A method and system for monitoring uterine contraction based on multi-dimensional information fusion

Through the six-axis sensor fusion of acceleration and angular velocity signals, combined with MECG and EHG signal separation technology, the target contraction curve is generated, which solves the monitoring problems in noise interference and dynamic environments in contraction monitoring, and achieves efficient contraction motion monitoring.

CN114795253BActive Publication Date: 2024-11-26THE GBA NAT INST FOR NANOTECHNOLOGY INNOVATION
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210427884.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-04-22
Publication Date
2024-11-26
Estimated Expiration
2042-04-22

AI Technical Summary

Technical Problem

Existing uterine contraction monitoring methods are difficult to suppress the noise interference of uterine myoelectric signal, especially in dynamic environments, which are difficult to effectively monitor uterine contraction movement.

Method used

A six-axis sensor is used to fuse the three-axis acceleration signal and angular velocity signal to perform uterine feature segment marking and curve construction, combining MECG signal separation and EHG characteristic signal segment marking to generate a target uterine contraction curve to suppress noise interference.

Benefits of technology

It effectively avoids gravity acceleration and posture interference, enhances the dimension of contraction motion monitoring, and can accurately monitor contraction motion in static and dynamic environments, improving the monitoring effect.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114795253B_ABST
    Figure CN114795253B_ABST
Patent Text Reader

Abstract

The present invention discloses a uterine contraction monitoring method and system based on multi-dimensional information fusion. On the one hand, a six-axis sensor is used to analyze the movement state of the pregnant woman's abdomen during uterine contraction and draw a uterine contraction movement curve, which can avoid the interference of gravity acceleration caused by the pregnant woman's movements and the interference of different postures on the extraction of uterine electromyographic signal features. Uterine contraction movements can also be detected in a dynamic environment, which increases the dimension of monitoring uterine contraction and effectively expands the use scenarios. On the other hand, the mixed bioelectric signal is first separated by MECG signal, and then the EHG characteristic signal segment is marked to extract the EHG signal, which can effectively suppress the interference of MECG signals and the interference caused by the motion artifacts of the sensor, solving the technical problems that the existing uterine contraction movement monitoring methods are difficult to suppress the noise interference of uterine electromyographic signals and effectively monitor uterine contraction movements in both static and dynamic environments.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of uterine contraction monitoring, and in particular to a uterine contraction monitoring method and system based on multi-dimensional information fusion. Background Art

[0002] Uterine contraction refers to the regular contraction of the uterus, which is an important monitoring indicator for clinical prenatal examinations. Electrodes placed on the abdomen of pregnant women can collect uterine electromyogram (EHG) signals during uterine contractions, but this signal will be affected by noise interference such as maternal electrocardiogram (MECG) signals and sensor motion artifacts. During dynamic monitoring, it will also be interfered by the electromyographic signals generated by the movements of pregnant women, making it difficult to effectively monitor uterine contractions.

[0003] The existing uterine contraction monitoring method is to use a three-axis acceleration sensor in conjunction with the RMS algorithm for monitoring. The RMS algorithm is currently one of the most commonly used methods for extracting uterine electromyographic signals. Its principle is to reflect the energy change trend of the entire EHG signal burst segment and non-burst segment signal by calculating the standard deviation of uterine electromyography. Although this method can obtain a relatively smooth EHG waveform, which is helpful for the identification of uterine contraction waves, it is easily interfered by sensor motion artifacts, thereby affecting the monitoring effect of uterine contraction movement, and the three-axis acceleration sensor cannot obtain body position signal information in a dynamic situation, and it is difficult to solve the problem of interference from the electromyographic signal generated by the movement of pregnant women.

[0004] In the current uterine contraction movement monitoring technology, how to suppress the noise interference of uterine myoelectric signals and effectively monitor uterine contraction movements in both static and dynamic environments are still technical problems that need to be urgently solved by technical personnel in this field. Summary of the invention

[0005] The present invention provides a uterine contraction monitoring method and system based on multi-dimensional information fusion, which is used to solve the technical problem that the existing uterine contraction movement monitoring method is difficult to suppress the noise interference of uterine myoelectric signals and effectively monitor uterine contraction movements in both static and dynamic environments.

