A data analysis-based fetal electrocardio monitoring system

Abstract

The application provides a fetal electrocardio monitoring system based on data analysis, and belongs to the technical field of electrocardio monitoring, and specifically comprises the following steps: collecting a mixed signal, removing power frequency interference by using an IIR notch filter, removing baseline drift by using a median filter, and removing artificial pulse interference by using empirical mode decomposition; extracting a maternal signal by using an independent component analysis method, constructing a singular value transfer matrix, taking the maternal abdominal wall electrocardio signal as spatial filtering, offsetting the maternal signal by singular value decomposition; extracting a fetal signal by using the independent component analysis method again; determining a fetal electrocardio R wave peak by using a time domain nonlinear transformation threshold method, performing clustering detection on the R wave peak, calculating a fetal heart rate according to the time difference of RR intervals; analyzing low frequency components LF and high frequency components HF in a maternal heart rate signal, calculating an LF / HF value of the maternal heart rate, and judging whether the maternal body has an anxiety emotion; and the application realizes more convenient and accurate fetal heart monitoring.

Description

A fetal heart rate monitoring system based on data analysis Technical Field

[0001] This invention relates to the field of electrocardiogram (ECG) monitoring technology, and more specifically to a fetal ECG monitoring system based on data analysis. Background Technology

[0002] Home-use portable fetal heart rate monitors do have some problems in the current market, the most prominent being the difficulty in accurately separating maternal and fetal heart rates from murmurs. This poses a significant challenge to monitoring the health of both the pregnant woman and the fetus. In recent years, with the advancement of medical research, screening for maternal mood changes and depression during pregnancy has been incorporated into routine prenatal checkups.

[0003] Current fetal heart rate monitors mainly collect signals through photoelectric sensors and microphones. However, due to signal interference and limitations in signal processing technology, the monitoring results are often affected by environmental noise and the mother's own heart rate, making it difficult to accurately separate the maternal and fetal heart rates.

[0004] On the other hand, prenatal depression screening also faces similarly serious challenges. Pregnancy depression has a significant impact on both maternal and infant health, making early detection and intervention crucial. However, current subjective screening methods are limited by the pregnant woman's own cognitive abilities and subjective feelings, lacking objective data support, which may lead to lower accuracy in screening results. To improve the objectivity and accuracy of screening... Summary of the Invention

[0005] The purpose of this invention is to provide a fetal heart rate monitoring system based on data analysis, and to solve the following technical problems:

[0006] Existing home fetal heart rate monitors are unable to accurately separate maternal and fetal heart rates and murmurs, and prenatal depression screening also faces significant challenges.

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] A fetal heart rate monitoring system based on data analysis, comprising:

[0009] A signal acquisition module is used to acquire mixed signals, which include maternal signals and fetal signals;

[0010] The signal filtering module is used to preprocess the mixed signal. The preprocessing includes using an IIR notch filter to remove power frequency interference, using a median filter to remove baseline drift, and using empirical mode decomposition to remove artificial impulse interference.

[0011] The signal separation module is used to extract maternal signals using independent component analysis, construct a singular value transfer matrix, use maternal abdominal wall electrocardiogram signals as spatial filters, and cancel maternal signals through singular value decomposition; then, independent component analysis is used again to extract fetal signals.

[0012] The heart rate calculation module is used to determine the fetal ECG R wave spikes using the time-domain nonlinear transformation threshold method, perform cluster detection on the R wave spikes, and then calculate the fetal heart rate based on the time difference of the RR intervals.

[0013] The maternal assessment module is used to analyze the low-frequency component (LF) and high-frequency component (HF) in the maternal heart rate signal using frequency domain analysis based on pyHRV, and calculate the LF / HF value of the maternal heart rate. When the LF / HF value is greater than 1.5, the mother is considered to have anxiety.

[0014] As a further aspect of the present invention: the maternal assessment module uses the open-source Python toolbox pyHRV to analyze maternal heart rate signals with heart rate indicators, including but not limited to calculating heart rate parameters through electrocardiogram, SpO2, blood volume pulse.

