting the mixed maternal-fetal electrocardiogram signal to obtain a segmented mixed maternal-fetal electrocardiogram signal includes:

[0015] Eliminate baseline drift, power frequency interference and high-frequency disturbance of mixed maternal-fetal ECG signals;

[0016] The mixed maternal-fetal electrocardiogram signal is segmented and processed using a sliding window with a preset sliding step size and a preset length to obtain a segmented mixed maternal-fetal electrocardiogram signal.

[0017] Optionally, the bioelectrical impedance of the monitoring site is measured by direct current or single frequency or multi-frequency method, and whether there is a Class 1 abnormality is determined based on the bioelectrical impedance.

[0018] Optionally, when the bioelectrical impedance measured at the monitoring site is greater than the impedance value threshold, it is determined that a Class 1 abnormality exists.

[0019] Optionally, the method for determining the second type of abnormality is:

[0020] For each segment of the mixed maternal-fetal ECG signal, determine whether the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and whether the mean of the absolute value of the derivative is greater than the second threshold, and determine whether there is a Class 2 abnormality based on the judgment result. When the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and the mean of the absolute value of the derivative is greater than the second threshold, it is determined that a Class 2 abnormality exists.

[0021] Optionally, the method for determining the third type of abnormality is:

[0022] For each segment of the mixed maternal-fetal ECG signal, determine whether the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and whether the mean of the absolute value of the derivative is greater than the second threshold, and whether the continuous time when the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and the mean of the absolute value of the derivative is greater than the second threshold is greater than the shortest contraction duration and less than the longest contraction duration, and determine whether there is a Class 3 abnormality based on the judgment result, wherein when the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and the mean of the absolute value of the derivative is greater than the second threshold, and the continuous time is greater than the shortest contraction duration and less than the longest contraction duration, it is determined that there is a Class 3 abnormality.

[0023] Optionally, the method for determining the fourth type of abnormality is:

[0024] Perform FFT transformation on the reference mixed maternal-fetal ECG signal, take the upper envelope, and mark it as the reference upper envelope;

[0025] Perform FFT transformation on the segmented mixed maternal-fetal ECG signal and obtain the upper envelope, which is marked as the upper envelope to be detected;

[0026] Take the absolute value of the difference between the element in the reference upper envelope and the corresponding element in the upper envelope to be detected, calculate the mean of the absolute values ​​of the differences, and if the mean of the absolute values ​​of the differences is greater than the third threshold, it is determined that a fourth type of abnormality exists.

[0027] Optionally, the method for determining the fifth type of abnormality is:

[0028] Performing maternal-fetal electrocardiogram separation on the segmented mixed maternal-fetal electrocardiogram signal by blind source separation or pattern recognition method, extracting the separated fetal electrocardiogram signal, or suppressing the maternal electrocardiogram signal on the segmented mixed maternal-fetal electrocardiogram signal by filtering signal processing method, and obtaining the fetal electrocardiogram signal;

[0029] The R wave peak of the fetal ECG signal is identified by Pan&Tompkins algorithm to obtain the R wave peak position;

[0030] Take the signal of the first time length before and the signal of the second time length after the R wave peak position, calculate the first slope mean absolute value of the signal of the first time length and the second slope mean absolute value of the signal of the second time length; if the first slope mean absolute value is less than the fourth threshold and the second slope mean absolute value is less than the fifth threshold, it is judged that a Category 5 abnormality exists.

[0031] Optionally, extracting the fetal heart rate signal and drawing a fetal heart rate curve includes:

[0032] The RR interval was calculated based on the position of the R wave peak;

[0033] Calculate the instantaneous heart rate value at each sampling point according to the RR interval;

[0034] The fetal heart rate signal is generated according to the instantaneous heart rate value, and the fetal heart rate curve is drawn.

[0035] A second aspect of the present invention provides a fetal electrocardiogram monitoring system based on a marker interference signal, comprising:

[0036] The original signal acquisition module is used to collect the mixed maternal and fetal ECG signals of pregnant women through non-invasive fetal ECG monitoring equipment;

[0037] A preprocessing module, used for filtering and segmenting the mixed maternal-fetal electrocardiogram signal to obtain a segmented mixed maternal-fetal electrocardiogram signal;