[0006] In view of this, a first aspect of the present invention provides a uterine contraction monitoring method based on multi-dimensional information fusion, comprising:

[0007] The three-axis acceleration signal and the three-axis angular velocity signal of the pregnant woman's abdomen obtained by the six-axis sensor are fused to obtain a three-axis linear acceleration signal;

[0008] Marking the uterine contraction characteristic segments of the three-axis linear acceleration signal to determine a set of uterine contraction time points;

[0009] Constructing a uterine contraction motion curve according to the three-axis linear acceleration signal and the set of uterine contraction time points;

[0010] Perform MECG signal separation on the mixed bioelectric signals obtained from the pregnant woman's abdomen to obtain intermediate state signals;

[0011] Marking the intermediate state signal with an EHG characteristic signal segment, and drawing an EHG curve according to the marked EHG characteristic signal segment;

[0012] The uterine contraction curve and the EHG curve are curve-fitted to generate a target uterine contraction curve.

[0013] Optionally, the target contraction curve is:

[0014]

[0015] Among them, C TOCO is the target uterine contraction curve, C EHG is the EHG curve, V is the column vector representing the EHG signal intensity, C ACC is the uterine contraction movement curve, and δ is the maximum displacement change of the abdominal surface skin relative to the trunk in the static state during uterine contraction.

[0016] Optionally, marking the uterine contraction characteristic segment of the three-axis linear acceleration signal to determine the uterine contraction time point includes:

[0017] Marking the signal segments that meet the preset constraint conditions in the three-axis linear acceleration signal as uterine contraction characteristic segments;

[0018] The preset constraints are:

[0019]

[0020] Among them, t∈T, T satisfies is a set of time points, m is the number of elements in T, v min is the minimum contraction velocity, t min is the shortest uterine contraction time, t max The longest uterine contraction time, is the three-axis linear acceleration, is the intensity of uterine contraction, δ is the maximum displacement change of the abdominal surface skin relative to the trunk in the static state during uterine contraction;

[0021] Get the set of uterine contraction time points corresponding to the uterine contraction feature segment.

[0022] Optionally, the uterine contraction movement curve is:

[0023]

[0024] Among them, C ACCis the uterine contraction curve, T α is the set of uterine contraction time points consisting of uterine contraction time points, T α ∈T.

[0025] Optionally, the minimum contraction speed is 0.1 m / s, the maximum displacement change of the abdominal surface skin relative to the trunk in a static state during contraction is 0.04 m, the shortest contraction time is 20 s, and the longest contraction time is 180 s.

[0026] Optionally, the mixed bioelectric signal acquired from the abdomen of the pregnant woman is subjected to MECG signal separation to obtain an intermediate state signal, including:

[0027] Obtaining a mixed bioelectric signal S from the abdomen of a pregnant woman;

[0028] The Pan&Tompkins algorithm is used to detect the R wave site of MECG in the mixed bioelectric signal S;

[0029] Mark the R wave location and corresponding signal intensity to generate a single QRS waveform;

[0030] The mixed bioelectric signal S is linearly subtracted from the generated single QRS waveform over the entire time period to obtain the intermediate state signal S f .

[0031] Optionally, mark the R wave location and the corresponding signal strength to generate a single QRS waveform, including:

[0032] Mark the R wave site;

[0033] Taking the R wave site as the center, the QRS wave sampling points are respectively shifted to the left by a preset number of sampling points and to the right by a preset number of sampling points, and a QRS matrix is ​​constructed according to the QRS wave sampling points and the signal strengths corresponding to the QRS wave sampling points;

[0034] Perform singular value decomposition on the QRS matrix to obtain the QRS wave template matrix;

[0035] Extract the column vector of the QRS wave template matrix to obtain a single QRS waveform.

[0036] Optionally, marking the intermediate state signal with an EHG characteristic signal segment, and drawing an EHG curve according to the marked EHG characteristic signal segment, comprises:

[0037] Use a preset length Hanning window to filter the intermediate signal S f Perform short-time Fourier transform to obtain a complex signal matrix, in which the column vectors are the signal strengths of each frequency;

[0038] Take out the element set C at the EHG characteristic frequency from the complex signal matrix f ;

[0039] The elements at the EHG characteristic frequency are combined into C f Subtract the corresponding EHG frequency characteristic threshold to obtain the characteristic frequency matrix W;

[0040] The elements of each row in the characteristic frequency matrix W are summed to obtain the column vector V representing the EHG signal strength. n is the total number of sampling points;

[0041] Take the column vector V whose elements are greater than 0 and whose continuous time is greater than t min Less than t max The subscript set of is I, where I is the set of sampling points marked as EHG characteristic signal segments;

[0042] Draw the EHG curve. The EHG curve is:

[0043]

[0044] Among them, the value of i is [1,n].