[0015] As a further aspect of the present invention: the specific process by which the signal separation module extracts fetal signals is as follows:

[0016] Singular value decomposition is performed on the mixed signal to obtain singular value vectors; the singular value vectors are arranged in descending order, and the left and right singular vectors corresponding to the first k singular values ​​are selected as spatial filters; the spatial filters are applied to the mixed signal to obtain the filtered fetal signal.

[0017] As a further aspect of the present invention: the wearable fetal heart rate monitor includes several MEMS sensing chips on flexible substrates, each of which is equipped with pressure-sensitive single-crystal silicon, and the position of the sensing chip is automatically corrected after the mother wears it.

[0018] As a further aspect of the present invention: the position correction process is as follows:

[0019] Each flexible bandage of the fetal heart rate monitor is equipped with two centrally symmetrical sensor chips. The signal intensities A1 and A2 measured by the two sensor chips in each flexible bandage are collected respectively. The distances d1 and d2 between the two sensor chips and the fetal heart rate are calculated according to the signal attenuation formula, which is:

[0020] A1 = A0·e^(-α·d1), A2= A0·e^(-α·d2);

[0021] Where A0 is the initial intensity of the fetal heart signal, e is the natural constant, α is the human body medium attenuation coefficient, and when the absolute value of the difference between d1 and d2 is greater than the preset threshold, the mother is reminded to adjust the position of the flexible bandage.

[0022] As a further aspect of the present invention, the specific adjustment process is as follows:

[0023] Based on the curvature change of the flexible bandage, the approximate arc formed by the flexible bandage is fitted to generate an ideal arc. The center position and diameter of the ideal circle containing the ideal arc are calculated, and the positions of distances d1 and d2 in the ideal circle are marked. A diameter line segment is drawn through the center position and the fetal heart position. The arc length L1 corresponding to the angle formed by distance d1 and the diameter line segment, and the arc length L2 corresponding to the angle formed by distance d2 and the diameter line segment are calculated respectively. When the absolute value of the difference between d1 and d2 is greater than a preset threshold, the position of the flexible bandage is abnormal. The value of |L1-L2| / 2 is calculated and marked as an adjustment distance value and sent to the fetal heart rate monitor display page.

[0024] As a further aspect of the present invention, it also includes a wireless transmission module for receiving and displaying monitoring results, including fetal heart rate, maternal mood, and the position of the flexible bandage, and transmitting the monitoring results to the devices of preset medical personnel and family members.

[0025] The beneficial effects of this invention are:

[0026] (1) This invention uses pyHRV to analyze the low-frequency and high-frequency components in the heart rate signal using frequency domain analysis and calculates LF / HF. The fetal heart rate monitor can provide a more sensitive assessment of anxiety status than the Hamilton Depression Rating Scale. In an anxious state, the maternal heart rate variability will decrease, the high-frequency part will decrease and the low-frequency part will increase relatively. Therefore, LF / HF can be used as one of the indicators for judging anxiety status. When the LF / HF value is greater than 1.5, it is considered that the mother has anxiety.

[0027] (2) This invention provides individualized assessment based on the heart rate characteristics of each mother. Heart rate variability may vary among different populations. When calculating the LF / HF value, pyHRV can customize the frequency band calculation parameters to provide heart rate analysis results for each mother, which helps to more accurately assess anxiety status.

[0028] (3) By monitoring the mother's heart rate LF / HF, the fetal heart rate monitor can provide timely information to help doctors or nurses take intervention measures. Once the monitoring results show that the mother may have anxiety, the doctor can take corresponding measures, such as providing psychological support, suggesting appropriate relaxation and adjustment methods, and arranging further evaluation or treatment. Attached Figure Description

[0029] The invention will now be further described with reference to the accompanying drawings.

[0030] Figure 1 is a schematic diagram of the process of the present invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] Please refer to Figure 1. This invention is a fetal heart rate monitoring system based on data analysis, comprising:

[0033] A signal acquisition module is used to acquire mixed signals, which include maternal signals and fetal signals;

[0034] The signal filtering module is used to preprocess the mixed signal. The preprocessing includes using an IIR notch filter to remove power frequency interference, using a median filter to remove baseline drift, and using empirical mode decomposition to remove artificial impulse interference.

