A high signal-to-noise ratio fetal electrocardio separation acquisition method and device

By preprocessing and quality inspection of the mixed maternal ECG signals, preliminary estimates of maternal and fetal ECGs are constructed. Subtraction calculation and reconstruction matrix processing are performed to solve the problem of interference between fetal ECG signals and maternal ECG signals, and high signal-to-noise ratio fetal ECG signals are obtained.

CN116269422BActive Publication Date: 2026-04-24JIANGSU UNIV OF SCI & TECH
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
JIANGSU UNIV OF SCI & TECH
Filing Date
2023-03-27
Publication Date
2026-04-24

AI Technical Summary

Technical Problem

Fetal electrocardiogram (ECG) signals are easily interfered with by maternal ECG signals, resulting in low accuracy of fetal ECG monitoring results.

Method used

By preprocessing the mixed maternal electrocardiogram (ECG) signals, performing R-peak detection and ECG signal quality detection, a preliminary estimate of maternal and fetal ECG signals is constructed. Subtraction calculation and reconstruction matrix processing are then performed to remove noise and obtain a fetal ECG signal with a high signal-to-noise ratio.

Benefits of technology

It improves the accuracy of fetal electrocardiogram signals and solves the problem of accuracy in fetal electrocardiogram monitoring results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116269422B_ABST
    Figure CN116269422B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of biomedical engineering, in particular to a high signal-to-noise ratio fetal electrocardiogram separation acquisition method and device, through preprocessing of mixed maternal electrocardiogram signals, extracting maternal electrocardiogram signals from the preprocessed mixed maternal electrocardiogram signals, quality detection of the maternal electrocardiogram signals to obtain maternal electrocardiogram signals with high accuracy, removing the maternal electrocardiogram signals with high accuracy from the preprocessed mixed maternal electrocardiogram signals to obtain fetal electrocardiogram signals, quality detection of the fetal electrocardiogram signals to obtain fetal electrocardiogram signals with high accuracy, and receiving signals by a DSP processing unit and performing digital conversion to output fetal electrocardiogram signals with high signal-to-noise ratio, thereby solving the problem that fetal electrocardiogram signals are easily disturbed by maternal electrocardiogram signals, resulting in low accuracy of fetal electrocardiogram monitoring results.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of biomedical engineering technology, and in particular to a method and apparatus for obtaining fetal electrocardiograms with a high signal-to-noise ratio. Background Technology

[0002] Fetal electrocardiogram (ECG) monitoring requires separating the maternal ECG signal (MECG, Maternal Electrocardiogram) and the fetal ECG signal (FECG, fetal electrocardiogram) from the mixed electrical signals collected from the maternal abdomen. During the separation of maternal and fetal ECG signals, the maternal abdominal signals collected first are often accompanied by 50Hz power frequency interference, baseline drift, and interference from some low-frequency signals. Therefore, the collected signals need to be preprocessed. Secondly, the R wave of the maternal QRS complex (i.e., MQRS) is located to construct the MECG template. After removing the MECG, the FECG signal is extracted.

[0003] However, the FECG signal is very weak and easily interfered with by the MECG signal. If the R wave of the MQRS is not accurately located, it will affect the accuracy of maternal-fetal signal separation, thereby seriously affecting the accuracy of fetal electrocardiogram monitoring results. Therefore, how to accurately locate the R wave of the MQRS to improve the accuracy of maternal-fetal signal separation and thus improve the accuracy of fetal electrocardiogram monitoring results is a technical problem that needs to be solved by those skilled in the art. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a method and apparatus for obtaining fetal electrocardiogram (ECG) with a high signal-to-noise ratio, so as to solve the problem that fetal ECG signals are easily interfered with by maternal ECG signals, resulting in low accuracy of fetal ECG monitoring results.