[0045] Optionally, the number of EHG characteristic frequencies is 2, 3, 4 or 5.

[0046] A second aspect of the present invention provides a uterine contraction monitoring system based on multi-dimensional information fusion, comprising:

[0047] The linear acceleration acquisition module is used to fuse the three-axis acceleration signal and the three-axis angular velocity signal of the pregnant woman's abdomen obtained by the six-axis sensor to obtain a three-axis linear acceleration signal;

[0048] A uterine contraction feature marking module is used to mark the uterine contraction feature segment of the three-axis linear acceleration signal and determine the uterine contraction time point set;

[0049] A uterine contraction motion curve module, used to construct a uterine contraction motion curve according to the three-axis linear acceleration signal and the uterine contraction time point set;

[0050] The MECG signal separation module is used to perform MECG signal separation on the mixed bioelectric signals acquired from the abdomen of the pregnant woman to obtain intermediate state signals;

[0051] An EHG curve module, used for marking the EHG characteristic signal segments of the intermediate state signal and drawing the EHG curve according to the marked EHG characteristic signal segments;

[0052] The uterine contraction curve module is used to perform curve fitting on the uterine contraction motion curve and the EHG curve to generate a target uterine contraction curve. From the above technical solutions, it can be seen that the uterine contraction monitoring method and system based on multi-dimensional information fusion provided by the present invention has the following advantages:

[0053] The uterine contraction monitoring method and system based on multi-dimensional information fusion provided by the present invention, on the one hand, use a six-axis sensor to analyze the movement state of the pregnant woman's abdomen during uterine contractions and draw a uterine contraction movement curve, which can avoid the interference of gravity acceleration caused by the pregnant woman's movements and the interference of different postures on the extraction of uterine electromyographic signal features. Uterine contraction movements can also be detected in a dynamic environment, which increases the dimension of monitoring uterine contractions and effectively expands the usage scenarios. On the other hand, the mixed bioelectric signal is first separated by MECG signals, and then the EHG characteristic signal segments are marked to extract the EHG signal, which can effectively suppress the interference of MECG signals and the interference caused by the motion artifacts of the sensor, solving the technical problems that the existing uterine contraction movement monitoring methods are difficult to suppress the noise interference of uterine electromyographic signals and effectively monitor uterine contraction movements in both static and dynamic environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] 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.

[0055] Figure 1 A schematic flow chart of a uterine contraction monitoring method based on multi-dimensional information fusion provided in the present invention;

[0056] Figure 2 It is a structural schematic diagram of a uterine contraction monitoring system based on multi-dimensional information fusion provided in the present invention. DETAILED DESCRIPTION

[0057] 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.

[0058] For easier understanding, see Figure 1 The present invention provides an embodiment of a uterine contraction monitoring method based on multi-dimensional information fusion, comprising:

[0059] Step 101: fuse the three-axis acceleration signal and the three-axis angular velocity signal of the pregnant woman's abdomen obtained by the six-axis sensor to obtain a three-axis linear acceleration signal.

[0060] It should be noted that in the embodiment of the present application, the six-axis sensor includes an acceleration sensor and a gyroscope sensor. The three-axis acceleration signal is obtained by the acceleration sensor, and the three-axis angular velocity signal is obtained by the gyroscope sensor. The Euler angle data obtained by fusing the three-axis acceleration signal and the three-axis angular velocity signal is: Among them, p, r, and y correspond to the pitch angle, roll angle, and yaw angle, respectively. In this posture, the component of gravity acceleration acting on the acceleration sensor is g is the acceleration due to gravity. Subtract the three-axis acceleration signal measured by the acceleration sensor from Get three-axis linear acceleration signal That is, the three-axis acceleration that eliminates the interference of gravity acceleration.

[0061] Step 102: Mark the uterine contraction characteristic segments of the three-axis linear acceleration signal to determine a set of uterine contraction time points.