[0038] The first abnormality detection module is used to detect whether each segment of the mixed maternal-fetal electrocardiogram signal has the first type of abnormality, the second type of abnormality, the third type of abnormality and the fourth type of abnormality, and mark the mixed maternal-fetal electrocardiogram signal data segment with abnormalities, wherein the first type of abnormality is the interface state abnormality, the second type of abnormality is the abdominal muscle electrical signal abnormality, the third type of abnormality is the uterine muscle electrical signal abnormality, and the fourth type of abnormality is the strong electromagnetic interference abnormality;

[0039] The maternal-fetal ECG signal separation module is used to perform signal separation on the segmented mixed maternal-fetal ECG signal to obtain a target signal after removing the maternal ECG signal from the mixed maternal-fetal ECG signal;

[0040] The fetal ECG R wave detection module is used to perform R wave recognition on the target signal and obtain the R wave recognition result of the target signal;

[0041] The second abnormality detection module is used to determine whether the target signal has a fifth-category abnormality according to the R wave recognition result of the target signal, and if the fifth-category abnormality exists, the target signal is marked as abnormal, wherein the fifth-category abnormality is that there is no abnormal fetal electrocardiogram signal;

[0042] A heart rate curve drawing module is used to draw a fetal heart rate curve according to the R wave recognition result of the target signal;

[0043] The abnormal marking module is used to not display the data segment marked as abnormal or to cover the fetal heart rate curve with different colors or color blocks of different transparency, wherein the same type of abnormality corresponds to the same color or transparency color block.

[0044] It can be seen from the above technical solutions that the fetal electrocardiogram monitoring method based on the marker interference signal provided by the present invention has the following advantages:

[0045] The fetal electrocardiogram monitoring method based on the marked interference signal provided by the present invention filters and segments the mixed maternal-fetal electrocardiogram signal after collecting it, then detects whether there are the first to fourth category abnormalities, and then performs R wave recognition on the segmented mixed maternal-fetal electrocardiogram signal to detect whether there is the fifth category abnormality. After extracting the target signal containing the fetal heart rate signal, a fetal heart rate curve is drawn according to the R wave recognition result of the target signal. According to the abnormal detection results of categories 1 to 5, the data segment marked as abnormal is not displayed in the fetal heart rate curve, or the data segment marked as abnormal is displayed in the fetal heart rate curve in a visual manner of different colors or different transparency color blocks, which can prevent users from clinically interpreting unreliable monitoring results, improve the accuracy and reliability of fetal electrocardiogram monitoring, and solve the technical problem that the monitoring signal abnormality caused by interference leads to unreliable fetal electrocardiogram monitoring results in the fetal electrocardiogram monitoring process. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 A schematic flow chart of a fetal electrocardiogram monitoring method based on a marker interference signal provided in the present invention;

[0048] Figure 2 It is a structural schematic diagram of a fetal electrocardiogram monitoring system based on a marker interference signal provided in the present invention. DETAILED DESCRIPTION

[0049] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] For easier understanding, see Figure 1 The present invention provides an embodiment of a fetal electrocardiogram monitoring method based on a marker interference signal, comprising:

[0051] Step 101: Collect mixed maternal-fetal ECG signals of a pregnant woman using a non-invasive fetal ECG monitoring device.

[0052] It should be noted that the embodiment of the present invention uses a non-invasive fetal ECG monitoring device to perform fetal ECG monitoring, and collects the mixed maternal-fetal ECG signal of the pregnant woman through electrodes fixed on the abdomen of the pregnant woman. The sampling frequency of the mixed maternal-fetal ECG signal can be set to 1kHz.

[0053] Step 102: filter and segment the mixed maternal-fetal electrocardiogram signal to obtain a segmented mixed maternal-fetal electrocardiogram signal.

[0054] It should be noted that after the mixed maternal-fetal ECG signal S is obtained, the mixed maternal-fetal ECG signal S is filtered, mainly including the elimination of baseline drift, power frequency interference and high-frequency disturbance. Specifically, a 6th-order high-pass Butterworth filter with a cutoff frequency of 1 Hz is used to eliminate baseline drift, a 10th-order notch filter with a cutoff frequency of 50 Hz / 60 Hz is used to eliminate power frequency signal interference, and a 2nd-order low-pass Chebyshev type II filter with a cutoff frequency of 200 Hz is used to eliminate high-frequency noise. The mixed maternal-fetal ECG signals of multiple channels are processed in the same way.

[0055] For the mixed maternal-fetal ECG signal S, it can be expressed as:

[0056]

[0057] Among them, s i ,i∈[1,n] is the signal value sampled once, the unit is uV, and n is the number of samples.