[0005] To achieve the above objectives, the present invention provides a method for obtaining fetal electrocardiograms with a high signal-to-noise ratio, characterized in that it includes:

[0006] A1: Preprocessing of mixed maternal electrocardiogram signals;

[0007] A2: R-peak detection is performed on the preprocessed mixed maternal ECG to construct a preliminary estimate of the maternal ECG;

[0008] A3: Collect the ECG signal sequence of the first time period in the preliminary estimation of the maternal ECG and perform ECG signal quality detection to obtain the maternal ECG quality index;

[0009] A4: Compare the maternal ECG quality index with the maternal ECG quality threshold to determine if the condition is met. If yes, obtain the ECG signal corresponding to the maternal ECG quality index, add the ECG signal to the target signal group, and construct the secondary estimate of the maternal ECG. If no, return to step A3, collect the ECG signal sequence of adjacent time periods in the preliminary estimate of the maternal ECG for detection until the condition is met, obtain the ECG signal corresponding to the maternal ECG quality index, add the ECG signal to the target signal group, and construct the secondary estimate of the maternal ECG.

[0010] A5: The difference between the secondary estimate of the maternal electrocardiogram and the preprocessed mixed maternal electrocardiogram signal is calculated, and the calculation result is set as the preliminary estimate of the fetal electrocardiogram;

[0011] A6: Collect the ECG signal sequence of the first time period in the preliminary estimate of the fetal ECG and perform ECG signal quality detection to obtain the fetal ECG quality index;

[0012] A7: Compare the fetal electrocardiogram quality index with the fetal electrocardiogram quality threshold to determine whether the conditions are met. If yes, obtain the electrocardiogram signal corresponding to the fetal electrocardiogram quality index, add the electrocardiogram signal to the target signal group, and construct the fetal electrocardiogram secondary estimation. If no, return to step A6, collect the electrocardiogram signal sequence of adjacent time periods in the preliminary fetal electrocardiogram estimation for detection until the conditions are met, obtain the electrocardiogram signal corresponding to the fetal electrocardiogram quality index, add the electrocardiogram signal to the target signal group, and construct the fetal electrocardiogram secondary estimation.

[0013] A8: Construct a reconstruction matrix for the secondary estimation of the fetal electrocardiogram to obtain a fetal electrocardiogram signal with a high signal-to-noise ratio.

[0014] Preferably, in A1, a 50Hz digital notch filter and a 0.05-100Hz bandpass filter are used to complete the preprocessing of the mixed maternal electrocardiogram signal.

[0015] Preferably, in A2, the method for detecting the R-peak includes detecting the R-peak on the maternal abdominal electrocardiogram. Let the length be T, the R-peak is located within a window of length T, where T approximates an RR interval. By calculating the phase, the RR interval is mapped to the phase domain, taking the midpoint of the RR interval as the starting point. The formula for the phase interval is:

[0016] Δθ=2πT S / T R

[0017] T S T is the sampling interval. R For RR interval;

[0018] The template signal is obtained using the arithmetic mean method. Let all QRS waves be x1, x2, ..., xn The formula for reconstructing the MECG template is:

[0019]

[0020] The average duration of the maternal QRS complex was obtained by averaging the total number of time-wound beats to construct a preliminary estimate of the maternal electrocardiogram.

[0021] Preferably, in A5, the difference calculation includes setting N to be the length of the sequence, and setting the preprocessed mixed maternal electrocardiogram signal sequence to be AECG = {x1, x2, ..., x...}. N Let the maternal electrocardiogram quadratic estimation sequence be MECG={y1,y2,...,y N The new sequence, FECG, is obtained by subtracting the mixed maternal electrocardiogram (ECG) signal sequence from the secondary estimated maternal ECG sequence. This new sequence is: FECG = {(x1-y1),(x2-y2),...,(x...}. N -y N )}.

[0022] Preferably, in A8, the construction of the reconstruction matrix includes:

[0023] B1: The length of the RR interval estimated by the second estimate of the fetal electrocardiogram is estimated by R-peak detection. The length of 2q RR intervals is estimated continuously, and the length of every two adjacent RR intervals is recorded as a column period, which is then denoted as l1, l2, ..., l q ;

[0024] B2: Continuously capture 2q segments of fetal electrocardiogram data, record the length of the RR interval between every two segments, and select the longest length l. max =max(l1,l2,…,l q ) as the reconstruction matrix A f The number of columns;