[0062] It should be noted that the main characteristics of uterine contractions on the skin surface are: (1) the speed of uterine contractions is less than the minimum speed of uterine contractions; (2) the displacement of the abdominal skin surface relative to the trunk of the pregnant woman in a static state is greater than 0 and less than the maximum displacement change of the abdominal skin surface relative to the trunk in a static state during uterine contractions; (3) the duration of a single uterine contraction is greater than the shortest uterine contraction time and less than the longest uterine contraction time. Movements that meet the above three conditions are characterized as uterine contraction intensity and are marked as uterine contraction characteristic segments.

[0063] Converted into mathematical expression:

[0064]

[0065] Among them, t∈T, T satisfies is a set of time points, m is the number of elements in T, v min is the minimum contraction velocity, t min is the shortest uterine contraction time, t max The longest contraction time, is the three-axis linear acceleration, is the intensity of uterine contraction, and δ is the maximum displacement change of the abdominal surface skin relative to the trunk in a static state during uterine contraction.

[0066] In the embodiment of the present invention, the minimum contraction speed is preferably 0.1 m / s, the maximum displacement change of the abdominal surface skin relative to the trunk in a static state during contraction is preferably 0.04 m, the shortest contraction time is preferably 20 s, and the longest contraction time is preferably 180 s. In the algorithm, the three-axis linear acceleration is expressed as:

[0067]

[0068] Wherein, x, y, and z represent the linear acceleration in the x-axis, y-axis, and z-axis directions, respectively.

[0069] The above three conditions can be converted to:

[0070] (1) Uterine contraction intensity The integral over time is <0.1m / s;

[0071] (2) Contraction intensity The time length of the second integral from 0 to 0 is >20s and <180s;

[0072] (3) Uterine contraction intensity The quadratic integral in time is less than 4 cm.

[0073] The specific signal processing flow can be described as:

[0074] There exists a set T, t∈T, where t is time, such that

[0075] Then T={t1,t2,…,t l}, l is the number of elements in T, t l is the lth time point.

[0076] There exists a set T α , T α ∈T, such that T α The elements in satisfy:

[0077]

[0078] Then, the set T α ={t 1α ,t 2α ,…,t kα The elements in} are the contraction time points, k is T α The number of elements in .

[0079] Step 103: construct a uterine contraction motion curve according to the three-axis linear acceleration signal and the uterine contraction time point set. ACC According to the three-axis linear acceleration signal and the set of contraction time points T consisting of contraction time points α Constructed:

[0080]

[0081] Step 104: perform MECG signal separation on the mixed bioelectric signal acquired from the pregnant woman's abdomen to obtain an intermediate state signal.

[0082] It should be noted that the mixed bioelectric signal S of the pregnant woman's abdomen is obtained by collecting electrodes, and the MECG signal S is obtained from the mixed bioelectric signal S. t Separate it and get the MECG signal S minus t The mixed bioelectric signal, that is, the intermediate state signal S f .

[0083] Specifically, the Pan & Tompkins algorithm is used to detect the R wave site of the MECG in the mixed bioelectric signal S, mark the R wave site and the corresponding signal strength, generate a single QRS waveform, and linearly subtract the mixed bioelectric signal S from the generated single QRS waveform over the entire time period to obtain the intermediate state signal S f The signal processing process is:

[0084] (1) Mark the R wave. Use an analog-to-digital converter to sample the mixed bioelectric signal S and generate a discrete time series signal, which is recorded as:

[0085]

[0086] Where n is the total number of sampling points and the sampling frequency is f s , unit is Hz.

[0087] Generate vector using Pan & Tompkins:

[0088]

[0089] Where K is the number of R waves in the mixed bioelectric signal S, t1~t K is the time point of the R wave peak, K <n。

[0090] Let α be the left offset sampling point of the QRS wave centered on the R wave, and β be the right offset sampling point of the QRS wave centered on the R wave. The matrix I A is the sampling point matrix of the QRS wave, I A It is expressed as:

[0091]

[0092] (2) Generate the QRS matrix. According to the vector T R Generate QRS matrix A S :

[0093]

[0094] Let element a satisfy a q =s q , if q∈I A And q∈{1,2,…,n}.

[0095] Among them, element a∈S,A S A is a matrix of size (α+β+1)×K, S The single column vector is the sampling signal of a single QRS wave.

[0096] α and β take empirical values: α=0.25f s , β=0.45f s .

[0097] (3) Construct the QRS wave template matrix. S For a matrix of size (α+β+1)×K, we can find UYV T ≈A S , U is A S The left singular value vector of A is a matrix of size (α+β+1)×J, where J is the number of singular value vectors and its empirical value is 2. S The singular value matrix of A is J×J, and V is S The right singular value vector of is a matrix of size K×J.