[0058] Use the sliding window method to segment the signal S, assuming that the window length is w and the single sliding length (i.e. sliding step length) is d. Then:

[0059]

[0060] Among them, S1 and S2 are the signals of the first and second segments obtained after segmentation, and the signals of other segments can be obtained by analogy, which will not be described in detail here.

[0061] In the present invention, the window length of the sliding window is 128, and the single sliding length (ie, sliding step length) is 16.

[0062] Step 103, detect whether each segment of the mixed maternal-fetal ECG signal has the first type of abnormality, the second type of abnormality, the third type of abnormality and the fourth type of abnormality, and mark the mixed maternal-fetal ECG signal data segments with abnormalities, wherein the first type of abnormality is the interface state abnormality, the second type of abnormality is the abdominal muscle electrical signal abnormality, the third type of abnormality is the uterine muscle electrical signal abnormality, and the fourth type of abnormality is the strong electromagnetic interference abnormality.

[0063] It should be noted that the first type of abnormality is an abnormal interface state, which means that the monitoring interface is in a poor state, the electrode pasting position is abnormal, and the signal utilization rate is low. This type of abnormality can be obtained through impedance measurement, and the extreme case is lead detachment. The bioimpedance of the monitoring site can be measured by DC or single-frequency or multi-frequency method, and the presence of a first type of abnormality can be determined based on the bioimpedance. When the bioimpedance of the monitoring site is greater than the impedance value threshold, it is determined that a first type of abnormality exists. For example, the normal human body impedance value, that is, the impedance value threshold is 100kΩ. When the bioimpedance of the monitoring site is greater than 100kΩ, it is marked as a first type of abnormality.

[0064] The second type of abnormality is the abnormality of the abdominal muscle electrical signal. The occurrence of this abnormality means that the signal is mixed with the abdominal muscle electrical signal, the signal utilization rate is low, the signal frequency component has obvious changes or the zero crossing in the time domain has increased significantly. The method used to detect this type of abnormality can be: for each segment of mixed maternal-fetal ECG signal, determine whether the zero crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and whether the mean of the absolute value of the derivative is greater than the second threshold, and determine whether there is a second type of abnormality based on the judgment result, wherein when the zero crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and the mean of the absolute value of the derivative is greater than the second threshold, it is determined that a second type of abnormality exists. Let the first threshold be T zero , the second threshold is M slop , f s is the sampling frequency, then the conditions for marking as type 2 anomalies are:

[0065]

[0066] The third type of abnormality is the abnormality of uterine muscle electrical signals. The occurrence of this abnormality means that the signal is mixed with uterine muscle electrical signals, the utilization rate of the signal is low, the frequency components of the signal have obvious changes or the number of zero crossings in the time domain has increased significantly, and the duration is long. That is, under the conditions of the second type of abnormality, the signal time length is greater than a time length, which is the shortest duration of uterine contractions, and the signal time length is less than a time length, which is the longest duration of uterine contractions. That is, the method used to detect this type of abnormality can be: for each segment of mixed maternal-fetal ECG signal, determine whether the zero crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold T zero Is the mean of the absolute value of the derivative greater than the second threshold M? slop , and when the zero crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold T zero And the mean of the absolute value of the derivative is greater than the second threshold M slop Whether the continuous time of the situation is greater than the shortest duration of uterine contraction t1 and less than the longest duration of uterine contraction t2, and whether there is a third type of abnormality according to the judgment result, wherein, when the zero crossing rate of the mixed maternal-fetal electrocardiogram signal is greater than the first threshold T zero And the mean of the absolute value of the derivative is greater than the second threshold M slop, and when the continuous time is greater than the shortest uterine contraction duration t1 and less than the longest uterine contraction duration t2, it is determined that there is a Class 3 abnormality. In one embodiment, the shortest uterine contraction duration t1 is 20 seconds, and the longest uterine contraction duration t2 is 3 minutes.