[0025] B3: If matrix A f The length l of the ECG data in the k-th row (1≤k≤2q) k equal to l max Let other lengths be less than l max ECG data is padded with zeros to make its length 1. max To obtain a period of l max The electrocardiogram signal was used to construct a 2q×l max Reconstruction matrix A f ,have:

[0026]

[0027] B4: After singular value decomposition, retain the largest singular value σ1 and set the remaining singular values ​​to 0 to obtain matrix A. f The estimates are as follows:

[0028]

[0029] u1 and All are singular value parameters;

[0030] B5: Discard the data corresponding to the positions where zeros were originally padded in the matrix, and take out the remaining data from top to bottom, row by row, and rearrange them into a one-dimensional signal to obtain the 2q segment fetal electrocardiogram signal after noise removal;

[0031] B6: Repeat steps B1 to B5 above, processing 2q segments of ECG data each time, thus achieving denoising of all fetal ECG signals and obtaining fetal ECG signals with high signal-to-noise ratio.

[0032] Preferably, the method for detecting electrocardiogram quality is as follows:

[0033] C1: Acquire the electrocardiogram signal sequence of the first time period of a preset length;

[0034] C2: Perform phase space reconstruction on the electrocardiogram (ECG) signal sequence, and calculate the ECG signal based on the number of occurrences of various permutations to obtain the permutation entropy;

[0035] C3: Obtain the entropy threshold, and calculate the quality index of the electrocardiogram signal based on the permutation entropy and the entropy threshold;

[0036] C4: Obtain the quality threshold, determine whether the quality index meets the standard based on the quality threshold, if yes, obtain the ECG signal corresponding to the quality index and add the ECG signal to the target signal group; if not, return to step C1, update the entropy threshold based on the quality index, and collect ECG signals from adjacent time periods for phase space reconstruction until the quality index meets the quality threshold, obtain the ECG signal corresponding to the quality index and add the ECG signal to the target signal group.

[0037] Preferably, in C2, the method for obtaining the permutation entropy includes setting a one-dimensional time series: {Y(i), i = 1, 2, ..., n} with length n, reconstructing the phase space for any element Y(i), and according to the phase space reconstruction delay coordinate method proposed by Takens, that is, taking m consecutive sample points for each sampling point and delaying the coordinates to obtain a reconstruction vector group in m-dimensional space:

[0038] Y i =[y(i),y(i+l),...,y(i+(m-1)l)]

[0039] In the formula: m and l are the embedding dimension and delay time, respectively, and Y i Arrange the m reconstructed components in ascending order to obtain:

[0040] y(i+(j1-1)l)≤y(i+(j2-1)l)≤...≤y(i+(j m -1)l)

[0041] The obtained vector Y i The arrangement is {j1,j2,...,j...} m The occurrence frequency of each possible arrangement in the entire sequence is statistically analyzed, and the relative frequency of each arrangement is calculated as its probability P1, P2, ..., P. k , k≤m!, calculate the permutation entropy using Shannon's definition of information entropy:

[0042]

[0043] To obtain the permutation entropy.

[0044] An apparatus for a high signal-to-noise ratio fetal electrocardiogram separation acquisition method, comprising:

[0045] The signal acquisition module is used to acquire preprocessed mixed maternal electrocardiogram signals;

[0046] The maternal electrocardiogram (ECG) acquisition module is used to acquire the preprocessed mixed maternal ECG signal to obtain the maternal ECG signal.

[0047] The first detection module is used to detect the maternal electrocardiogram (ECG) signal and obtain a highly accurate maternal ECG signal.

[0048] The fetal electrocardiogram (ECG) separation module is used to separate the fetal ECG signal from the preprocessed mixed maternal ECG signal by combining the highly accurate maternal ECG signal.

[0049] The second detection module is used to detect fetal electrocardiogram signals and obtain highly accurate fetal electrocardiogram signals.

[0050] The DSP processing unit is used to receive high-accuracy fetal electrocardiogram (ECG) signals and output high signal-to-noise ratio fetal ECG signals.

[0051] Preferably, the signal acquisition module includes:

[0052] The electrode detachment detection module is used to detect whether the electrode pads have detached. If they have detached, the mixed maternal electrocardiogram signal is collected.