[0035] The signal separation module is used to extract maternal signals using independent component analysis, construct a singular value transfer matrix, use maternal abdominal wall electrocardiogram signals as spatial filters, and cancel maternal signals through singular value decomposition; then, independent component analysis is used again to extract fetal signals.

[0036] The heart rate calculation module is used to determine the fetal ECG R wave spikes using the time-domain nonlinear transformation threshold method, perform cluster detection on the R wave spikes, and then calculate the fetal heart rate based on the time difference of the RR intervals.

[0037] The maternal assessment module is used to analyze the low-frequency component (LF) and high-frequency component (HF) in the maternal heart rate signal using frequency domain analysis based on pyHRV, and calculate the LF / HF value of the maternal heart rate. When the LF / HF value is greater than 1.5, the mother is considered to have anxiety.

[0038] This project aims to optimize the design of microelectromechanical systems (MEMS) sensor chips to expand the functionality of fetal heart rate monitors. This will enable independent monitoring of the heart rates of both the pregnant woman and the fetus, and analyze the low-frequency to high-frequency (LF / HF) ratio in the heart rate signal to assess the pregnant woman's anxiety level. By simultaneously monitoring the fetal heart rate, the project will provide an objective assessment of the pregnant woman's anxiety, thereby assisting doctors in diagnosis and treatment and reducing the occurrence of psychological disorders during pregnancy.

[0039] Analyzing the low-frequency to high-frequency (LF / HF) ratio of a pregnant woman's heart rate to assess anxiety stems from research on heart rate variability and autonomic nervous system activity, as well as scientific knowledge from related fields of psychophysiology, neuroscience, and clinical medicine. Heart rate variability refers to the change in heart rate at different times, reflecting the autonomic nervous system's control over the heart. The autonomic nervous system is divided into the sympathetic and parasympathetic nervous systems. The sympathetic nervous system is responsible for triggering stress responses, promoting bodily arousal and coping with stress, while the parasympathetic nervous system is responsible for balancing and regulating the activity of the sympathetic nervous system, promoting relaxation and recovery.

[0040] In a state of anxiety, sympathetic nervous system activity increases while parasympathetic nervous system activity decreases, leading to reduced heart rate variability. This means smaller heart rate variations and a more regular heart rate waveform. Conversely, in a relaxed state, parasympathetic nervous system activity increases, heart rate variability increases, and the heart rate waveform becomes more complex and irregular. Analyzing the LF / HF ratio of heart rate signals can indirectly reflect the proportion of autonomic nervous system activity. High-frequency components primarily reflect parasympathetic nervous system activity, while low-frequency components primarily reflect sympathetic nervous system activity. Therefore, in a state of anxiety, when sympathetic nervous system activity increases and parasympathetic nervous system activity decreases, the LF / HF ratio increases. An LF / HF ratio greater than 1.5 is considered indicative of anxiety in pregnant women.

[0041] Research has found that analyzing the LF / HF ratio of pregnant women's heart rate can reflect anxiety levels that the Hamilton Depression Rating Scale cannot measure, specifically anxiety that pregnant women subjectively believe they do not possess. This provides doctors with an objective indicator to assess anxiety levels, helping to prevent anxiety and other psychological problems in pregnant women and aiding in the assessment and diagnosis of their mental health.

[0042] In another preferred embodiment of the present invention, the maternal assessment module uses the open-source Python toolbox pyHRV to analyze maternal heart rate signals with heart rate indicators, including but not limited to calculating heart rate parameters through electrocardiogram, SpO2, blood volume pulse.

[0043] The improvement of MEMS sensors for fetal heart rate monitors involves a shift from rigid MEMS substrates to flexible NEMS substrates based on flexibility and interfaces. Currently, most home-use fetal heart rate monitors are bandage-style, which can cause discomfort for some pregnant women. Therefore, improvements in flexibility are needed to integrate them with wearable sensors. These flexible sensors can then be used as wearable devices to achieve monitoring functions, improving comfort and ease of use. Furthermore, each sensor will be enhanced with pressure-sensitive monocrystalline silicon, allowing for accurate measurement of local pressure when attached to the body, thus improving sensitivity.