[0067] The fourth type of abnormality is a strong electromagnetic interference abnormality, which means that there is strong external electromagnetic interference during the monitoring process, the frequency component of the signal is incorrect for a long period of time, and the quality of the mother's QRS waveform (MQRS) detection is low. The method used to detect this type of abnormality can be: first obtain a reference mixed maternal-fetal ECG signal, which is a mixed maternal-fetal ECG signal obtained in an electromagnetic shielding room using an ECG monitor without a radio frequency transmission function at the same lead position. Then perform an FFT transformation on the reference mixed maternal-fetal ECG signal, take the upper envelope, and mark it as the reference upper envelope E. d , such as E d =[0.000204,0.000203,…,0.000219]. Perform FFT transformation on the segmented mixed maternal-fetal ECG signal and take the upper envelope, which is marked as the upper envelope to be detected E. y , refer to the upper envelope E d The elements in the upper envelope E to be detected y The corresponding elements in are de-interpreted, and the mean absolute value of the difference is calculated. If the mean absolute value of the difference is greater than the third threshold, it is determined that there is a fourth type of abnormality. Specifically, the reference mixed maternal-fetal ECG signal is transformed by FFT to obtain a complex signal vector E containing frequency components = [e1 e2 …e n ], taking the elements of vector E modulo, we get D = [d1 d2 … d n ]. D is made into an RMS envelope of length L, and the upper envelope is denoted as E d Perform FFT on the sampled and segmented signal, and take the modulus of the auxiliary signal elements to obtain Y = [y1 y2 …y n ]. Let Y be an RMS envelope of length L, where the upper envelope is denoted by E y The value of L is 5≤L≤20. If the average of the absolute values ​​of the differences is greater than the third threshold value f, it is determined that a fourth type of abnormality exists. Among them, 0.1≤f≤0.6, and f is preferably 0.5.

[0068] Step 104 , performing signal separation on the segmented mixed maternal-fetal ECG signal to obtain a target signal after removing the maternal ECG signal from the mixed maternal-fetal ECG signal.

[0069] Step 105: Perform R wave recognition on the target signal to obtain an R wave recognition result of the target signal.

[0070] Step 106: determine whether the target signal has a fifth-category abnormality based on the R wave recognition result of the target signal. If the fifth-category abnormality exists, mark the target signal as abnormal, wherein the fifth-category abnormality is the absence of a fetal ECG signal abnormality.

[0071] It should be noted that after completing the detection of the first to fourth types of abnormalities, the mixed maternal and fetal ECG signals obtained after filtering and segmentation are subjected to signal separation and R wave identification to determine whether there is a fifth type of abnormality. The fifth type of abnormality is the absence of fetal ECG signal abnormality, which means that the R wave morphology or RR interval of the separated and detected fetal QRS waveform (FQRS) is abnormal. The method used to detect this type of abnormality can be:

[0072] (1) performing maternal-fetal electrocardiogram separation on the segmented mixed maternal-fetal electrocardiogram signal by blind source separation or pattern recognition method, and extracting the target signal containing the fetal electrocardiogram signal after the maternal electrocardiogram signal is separated, or using a filtering signal processing method to suppress the maternal electrocardiogram signal on the segmented mixed maternal-fetal electrocardiogram signal to obtain the fetal electrocardiogram signal, such as extended Kalman filtering and particle filtering, preferably particle filtering;

[0073] (2) The R wave peak of the target signal of the fetal ECG signal is identified by the Pan & Tompkins algorithm to obtain the R wave peak position;

[0074] (3) Take the first time length t before the peak position of the R wave a (1ms<t a ≤10ms) and the first time length t b (1ms<t b ≤10ms) signal, calculate the previous t a The time length of the signal's first slope mean absolute value M1 and after t b The second slope mean absolute value M2 of the signal of the time length, if the first slope mean absolute value M1 is less than the fourth threshold and the second slope mean absolute value M2 is less than the fifth threshold, it is judged that there is a fifth type of abnormality. Among them, the fourth threshold is preferably 8, the fifth threshold is preferably 4, t a 10ms is preferred, t b 10ms.

[0075] Step 107: Draw a fetal heart rate curve based on the R wave recognition result of the target signal.

[0076] It should be noted that the RR interval, i.e., the time interval between single heartbeats, is calculated for the calculated fetal R wave peak position. Then, the instantaneous heart rate value of the sampling point is obtained by dividing 60 by the time interval between single heartbeats. In the entire sampling time window, the instantaneous heart rate value of each sampling point is calculated according to this method, thereby obtaining the fetal heart rate signal and drawing the fetal heart rate curve.

[0077] Step 108: The data segment marked as abnormal is not displayed or the fetal heart rate curve is covered with blocks of different colors or different transparency, wherein the same type of abnormality corresponds to the same color or transparency block.

[0078] It should be noted that abnormal data segments are visually marked on the fetal heart rate curve. For example, green or green translucent blocks are used to mark Category 1 abnormalities, blue or blue translucent blocks are used to mark Category 2 abnormalities, yellow or yellow translucent blocks are used to mark Category 3 abnormalities, gray or gray translucent blocks are used to mark Category 4 abnormalities, and red or red translucent blocks are used to mark Category 5 abnormalities.