[0053] The signal conditioning module is used to amplify, filter, and perform other preprocessing on the acquired mixed maternal electrocardiogram signals.

[0054] The beneficial effects of this invention are as follows: By preprocessing the mixed maternal electrocardiogram (ECG) signals, acquiring maternal ECG signals from the preprocessed mixed maternal ECG signals, and performing quality detection on the maternal ECG signals to obtain highly accurate maternal ECG signals, by removing highly accurate maternal ECG signals from the preprocessed mixed maternal ECG signals to obtain fetal ECG signals, and by performing quality detection on the fetal ECG signals to obtain highly accurate fetal ECG signals, the DSP processing unit receives the signals and performs digital conversion to output a fetal ECG signal with a high signal-to-noise ratio. This solves the problem that fetal ECG signals are easily interfered with by maternal ECG signals, leading to low accuracy of fetal ECG monitoring results. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is a flowchart illustrating the high signal-to-noise ratio fetal electrocardiogram separation acquisition method according to an embodiment of the present invention;

[0057] Figure 2 This is a schematic flowchart of the R-peak detection method according to an embodiment of the present invention;

[0058] Figure 3 This is a schematic diagram illustrating the process of constructing the reconstruction matrix according to an embodiment of the present invention;

[0059] Figure 4 This is a flowchart illustrating the method for detecting electrocardiogram quality according to an embodiment of the present invention;

[0060] Figure 5 This is a schematic diagram of the fetal electrocardiogram separation device according to an embodiment of the present invention. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0062] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0063] like Figures 1 to 3 As shown, a method for obtaining fetal electrocardiograms with high signal-to-noise ratio includes:

[0064] A1: Preprocessing of mixed maternal electrocardiogram signals;

[0065] A2: R-peak detection is performed on the preprocessed mixed maternal ECG to construct a preliminary estimate of the maternal ECG;

[0066] A3: Collect the ECG signal sequence of the first time period in the preliminary estimation of the maternal ECG and perform ECG signal quality detection to obtain the maternal ECG quality index;

[0067] A4: Compare the maternal ECG quality index with the maternal ECG quality threshold to determine if the condition is met. If yes, obtain the ECG signal corresponding to the maternal ECG quality index, add the ECG signal to the target signal group, and construct the secondary estimate of the maternal ECG. If no, return to step A3, collect the ECG signal sequence of adjacent time periods in the preliminary estimate of the maternal ECG for detection until the condition is met, obtain the ECG signal corresponding to the maternal ECG quality index, add the ECG signal to the target signal group, and construct the secondary estimate of the maternal ECG.

[0068] A5: The difference between the secondary estimate of the maternal electrocardiogram and the preprocessed mixed maternal electrocardiogram signal is calculated, and the calculation result is set as the preliminary estimate of the fetal electrocardiogram;

[0069] A6: Collect the ECG signal sequence of the first time period in the preliminary estimate of the fetal ECG and perform ECG signal quality detection to obtain the fetal ECG quality index;

[0070] A7: Compare the fetal electrocardiogram quality index with the fetal electrocardiogram quality threshold to determine whether the conditions are met. If yes, obtain the electrocardiogram signal corresponding to the fetal electrocardiogram quality index, add the electrocardiogram signal to the target signal group, and construct the fetal electrocardiogram secondary estimation. If no, return to step A6, collect the electrocardiogram signal sequence of adjacent time periods in the preliminary fetal electrocardiogram estimation for detection until the conditions are met, obtain the electrocardiogram signal corresponding to the fetal electrocardiogram quality index, add the electrocardiogram signal to the target signal group, and construct the fetal electrocardiogram secondary estimation.

[0071] A8: Construct a reconstruction matrix for the secondary estimation of the fetal electrocardiogram to obtain a fetal electrocardiogram signal with a high signal-to-noise ratio.

[0072] As an optional embodiment, in A1, a 50Hz digital notch filter and a 0.05-100Hz bandpass filter are used to preprocess the mixed maternal ECG signal to eliminate power frequency interference and suppress baseline drift.