[0098] We can obtain: A t The elements in are denoted by b, and they are:

[0099]

[0100] Matrix A t The size of is (α+β+1)×K.

[0101] Matrix A t The column vector in the signal S corresponds to the template generated by the pregnant woman's ECG QRS wave. This template can be used to construct the vector S t :

[0102]

[0103] (4) Substitute the signal S with the vector S t Subtract and get S f , S f That is, the electrical signal of the pregnant woman's electrocardiogram MECG is removed, and the signal can be defined as an intermediate state signal.

[0104] If the signal acquisition contains multiple channels, the same T R and S of different channels, repeat steps (1)-(4).

[0105] Step 105: Mark the EHG characteristic signal segments of the intermediate state signal, and draw an EHG curve according to the marked EHG characteristic signal segments.

[0106] It should be noted that the electrical signal S of the MECG of the pregnant woman’s ECG was removed.f The main interferences are sensor motion artifacts, fetal ECG signal interference, uterine myoelectric signals, and sensor internal noise. The simplest and most obvious feature of uterine myoelectric signals and other interferences is the frequency feature. Therefore, multiple frequency feature points can be selected as reference points to mark the EHG feature signal segment.

[0107] Specifically, this embodiment uses short-time Fourier transform to extract the signal segment whose frequency characteristic point is greater than the threshold and mark it as the EHG characteristic signal segment. The signal processing flow is:

[0108] Use a Hanning window with a preset length N to filter the intermediate signal S f Perform short-time Fourier transform to obtain the complex signal matrix and frequency column vector.

[0109] The complex signal matrix is ​​denoted by C:

[0110]

[0111] The frequency column vector is denoted by F:

[0112]

[0113] Among them, the frequency column vector F is the frequency vector corresponding to the complex signal C, E is the number of frequencies, satisfying the Nyquist theorem, and n is the total number of points used.

[0114] A preset number (preferably 2 or 3 or 4 or 5) of signal segments with frequency feature points greater than the frequency feature threshold are taken out from the frequency column vector, and recorded as:

[0115]

[0116] Set the frequency feature threshold:

[0117]

[0118] Get the frequency feature threshold matrix:

[0119]

[0120] Find the corresponding frequency eigenvector F in matrix C f The point modulus is obtained, and the set C is obtained. f :

[0121]

[0122] Subtract the frequency characteristic value of the EHG characteristic signal segment from the frequency characteristic threshold to obtain the characteristic frequency matrix W:

[0123]

[0124] Sum the elements of each row in the characteristic frequency matrix W to obtain the column vector V representing the EHG signal strength:

[0125]

[0126] Among them, the number of consecutive subscripts of elements in V that are greater than zero is greater than 20f s And less than 180f s The set is I, where I is marked as the set of sampling points of the EHG characteristic signal segment.

[0127] Use column vector V to draw the EHG curve and set zeros in the unmarked area. The EHG curve is:

[0128]

[0129] Among them, the value of i is [1,n].

[0130] It should be noted that in the embodiment of the present application, steps 101-103 can be performed simultaneously with steps 104-105, and there is no execution time sequence relationship between steps 101-103 and steps 104-105, that is, steps 104-105 can be executed simultaneously when steps 101-103 are executed, or steps 101-103 can be executed first and then steps 104-105.

[0131] Step 106: Perform curve fitting on the uterine contraction curve and the EHG curve to generate a target uterine contraction curve.

[0132] It should be noted that after obtaining the uterine contraction curve in step 103 and the EHG curve in step 105, the uterine contraction curve and the EHG curve are curve-fitted using a linear superposition method to generate a target uterine contraction curve:

[0133]

[0134] Among them, C TOCO is the target uterine contraction curve, C EHG is the EHG curve, V is the column vector representing the EHG signal intensity, C ACC is the uterine contraction movement curve, δ is the maximum displacement change of the abdominal surface skin relative to the trunk in the static state during uterine contraction, and the empirical value can be 0.04.