[0079] The fetal ECG monitoring method based on the marked interference signal provided by the embodiment of the present invention, after collecting the mixed maternal-fetal ECG signal, filters and segments the mixed maternal-fetal ECG signal, then detects whether there are abnormalities of categories 1 to 4, and then performs R wave recognition on the segmented mixed maternal-fetal ECG signal to detect whether there is abnormality of category 5. After extracting the target signal containing the fetal heart rate signal, a fetal heart rate curve is drawn according to the R wave recognition result of the target signal. According to the abnormal detection results of categories 1 to 5, the data segment marked as abnormal is not displayed in the fetal heart rate curve, or the data segment marked as abnormal is displayed in the fetal heart rate curve in a visual manner of different colors or different transparency color blocks, which can prevent users from clinically interpreting unreliable monitoring results, improve the accuracy and reliability of fetal ECG monitoring, and solve the technical problem that the monitoring signal abnormality caused by interference leads to unreliable fetal ECG monitoring results during fetal ECG monitoring.

[0080] For easier understanding, see Figure 2 The present invention provides an embodiment of a fetal electrocardiogram monitoring system based on a marker interference signal, comprising:

[0081] The original signal acquisition module is used to collect the mixed maternal and fetal ECG signals of pregnant women through non-invasive fetal ECG monitoring equipment;

[0082] A preprocessing module, used for filtering and segmenting the mixed maternal-fetal electrocardiogram signal to obtain a segmented mixed maternal-fetal electrocardiogram signal;

[0083] The first abnormality detection module is used to detect whether each segment of the mixed maternal-fetal electrocardiogram signal has the first type of abnormality, the second type of abnormality, the third type of abnormality and the fourth type of abnormality, and mark the mixed maternal-fetal electrocardiogram signal data segment with abnormalities, wherein the first type of abnormality is the interface state abnormality, the second type of abnormality is the abdominal muscle electrical signal abnormality, the third type of abnormality is the uterine muscle electrical signal abnormality, and the fourth type of abnormality is the strong electromagnetic interference abnormality;

[0084] The maternal-fetal ECG signal separation module is used to perform signal separation on the segmented mixed maternal-fetal ECG signal to obtain a target signal after removing the maternal ECG signal from the mixed maternal-fetal ECG signal;

[0085] The fetal ECG R wave detection module is used to perform R wave recognition on the target signal and obtain the R wave recognition result of the target signal;

[0086] The second abnormality detection module is used to determine whether the target signal has a fifth-category abnormality according to the R wave recognition result of the target signal, and if the fifth-category abnormality exists, the target signal is marked as abnormal, wherein the fifth-category abnormality is that there is no abnormal fetal electrocardiogram signal;

[0087] A heart rate curve drawing module is used to draw a fetal heart rate curve according to the R wave recognition result of the target signal;

[0088] The abnormal marking module is used to not display the data segment marked as abnormal or to cover the fetal heart rate curve with different colors or color blocks of different transparency, wherein the same type of abnormality corresponds to the same color or transparency color block.

[0089] The preprocessing module is specifically used for:

[0090] Eliminate baseline drift, power frequency interference and high-frequency disturbance of mixed maternal-fetal ECG signals;

[0091] The mixed maternal-fetal electrocardiogram signal is segmented and processed using a sliding window with a preset sliding step size and a preset length to obtain a segmented mixed maternal-fetal electrocardiogram signal.

[0092] Specifically, the bioelectrical impedance of the monitoring site is measured by direct current or single frequency or multi-frequency method, and whether there is a Class 1 abnormality is determined based on the bioelectrical impedance.

[0093] When the bioelectrical impedance measured at the monitoring site is greater than the impedance value threshold, it is determined that a Class 1 abnormality exists.

[0094] The judgment method for the second type of abnormality is:

[0095] For each segment of the mixed maternal-fetal ECG signal, determine whether the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and whether the mean of the absolute value of the derivative is greater than the second threshold, and determine whether there is a Class 2 abnormality based on the judgment result. When the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and the mean of the absolute value of the derivative is greater than the second threshold, it is determined that a Class 2 abnormality exists.