[0073] As an optional embodiment, in A2, the method for detecting the R-peak includes detecting the R-peak on the maternal abdominal electrocardiogram. Let the length be T, the R-peak is located within a window of length T, where T approximates an RR interval. By calculating the phase, the RR interval is mapped to the phase domain, taking the midpoint of the RR interval as the starting point. The formula for the phase interval is:

[0074] Δθ=2πT S / T R

[0075] T S T is the sampling interval. R The RR interval is used to achieve ECG beat separation;

[0076] The template signal is obtained using the arithmetic mean method. Let all QRS waves be x1, x2, ..., x n The formula for reconstructing the MECG template is:

[0077]

[0078] The average duration of the maternal QRS complex was obtained by averaging the total number of time-wound beats to construct a preliminary estimate of the maternal electrocardiogram.

[0079] As an optional embodiment, in A5, the difference calculation includes setting N to be the length of the sequence, and setting the preprocessed mixed maternal electrocardiogram signal sequence to be AECG = {x1, x2, ..., x...}. N Let the maternal electrocardiogram quadratic estimation sequence be MECG={y1,y2,...,y NThe new sequence, FECG, is obtained by subtracting the mixed maternal electrocardiogram (ECG) signal sequence from the secondary estimated maternal ECG sequence. This new sequence is: FECG = {(x1-y1),(x2-y2),...,(x...}. N -y N )}.

[0080] As an optional embodiment, in A8, the construction of the reconstruction matrix includes:

[0081] B1: The length of the RR interval estimated by the second estimate of the fetal electrocardiogram is estimated by R-peak detection. The length of 2q RR intervals is estimated continuously, and the length of every two adjacent RR intervals is recorded as a column period, which is then denoted as l1, l2, ..., l q ;

[0082] B2: Continuously capture 2q segments of fetal electrocardiogram data, record the length of the RR interval between every two segments, and select the longest length l. max =max(l1,l2,…,l q ) as the reconstruction matrix A f The number of columns;

[0083] B3: If matrix A f The length l of the ECG data in the k-th row (1≤k≤2q) k equal to l max Let other lengths be less than l max ECG data is padded with zeros to make its length 1. max To obtain a period of l max The electrocardiogram signal was used to construct a 2q×l max Reconstruction matrix A f ,have:

[0084]

[0085] B4: After singular value decomposition, retain the largest singular value σ1 and set the remaining singular values ​​to 0 to obtain matrix A. f The estimates are as follows:

[0086]

[0087] u1 and All are singular value parameters;

[0088] B5: Discard the data corresponding to the positions where zeros were originally padded in the matrix, and take out the remaining data from top to bottom, row by row, and rearrange them into a one-dimensional signal to obtain the 2q segment fetal electrocardiogram signal after noise removal;

[0089] B6: Repeat steps B1 to B5 above, processing 2q segments of ECG data each time, thus achieving denoising of all fetal ECG signals and obtaining fetal ECG signals with high signal-to-noise ratio.

[0090] In general, the process involves detecting the R-peak in the secondary estimation of the fetal electrocardiogram (ECG), interpolating the two consecutive RR intervals by padding with zeros at the end, and constructing a fetal ECG signal reconstruction matrix A. f Then, perform the corresponding SVD operation on it, retain the larger singular values ​​for noise reduction, and obtain A′. f Extract matrix A′ f The data corresponding to the positions where zeros were originally padded are used to obtain a fetal electrocardiogram signal with a high signal-to-noise ratio.

[0091] As an optional embodiment, the method for ECG quality detection is as follows:

[0092] C1: Acquire the electrocardiogram signal sequence of the first time period of a preset length;

[0093] C2: Perform phase space reconstruction on the electrocardiogram (ECG) signal sequence, and calculate the ECG signal based on the number of occurrences of various permutations to obtain the permutation entropy;

[0094] C3: Obtain the entropy threshold, and calculate the quality index of the electrocardiogram signal based on the permutation entropy and the entropy threshold;

[0095] C4: Obtain the quality threshold, determine whether the quality index meets the standard based on the quality threshold, if yes, obtain the ECG signal corresponding to the quality index and add the ECG signal to the target signal group; if not, return to step C1, update the entropy threshold based on the quality index, and collect ECG signals from adjacent time periods for phase space reconstruction until the quality index meets the quality threshold, obtain the ECG signal corresponding to the quality index and add the ECG signal to the target signal group.