[0135] The uterine contraction monitoring method based on multi-dimensional information fusion provided by the embodiment of the present invention, on the one hand, uses a six-axis sensor to analyze the movement state of the pregnant woman's abdomen during uterine contractions and draws a uterine contraction movement curve, which can avoid the interference of gravity acceleration caused by the pregnant woman's movements and the interference of different postures on the extraction of uterine electromyographic signal features. Uterine contraction movements can also be detected in a dynamic environment, which increases the dimension of monitoring uterine contractions and effectively expands the usage scenarios. On the other hand, the mixed bioelectric signal is first separated from the MECG signal, and then the EHG characteristic signal segment is marked to extract the EHG signal, which can effectively suppress the interference of the MECG signal and the interference caused by the motion artifacts of the sensor, solving the technical problems that the existing uterine contraction movement monitoring methods are difficult to suppress the noise interference of uterine electromyographic signals and effectively monitor uterine contraction movements in both static and dynamic environments.

[0136] The present invention also provides an embodiment of a uterine contraction monitoring system based on multi-dimensional information fusion, comprising:

[0137] The linear acceleration acquisition module is used to fuse the three-axis acceleration signal and the three-axis angular velocity signal of the pregnant woman's abdomen obtained by the six-axis sensor to obtain a three-axis linear acceleration signal;

[0138] A uterine contraction feature marking module is used to mark the uterine contraction feature segment of the three-axis linear acceleration signal and determine the uterine contraction time point set;

[0139] A uterine contraction motion curve module, used to construct a uterine contraction motion curve according to the three-axis linear acceleration signal and the uterine contraction time point set;

[0140] The MECG signal separation module is used to perform MECG signal separation on the mixed bioelectric signals acquired from the abdomen of the pregnant woman to obtain intermediate state signals;

[0141] An EHG curve module, used for marking the EHG characteristic signal segments of the intermediate state signal and drawing the EHG curve according to the marked EHG characteristic signal segments;

[0142] The uterine contraction curve module is used to perform curve fitting on the uterine contraction movement curve and the EHG curve to generate a target uterine contraction curve.

[0143] The target contraction curve is:

[0144]

[0145] Among them, C TOCO is the target uterine contraction curve, C EHG is the EHG curve, V is the column vector representing the EHG signal intensity, C ACC is the uterine contraction movement curve, and δ is the maximum displacement change of the abdominal surface skin relative to the trunk in the static state during uterine contraction.

[0146] The uterine contraction feature marking module is specifically used for:

[0147] The signal segments that meet the preset constraint conditions in the three-axis linear acceleration signal are marked as uterine contraction characteristic segments; wherein the preset constraint conditions are:

[0148]

[0149] Among them, t∈T, T satisfies is a set of time points, m is the number of elements in T, v min is the minimum contraction velocity, t min is the shortest uterine contraction time, t max The longest contraction time, is the three-axis linear acceleration, is the intensity of uterine contraction, δ is the maximum displacement change of the abdominal surface skin relative to the trunk in the static state during uterine contraction;

[0150] Get the uterine contraction time point corresponding to the uterine contraction characteristic segment.

[0151] The uterine contraction curve is:

[0152]

[0153] Among them, C ACC is the uterine contraction curve, T α is the set of uterine contraction time points consisting of uterine contraction time points, T α ∈T.

[0154] The minimum contraction speed was 0.1 m / s, the maximum displacement change of the abdominal surface skin relative to the trunk in a static state during contractions was 0.04 m, the shortest contraction time was 20 s, and the longest contraction time was 180 s.

[0155] The mixed bioelectric signals obtained from the pregnant woman's abdomen are separated by MECG signals to obtain intermediate state signals, including:

[0156] Obtaining a mixed bioelectric signal S from the abdomen of a pregnant woman;

[0157] The Pan&Tompkins algorithm is used to detect the R wave site of MECG in the mixed bioelectric signal S;

[0158] Mark the R wave location and corresponding signal intensity to generate a single QRS waveform;

[0159] The mixed bioelectric signal S is linearly subtracted from the generated single QRS waveform over the entire time period to obtain the intermediate state signal S f .

[0160] Mark the R wave position and the corresponding signal strength to generate a single QRS waveform, including:

[0161] Mark the R wave site;

[0162] Taking the R wave site as the center, the QRS wave sampling points are respectively shifted to the left by a preset number of sampling points and to the right by a preset number of sampling points, and a QRS matrix is ​​constructed according to the QRS wave sampling points and the signal strengths corresponding to the QRS wave sampling points;

[0163] Perform singular value decomposition on the QRS matrix to obtain the QRS wave template matrix;

[0164] Extract the column vector of the QRS wave template matrix to obtain a single QRS waveform.