[0096] The judgment method for the third type of abnormality is:

[0097] For each segment of the mixed maternal-fetal ECG signal, determine whether the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and whether the mean of the absolute value of the derivative is greater than the second threshold, and whether the continuous time when the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and the mean of the absolute value of the derivative is greater than the second threshold is greater than the shortest contraction duration and less than the longest contraction duration, and determine whether there is a Class 3 abnormality based on the judgment result, wherein when the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and the mean of the absolute value of the derivative is greater than the second threshold, and the continuous time is greater than the shortest contraction duration and less than the longest contraction duration, it is determined that there is a Class 3 abnormality.

[0098] The judgment method for the fourth type of abnormality is:

[0099] Perform FFT transformation on the reference mixed maternal-fetal ECG signal, take the upper envelope, and mark it as the reference upper envelope;

[0100] Perform FFT transformation on the segmented mixed maternal-fetal ECG signal and obtain the upper envelope, which is marked as the upper envelope to be detected;

[0101] Take the absolute value of the difference between the element in the reference upper envelope and the corresponding element in the upper envelope to be detected, calculate the mean of the absolute values ​​of the differences, and if the mean of the absolute values ​​of the differences is greater than the third threshold, it is determined that a fourth type of abnormality exists.

[0102] The judgment method for the fifth type of abnormality is:

[0103] Performing maternal-fetal electrocardiogram separation on the segmented mixed maternal-fetal electrocardiogram signal by blind source separation or pattern recognition method, extracting the separated fetal electrocardiogram signal, or suppressing the maternal electrocardiogram signal on the segmented mixed maternal-fetal electrocardiogram signal by filtering signal processing method, and obtaining the fetal electrocardiogram signal;

[0104] The R wave peak of the fetal ECG signal is identified by Pan&Tompkins algorithm to obtain the R wave peak position;

[0105] Take the signal of the first time length before and the signal of the second time length after the R wave peak position, calculate the first slope mean absolute value of the signal of the first time length and the second slope mean absolute value of the signal of the second time length; if the first slope mean absolute value is less than the fourth threshold and the second slope mean absolute value is less than the fifth threshold, it is judged that a Category 5 abnormality exists.

[0106] The heart rate curve drawing module is specifically used for:

[0107] The RR interval was calculated based on the peak position of the R wave;

[0108] The instantaneous heart rate value calculated based on the RR interval;

[0109] The fetal heart rate signal is generated according to the instantaneous heart rate value, and the fetal heart rate curve is drawn.

[0110] The fetal ECG monitoring system based on the marked interference signal provided by the embodiment of the present invention filters and segments the mixed maternal-fetal ECG signal after collecting it, and then detects whether there are abnormalities of categories 1 to 4, and then performs R wave recognition on the segmented mixed maternal-fetal ECG signal to detect whether there is abnormality of category 5. After extracting the target signal containing the fetal heart rate signal, a fetal heart rate curve is drawn according to the R wave recognition result of the target signal. According to the abnormal detection results of categories 1 to 5, the data segment marked as abnormal is not displayed in the fetal heart rate curve, or the data segment marked as abnormal is displayed in the fetal heart rate curve in a visual manner of different colors or different transparency color blocks, which can prevent users from clinically interpreting unreliable monitoring results, improve the accuracy and reliability of fetal ECG monitoring, and solve the technical problem that the monitoring signal abnormality caused by interference leads to unreliable fetal ECG monitoring results during fetal ECG monitoring.

[0111] The fetal ECG monitoring system based on a marker interference signal provided in an embodiment of the present invention is used to execute the fetal ECG monitoring method based on a marker interference signal in the aforementioned embodiment of the fetal ECG monitoring method based on a marker interference signal. The principle is the same as the fetal ECG monitoring method based on a marker interference signal in the aforementioned embodiment of the fetal ECG monitoring method based on a marker interference signal, and will not be repeated here.