[0096] Furthermore, it is necessary to calculate the quality ratio of permutation entropy to entropy threshold based on a preset algorithm, and finally determine the quality index of the electrocardiogram signal.

[0097] Furthermore, it is necessary to calculate the current quality weight of the electrocardiogram signal based on the quality index and the entropy threshold, calculate the update coefficient, and update the entropy threshold to obtain the updated entropy threshold.

[0098] As an optional embodiment, in C2, the method for obtaining the permutation entropy includes setting a one-dimensional time series: {Y(i), i = 1, 2, ..., n} with length n, reconstructing the phase space for any element Y(i), and according to the phase space reconstruction delay coordinate method proposed by Takens, that is, taking m consecutive samples for each sampling point and delaying the coordinates to obtain a reconstruction vector group in m-dimensional space:

[0099] Y i =[y(i),y(i+l),...,y(i+(m-1)l)]

[0100] In the formula: m and l are the embedding dimension and delay time, respectively, and Y i Arrange the m reconstructed components in ascending order to obtain:

[0101] y(i+(j1-1)l)≤y(i+(j2-1)l)≤...≤y(i+(j m -1)l)

[0102] The obtained vector Y i The arrangement is {j1,j2,...,j...} m The occurrence frequency of each possible arrangement in the entire sequence is statistically analyzed, and the relative frequency of each arrangement is calculated as its probability P1, P2, ..., P. k , k≤m!, calculate the permutation entropy using Shannon's definition of information entropy:

[0103]

[0104] To obtain the permutation entropy.

[0105] An apparatus for a high signal-to-noise ratio fetal electrocardiogram separation acquisition method, comprising:

[0106] The signal acquisition module is used to acquire preprocessed mixed maternal electrocardiogram signals;

[0107] The maternal electrocardiogram (ECG) acquisition module is used to extract the preprocessed mixed maternal ECG signal to obtain the maternal ECG signal.

[0108] The first detection module is used to detect the maternal electrocardiogram (ECG) signal and obtain a highly accurate maternal ECG signal.

[0109] The fetal electrocardiogram (ECG) separation module is used to separate the fetal ECG signal from the preprocessed mixed maternal ECG signal by combining the highly accurate maternal ECG signal.

[0110] The second detection module is used to detect fetal electrocardiogram signals and obtain highly accurate fetal electrocardiogram signals.

[0111] The DSP processing unit is used to receive high-accuracy fetal electrocardiogram (ECG) signals and output high signal-to-noise ratio fetal ECG signals.

[0112] As an optional embodiment, the signal acquisition module includes:

[0113] The electrode detachment detection module is used to detect whether the electrode pads have detached. If they have detached, the mixed maternal electrocardiogram signal is collected.

[0114] The signal conditioning module is used to amplify, filter, and perform other preprocessing on the acquired mixed maternal electrocardiogram signals.

[0115] By preprocessing the mixed maternal ECG signals, acquiring maternal ECG signals from the preprocessed mixed maternal ECG signals, and performing quality detection on the maternal ECG signals to obtain highly accurate maternal ECG signals, and then removing highly accurate maternal ECG signals from the preprocessed mixed maternal ECG signals to obtain fetal ECG signals, and performing quality detection on the fetal ECG signals to obtain highly accurate fetal ECG signals, the signal is received and digitally converted by the DSP processing unit to output a high signal-to-noise ratio fetal ECG signal. This solves the problem that fetal ECG signals are easily interfered with by maternal ECG signals, resulting in low accuracy of fetal ECG monitoring results.

[0116] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention (including the claims) is limited to these examples; within the framework of the invention, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of the different aspects of the invention as described above, which are not provided in the details for the sake of brevity.