[0165] Marking the EHG characteristic signal segment of the intermediate state signal, and drawing the EHG curve according to the marked EHG characteristic signal segment, including:

[0166] Use a preset length Hanning window to filter the intermediate signal S f Perform short-time Fourier transform to obtain a complex signal matrix, in which the column vectors are the signal strengths of each frequency;

[0167] Take out the element set C at the EHG characteristic frequency from the complex signal matrix f ;

[0168] The elements at the EHG characteristic frequency are combined into C f Subtract the corresponding EHG frequency characteristic threshold to obtain the characteristic frequency matrix W;

[0169] The elements of each row in the characteristic frequency matrix W are summed to obtain the column vector V representing the EHG signal strength. n is the total number of sampling points;

[0170] Take the column vector V whose elements are greater than 0 and whose continuous time is greater than t min Less than t max The subscript set of is I, where I is the set of sampling points marked as EHG characteristic signal segments;

[0171] Draw the EHG curve. The EHG curve is:

[0172]

[0173] Among them, the value of i is [1,n].

[0174] The EHG characteristic frequencies are 2, 3, 4 or 5.

[0175] The uterine contraction monitoring system based on multi-dimensional information fusion provided by the embodiment of the present invention, on the one hand, uses a six-axis sensor to analyze the movement state of the pregnant woman's abdomen during uterine contractions and draws a uterine contraction movement curve, which can avoid the interference of gravity acceleration caused by the pregnant woman's movements and the interference of different postures on the extraction of uterine electromyographic signal features. Uterine contraction movements can also be detected in a dynamic environment, which increases the dimension of monitoring uterine contractions and effectively expands the usage scenarios. On the other hand, the mixed bioelectric signal is first separated by MECG signals, and then the EHG characteristic signal segments are marked to extract the EHG signal, which can effectively suppress the interference of MECG signals and the interference caused by the motion artifacts of the sensor, solving the technical problems that the existing uterine contraction movement monitoring methods are difficult to suppress the noise interference of uterine electromyographic signals and effectively monitor uterine contraction movements in both static and dynamic environments.

[0176] The uterine contraction monitoring system based on multi-dimensional information fusion provided in an embodiment of the present invention is used to execute the uterine contraction monitoring method based on multi-dimensional information fusion in the aforementioned embodiment of the uterine contraction monitoring method based on multi-dimensional information fusion. Its principle is the same as the uterine contraction monitoring method based on multi-dimensional information fusion in the aforementioned embodiment of the uterine contraction monitoring method based on multi-dimensional information fusion, and will not be repeated here.

[0177] 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 uterine contraction monitoring method based on multi-dimensional information fusion, It is characterized in that include: The three-axis acceleration signal and the three-axis angular velocity signal of the pregnant woman's abdomen obtained by the six-axis sensor are fused to obtain a three-axis linear acceleration signal; Marking the uterine contraction characteristic segments of the three-axis linear acceleration signal to determine a set of uterine contraction time points; Constructing a uterine contraction motion curve according to the three-axis linear acceleration signal and the set of uterine contraction time points; Perform MECG signal separation on the mixed bioelectric signals obtained from the pregnant woman's abdomen to obtain intermediate state signals; Marking the intermediate state signal with an EHG characteristic signal segment, and drawing an EHG curve according to the marked EHG characteristic signal segment; The uterine contraction curve and the EHG curve are curve-fitted to generate a target uterine contraction curve.

2. The uterine contraction monitoring method based on multi-dimensional information fusion according to claim 1, It is characterized in that The target contraction curve is: Among them, C TOCO is the target uterine contraction curve, C EHG is the EHG curve, V is the column vector representing the EHG signal intensity, C ACC is the uterine contraction movement curve, and δ is the maximum displacement change of the abdominal surface skin relative to the trunk in the static state during uterine contraction.

3. The uterine contraction monitoring method based on multi-dimensional information fusion according to claim 1, It is characterized in that Mark the uterine contraction characteristic segments of the three-axis linear acceleration signal to determine the uterine contraction time point, including: The signal segments that meet the preset constraint conditions in the three-axis linear acceleration signal are marked as uterine contraction characteristic segments; wherein the preset constraint conditions are: Among them, t∈T, T satisfies is a set of time points, m is the number of elements in T, v min is the minimum contraction velocity, t min is the shortest uterine contraction time, t max The longest uterine contraction time, is the three-axis linear acceleration, is the intensity of uterine contraction, δ is the maximum displacement change of the abdominal surface skin relative to the trunk in the static state during uterine contraction; Get the set of uterine contraction time points corresponding to the uterine contraction feature segment.