[0112] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fetal electrocardiogram monitoring method based on a marker interference signal, characterized in that: include: The mixed maternal and fetal ECG signals of pregnant women are collected by non-invasive fetal ECG monitoring equipment; Performing filtering and segmentation processing on the mixed maternal-fetal electrocardiogram signal to obtain a segmented mixed maternal-fetal electrocardiogram signal; Detect whether each segment of the mixed maternal-fetal electrocardiogram signal has the first type of abnormality, the second type of abnormality, the third type of abnormality and the fourth type of abnormality, and mark the mixed maternal-fetal electrocardiogram signal data segment with abnormalities, wherein the first type of abnormality is the interface state abnormality, the second type of abnormality is the abdominal muscle electrical signal abnormality, the third type of abnormality is the uterine muscle electrical signal abnormality, and the fourth type of abnormality is the strong electromagnetic interference abnormality; Performing signal separation on the segmented mixed maternal-fetal electrocardiogram signal to obtain a target signal after removing the maternal electrocardiogram signal from the mixed maternal-fetal electrocardiogram signal; Perform R wave recognition on the target signal to obtain the R wave recognition result of the target signal; According to the R wave recognition result of the target signal, it is judged whether the target signal has the fifth type of abnormality. If the fifth type of abnormality exists, the target signal is marked as abnormal, wherein the fifth type of abnormality is the absence of fetal ECG signal abnormality; According to the R wave recognition result of the target signal, the fetal heart rate curve is drawn; The data segments marked as abnormal are not displayed or the fetal heart rate curve is covered with blocks of different colors or transparency, wherein the same type of abnormality corresponds to the same color or transparency block; The judgment method for the second type of abnormality is: For each segment of the mixed maternal-fetal electrocardiogram signal, determine whether the zero-crossing rate of the mixed maternal-fetal electrocardiogram signal is greater than a first threshold and whether the average of the absolute value of the derivative is greater than a second threshold, and determine whether there is a second type of abnormality according to the determination result, wherein when the zero-crossing rate of the mixed maternal-fetal electrocardiogram signal is greater than the first threshold and the average of the absolute value of the derivative is greater than the second threshold, it is determined that there is a second type of abnormality; The judgment method for the third type of abnormality is: For each segment of the mixed maternal-fetal ECG signal, determine whether the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than a first threshold and whether the mean of the absolute value of the derivative is greater than a second threshold, and determine whether the continuous time when the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and the mean of the absolute value of the derivative is greater than the second threshold is greater than the shortest contraction duration and less than the longest contraction duration, and determine whether there is a Class 3 abnormality based on the judgment result, wherein when the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and the mean of the absolute value of the derivative is greater than the second threshold, and the continuous time is greater than the shortest contraction duration and less than the longest contraction duration, it is determined that there is a Class 3 abnormality; The judgment method for the fifth type of abnormality is: Performing maternal-fetal electrocardiogram separation on the segmented mixed maternal-fetal electrocardiogram signal by blind source separation or pattern recognition method, extracting the separated fetal electrocardiogram signal, or suppressing the maternal electrocardiogram signal on the segmented mixed maternal-fetal electrocardiogram signal by filtering signal processing method, and obtaining the fetal electrocardiogram signal; The R wave peak of the fetal ECG signal is identified by Pan&Tompkins algorithm to obtain the R wave peak position; Take the signal of the first time length before and the signal of the second time length after the R wave peak position, calculate the first slope mean absolute value of the signal of the first time length and the second slope mean absolute value of the signal of the second time length; if the first slope mean absolute value is less than the fourth threshold and the second slope mean absolute value is less than the fifth threshold, it is judged that a Category 5 abnormality exists.

2. The method for fetal electrocardiogram monitoring based on a marker interference signal according to claim 1, characterized in that: The mixed maternal-fetal electrocardiogram signal is filtered and segmented to obtain a segmented mixed maternal-fetal electrocardiogram signal, including: Eliminate baseline drift, power frequency interference and high-frequency disturbance of mixed maternal-fetal ECG signals; The mixed maternal-fetal electrocardiogram signal is segmented and processed using a sliding window with a preset sliding step size and a preset length to obtain a segmented mixed maternal-fetal electrocardiogram signal.

3. The method for fetal electrocardiogram monitoring based on a marker interference signal according to claim 1, characterized in that: The bioelectrical impedance of the monitoring site is measured by direct current or single-frequency or multi-frequency method, and whether there is a Class 1 abnormality is determined based on the bioelectrical impedance.

4. The method for fetal electrocardiogram monitoring based on a marker interference signal according to claim 3, characterized in that: When the bioelectrical impedance measured at the monitoring site is greater than the impedance value threshold, it is determined that a Class 1 abnormality exists.

5. The method for fetal electrocardiogram monitoring based on a marker interference signal according to claim 1, characterized in that: The judgment method for the fourth type of abnormality is: Perform FFT transformation on the reference mixed maternal-fetal ECG signal, take the upper envelope, and mark it as the reference upper envelope; Perform FFT transformation on the segmented mixed maternal-fetal ECG signal and obtain the upper envelope, which is marked as the upper envelope to be detected; Take the absolute value of the difference between the element in the reference upper envelope and the corresponding element in the upper envelope to be detected, calculate the mean of the absolute values ​​of the differences, and if the mean of the absolute values ​​of the differences is greater than the third threshold, it is determined that a fourth type of abnormality exists.