[0117] This invention is intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for obtaining fetal electrocardiograms with high signal-to-noise ratio, characterized in that, include: A1: Preprocessing of mixed maternal electrocardiogram signals; A2: R-peak detection is performed on the preprocessed mixed maternal ECG to construct a preliminary estimate of the maternal ECG; A3: Collect the ECG signal sequence of the first time period in the preliminary estimation of the maternal ECG and perform ECG signal quality detection to obtain the maternal ECG quality index; A4: Compare the maternal ECG quality index with the maternal ECG quality threshold to determine if the condition is met. If yes, obtain the ECG signal corresponding to the maternal ECG quality index, add the ECG signal to the target signal group, and construct the secondary estimate of the maternal ECG. If no, return to step A3, collect the ECG signal sequence of adjacent time periods in the preliminary estimate of the maternal ECG for detection until the condition is met, obtain the ECG signal corresponding to the maternal ECG quality index, add the ECG signal to the target signal group, and construct the secondary estimate of the maternal ECG. A5: The difference between the secondary estimate of the maternal electrocardiogram and the preprocessed mixed maternal electrocardiogram signal is calculated, and the calculation result is set as the preliminary estimate of the fetal electrocardiogram; A6: Collect the ECG signal sequence of the first time period in the preliminary estimate of the fetal ECG and perform ECG signal quality detection to obtain the fetal ECG quality index; A7: Compare the fetal electrocardiogram quality index with the fetal electrocardiogram quality threshold to determine whether the conditions are met. If yes, obtain the electrocardiogram signal corresponding to the fetal electrocardiogram quality index, add the electrocardiogram signal to the target signal group, and construct the fetal electrocardiogram secondary estimation. If no, return to step A6, collect the electrocardiogram signal sequence of adjacent time periods in the preliminary fetal electrocardiogram estimation for detection until the conditions are met, obtain the electrocardiogram signal corresponding to the fetal electrocardiogram quality index, add the electrocardiogram signal to the target signal group, and construct the fetal electrocardiogram secondary estimation. A8: Construct a reconstruction matrix for the secondary estimation of the fetal electrocardiogram to obtain a fetal electrocardiogram signal with a high signal-to-noise ratio.

2. The method for obtaining fetal electrocardiograms with high signal-to-noise ratio according to claim 1, characterized in that, In A1, a 50Hz digital notch filter and a 0.05~100Hz bandpass filter are used to complete the preprocessing of the mixed maternal electrocardiogram signal.

3. The method for obtaining fetal electrocardiograms with high signal-to-noise ratio according to claim 1, characterized in that, In A2, the method for detecting the R-peak includes detecting the R-peak on the maternal abdominal electrocardiogram. Let the length be T. The R-peak is located within a window of length T, where T approximates an RR interval. By calculating the phase, the RR interval is mapped to the phase domain, taking the midpoint of the RR interval as the starting point. The formula for the phase interval is: ; The sampling interval is... For RR interval; The template signal is obtained using the arithmetic mean method. Let all QRS waves be as follows: The formula for reconstructing the MECG template is: ; The average duration of the maternal QRS complex was obtained by averaging the total number of time-wound beats to construct a preliminary estimate of the maternal electrocardiogram.

4. The method for obtaining fetal electrocardiograms with high signal-to-noise ratio according to claim 1, characterized in that, In A5, the difference calculation includes setting N to be the length of the sequence, and setting the preprocessed mixed maternal electrocardiogram signal sequence as... Let the maternal electrocardiogram quadratic estimation sequence be... A new sequence, namely the preliminary fetal electrocardiogram (ECG) estimation sequence, is obtained by subtracting the mixed maternal ECG signal sequence from the secondary estimated maternal ECG sequence. .