4. The uterine contraction monitoring method based on multi-dimensional information fusion according to claim 3, It is characterized in that The uterine contraction curve is: Among them, C ACC is the uterine contraction curve, T α is the set of uterine contraction time points consisting of uterine contraction time points, T α ∈T.

5. The uterine contraction monitoring method based on multi-dimensional information fusion according to claim 3, It is characterized in that The minimum contraction speed was 0.1 m / s, the maximum displacement change of the abdominal surface skin relative to the trunk in a static state during contractions was 0.04 m, the shortest contraction time was 20 s, and the longest contraction time was 180 s.

6. The uterine contraction monitoring method based on multi-dimensional information fusion according to claim 3, It is characterized in that The mixed bioelectric signals obtained from the pregnant woman's abdomen are separated by MECG signals to obtain intermediate state signals, including: Obtaining a mixed bioelectric signal S from the abdomen of a pregnant woman; The Pan&Tompkins algorithm is used to detect the R wave site of MECG in the mixed bioelectric signal S; Mark the R wave location and corresponding signal intensity to generate a single QRS waveform; The mixed bioelectric signal S is linearly subtracted from the generated single QRS waveform over the entire time period to obtain the intermediate state signal S f .

7. The uterine contraction monitoring method based on multi-dimensional information fusion according to claim 6, It is characterized in that Mark the R wave position and the corresponding signal strength to generate a single QRS waveform, including: Mark the R wave site; Taking the R wave site as the center, the QRS wave sampling points are respectively shifted to the left by a preset number of sampling points and to the right by a preset number of sampling points, and a QRS matrix is ​​constructed according to the QRS wave sampling points and the signal strengths corresponding to the QRS wave sampling points; Perform singular value decomposition on the QRS matrix to obtain the QRS wave template matrix; Extract the column vector of the QRS wave template matrix to obtain a single QRS waveform.

8. The uterine contraction monitoring method based on multi-dimensional information fusion according to claim 6, It is characterized in that Marking the EHG characteristic signal segment of the intermediate state signal, and drawing the EHG curve according to the marked EHG characteristic signal segment, including: Use a preset length Hanning window to filter the intermediate signal S f Perform short-time Fourier transform to obtain a complex signal matrix, in which the column vectors are the signal strengths of each frequency; Take out the element set C at the EHG characteristic frequency from the complex signal matrix f ; The elements at the EHG characteristic frequency are combined into C f Subtract the corresponding EHG frequency characteristic threshold to obtain the characteristic frequency matrix W; The elements of each row in the characteristic frequency matrix W are summed to obtain the column vector V representing the EHG signal strength. n is the total number of sampling points; Take the column vector V whose elements are greater than 0 and whose continuous time is greater than t min Less than t max The subscript set of is I, where I is the set of sampling points marked as EHG characteristic signal segments; Draw the EHG curve. The EHG curve is: Among them, the value of i is [1,n].

9. The uterine contraction monitoring method based on multi-dimensional information fusion according to claim 8, It is characterized in that The EHG characteristic frequencies are 2, 3, 4 or 5.

10. A uterine contraction monitoring system based on multi-dimensional information fusion, It is characterized in that include: The linear acceleration acquisition module is used to fuse the three-axis acceleration signal and the three-axis angular velocity signal of the pregnant woman's abdomen obtained by the six-axis sensor to obtain a three-axis linear acceleration signal; A uterine contraction feature marking module is used to mark the uterine contraction feature segment of the three-axis linear acceleration signal and determine the uterine contraction time point set; A uterine contraction motion curve module is used to construct a uterine contraction motion curve according to the three-axis linear acceleration signal and the uterine contraction time point set; The MECG signal separation module is used to perform MECG signal separation on the mixed bioelectric signals acquired from the abdomen of the pregnant woman to obtain intermediate state signals; An EHG curve module, used for marking the EHG characteristic signal segments of the intermediate state signal, and drawing the EHG curve according to the marked EHG characteristic signal segments; The uterine contraction curve module is used to perform curve fitting on the uterine contraction movement curve and the EHG curve to generate a target uterine contraction curve.