6. The method for fetal electrocardiogram monitoring based on a marker interference signal according to claim 1, characterized in that: Extract the fetal heart rate signal and draw the fetal heart rate curve, including: The RR interval was calculated based on the position of the R wave peak; Calculate the instantaneous heart rate value based on the RR interval; The fetal heart rate signal is generated according to the instantaneous heart rate value, and the fetal heart rate curve is drawn.

7. A fetal electrocardiogram monitoring system based on a marker interference signal, characterized in that: include: The original signal acquisition module is used to collect the mixed maternal and fetal ECG signals of pregnant women through non-invasive fetal ECG monitoring equipment; A preprocessing module, used for filtering and segmenting the mixed maternal-fetal electrocardiogram signal to obtain a segmented mixed maternal-fetal electrocardiogram signal; The first abnormality detection module is used to detect whether each segment of the mixed maternal-fetal electrocardiogram signal has the first type of abnormality, the second type of abnormality, the third type of abnormality and the fourth type of abnormality, and mark the mixed maternal-fetal electrocardiogram signal data segment with abnormalities, wherein the first type of abnormality is the interface state abnormality, the second type of abnormality is the abdominal muscle electrical signal abnormality, the third type of abnormality is the uterine muscle electrical signal abnormality, and the fourth type of abnormality is the strong electromagnetic interference abnormality; The maternal-fetal ECG signal separation module is used to perform signal separation on the segmented mixed maternal-fetal ECG signal to obtain a target signal after removing the maternal ECG signal from the mixed maternal-fetal ECG signal; The fetal ECG R wave detection module is used to perform R wave recognition on the target signal and obtain the R wave recognition result of the target signal; The second abnormality detection module is used to determine whether the target signal has a fifth-category abnormality according to the R wave recognition result of the target signal, and if the fifth-category abnormality exists, the target signal is marked as abnormal, wherein the fifth-category abnormality is that there is no abnormal fetal electrocardiogram signal; A heart rate curve drawing module is used to draw a fetal heart rate curve according to the R wave recognition result of the target signal; An abnormal marking module, used to not display the data segment marked as abnormal or to cover the fetal heart rate curve with blocks of different colors or different transparency, wherein the same type of abnormality corresponds to the same color or transparency block; The judgment method for the second type of abnormality is: For each segment of the mixed maternal-fetal electrocardiogram signal, determine whether the zero-crossing rate of the mixed maternal-fetal electrocardiogram signal is greater than a first threshold and whether the average of the absolute value of the derivative is greater than a second threshold, and determine whether there is a second type of abnormality according to the determination result, wherein when the zero-crossing rate of the mixed maternal-fetal electrocardiogram signal is greater than the first threshold and the average of the absolute value of the derivative is greater than the second threshold, it is determined that there is a second type of abnormality; The judgment method for the third type of abnormality is: For each segment of the mixed maternal-fetal ECG signal, determine whether the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than a first threshold and whether the mean of the absolute value of the derivative is greater than a second threshold, and determine whether the continuous time when the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and the mean of the absolute value of the derivative is greater than the second threshold is greater than the shortest contraction duration and less than the longest contraction duration, and determine whether there is a Class 3 abnormality based on the judgment result, wherein when the zero-crossing rate of the mixed maternal-fetal ECG signal is greater than the first threshold and the mean of the absolute value of the derivative is greater than the second threshold, and the continuous time is greater than the shortest contraction duration and less than the longest contraction duration, it is determined that there is a Class 3 abnormality; The judgment method for the fifth type of abnormality is: Performing maternal-fetal electrocardiogram separation on the segmented mixed maternal-fetal electrocardiogram signal by blind source separation or pattern recognition method, extracting the separated fetal electrocardiogram signal, or suppressing the maternal electrocardiogram signal on the segmented mixed maternal-fetal electrocardiogram signal by filtering signal processing method, and obtaining the fetal electrocardiogram signal; The R wave peak of the fetal ECG signal is identified by Pan&Tompkins algorithm to obtain the R wave peak position; Take the signal of the first time length before and the signal of the second time length after the R wave peak position, calculate the first slope mean absolute value of the signal of the first time length and the second slope mean absolute value of the signal of the second time length; if the first slope mean absolute value is less than the fourth threshold and the second slope mean absolute value is less than the fifth threshold, it is judged that a Category 5 abnormality exists.

Images