5. The method for obtaining fetal electrocardiograms with high signal-to-noise ratio according to claim 1, characterized in that, In A8, the construction of the reconstruction matrix includes: B1: The length of the RR interval estimated by the second estimate of the fetal electrocardiogram is estimated by R-peak detection. The length of 2q RR intervals is estimated continuously, and the length of every two adjacent RR intervals is recorded as a column period, and so on. ; B2: Continuously capture 2q segments of fetal electrocardiogram data, record the length of the RR interval between every two segments, and select the longest length. As a reconstruction matrix The number of columns; B3: If the matrix row k ECG data length equal Let other lengths be less than ECG data is lengthened by padding with zeros. To obtain the period as The electrocardiogram signal was used to construct a Reconstruction matrix ,have: ; B4: After singular value decomposition, the largest singular value is retained. Set the remaining singular values ​​to 0 to obtain the matrix. The estimates are as follows: ; and All are singular value parameters; B5: Discard the data corresponding to the positions where zeros were originally padded in the matrix, and take out the remaining data from top to bottom, row by row, and rearrange them into a one-dimensional signal to obtain the 2q segment fetal electrocardiogram signal after noise removal; B6: Repeat steps B1 to B5 above, processing 2q segments of ECG data each time, thus achieving denoising of all fetal ECG signals and obtaining fetal ECG signals with high signal-to-noise ratio.

6. The method for obtaining fetal electrocardiograms with high signal-to-noise ratio according to claim 1, characterized in that, The method for detecting the quality of the electrocardiogram signal is as follows: C1: Acquire the electrocardiogram signal sequence of the first time period of a preset length; C2: Perform phase space reconstruction on the electrocardiogram (ECG) signal sequence, and calculate the ECG signal based on the number of occurrences of various permutations to obtain the permutation entropy; C3: Obtain the entropy threshold, and calculate the quality index of the electrocardiogram signal based on the permutation entropy and the entropy threshold; C4: Obtain the quality threshold, determine whether the quality index meets the standard based on the quality threshold, if yes, obtain the ECG signal corresponding to the quality index and add the ECG signal to the target signal group; if not, return to step C1, update the entropy threshold based on the quality index, and collect ECG signals from adjacent time periods for phase space reconstruction until the quality index meets the quality threshold, obtain the ECG signal corresponding to the quality index and add the ECG signal to the target signal group.

7. The method for obtaining fetal electrocardiograms with high signal-to-noise ratio according to claim 6, characterized in that, In C2, the method for obtaining the permutation entropy includes setting a one-dimensional time series: Given a length of n, for any element Phase space reconstruction is performed using the phase space reconstruction delay coordinate method proposed by Takens. This involves taking m consecutive samples for each sampling point, delaying their coordinates, and obtaining a reconstruction vector set in m-dimensional space. ; In the formula: and These represent the embedding dimension and latency, respectively. of Arrange the reconstructed components in ascending order to obtain: ; The resulting vector The arrangement is as follows The frequency of each possible arrangement in the entire sequence is statistically analyzed, and the relative frequency of each arrangement is calculated as its probability. The permutation entropy is calculated using Shannon's definition of information entropy: ; to obtain the permutation entropy.

8. An apparatus employing the high signal-to-noise ratio fetal electrocardiogram separation and acquisition method according to any one of claims 1-6, characterized in that, include: The signal acquisition module is used to acquire preprocessed mixed maternal electrocardiogram signals; The maternal electrocardiogram (ECG) acquisition module is used to acquire the preprocessed mixed maternal ECG signal to obtain the maternal ECG signal. The first quality detection module is used to detect the maternal electrocardiogram (ECG) signal and obtain a highly accurate maternal ECG signal. The fetal electrocardiogram (ECG) separation module is used to separate the fetal ECG signal from the preprocessed mixed maternal ECG signal by combining the highly accurate maternal ECG signal. The second quality detection module is used to detect fetal electrocardiogram signals and obtain highly accurate fetal electrocardiogram signals. The DSP processing unit is used to receive high-accuracy fetal electrocardiogram (ECG) signals and output high signal-to-noise ratio fetal ECG signals.

9. The apparatus for a high signal-to-noise ratio fetal electrocardiogram separation and acquisition method according to claim 8, characterized in that, The signal acquisition module includes: The electrode detachment detection module is used to detect whether the electrode pads have detached. If they have detached, the mixed maternal electrocardiogram signal is collected. The signal conditioning module is used to amplify and filter the acquired mixed maternal electrocardiogram signals.

Citation Information

Patent Citations

  • System for monitoring heart rate of fetus in mother body

    CN108013872A

  • Separating mixed signals

    US20070027396A1