[0044] In another preferred embodiment of the present invention, the specific process of the signal separation module extracting fetal signals is as follows:

[0045] Singular value decomposition is performed on the mixed signal to obtain singular value vectors; the singular value vectors are arranged in descending order, and the left and right singular vectors corresponding to the first k singular values ​​are selected as spatial filters; the spatial filters are applied to the mixed signal to obtain the filtered fetal signal.

[0046] In another preferred embodiment of the present invention, the wearable fetal heart rate monitor includes a plurality of MEMS sensing chips on flexible substrates, each of which is equipped with pressure-sensitive single-crystal silicon, and the position of the sensing chip is automatically corrected after the mother wears it.

[0047] In another preferred embodiment of the present invention, the position correction process is as follows:

[0048] Each flexible bandage of the fetal heart rate monitor is equipped with two centrally symmetrical sensor chips. The signal intensities A1 and A2 measured by the two sensor chips in each flexible bandage are collected respectively. The distances d1 and d2 between the two sensor chips and the fetal heart rate are calculated according to the signal attenuation formula, which is:

[0049] A1 = A0·e^(-α·d1), A2= A0·e^(-α·d2);

[0050] Where A0 is the initial intensity of the fetal heart signal, e is the natural constant, α is the human body medium attenuation coefficient, and when the absolute value of the difference between d1 and d2 is greater than the preset threshold, the mother is reminded to adjust the position of the flexible bandage.

[0051] In another preferred embodiment of the present invention, the specific adjustment process is as follows:

[0052] Based on the curvature change of the flexible bandage, the approximate arc formed by the flexible bandage is fitted to generate an ideal arc. The center position and diameter of the ideal circle containing the ideal arc are calculated, and the positions of distances d1 and d2 in the ideal circle are marked. A diameter line segment is drawn through the center position and the fetal heart position. The arc length L1 corresponding to the angle formed by distance d1 and the diameter line segment, and the arc length L2 corresponding to the angle formed by distance d2 and the diameter line segment are calculated respectively. When the absolute value of the difference between d1 and d2 is greater than a preset threshold, the position of the flexible bandage is abnormal. The value of |L1-L2| / 2 is calculated and marked as an adjustment distance value and sent to the fetal heart rate monitor display page.

[0053] In another preferred embodiment of the present invention, a wireless transmission module is further included for receiving and displaying monitoring results, including fetal heart rate, maternal mood and flexible bandage position, and transmitting the monitoring results to preset devices of medical personnel and family members.

[0054] The key problem solved by this invention:

[0055] (1) Optimize the adaptability and response accuracy of the fetal heart rate measurement position to reduce the inaccuracy of fetal heart rate measurement caused by probe position deviation.

[0056] (2) Further optimize the fetal heart signal denoising and equalization processing algorithm, explore the accurate detection of signal endpoints when the fetal heart signal signal-to-noise ratio is low, study the suppression method of jitter phenomenon in the passband range of the equalizer, and identify the noise of the fetal heart signal.

[0057] (3) By improving the materials and the connection method, the sensor’s ability to sense pressure changes is optimized, and wearable technology can replace traditional bandage-type detection.

[0058] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A fetal heart rate monitoring system based on data analysis, using a wearable fetal heart rate monitor, characterized in that, include: A signal acquisition module is used to acquire mixed signals, which include maternal signals and fetal signals; The signal filtering module is used to preprocess the mixed signal. The preprocessing includes using an IIR notch filter to remove power frequency interference, using a median filter to remove baseline drift, and using empirical mode decomposition to remove artificial impulse interference. The signal separation module extracts maternal signals using independent component analysis (ICA), constructs a singular value transfer matrix (SVM), uses maternal abdominal wall ECG signals as spatial filters, and cancels maternal signals through SVM decomposition. Then, ICA is used again to extract fetal signals. The heart rate calculation module uses a time-domain nonlinear transformation thresholding method to determine fetal ECG R-wave spikes, performs cluster detection on the R-wave spikes, and calculates the fetal heart rate based on the time difference of the RR interval. The maternal assessment module uses frequency domain analysis based on pyHRV to analyze the low-frequency (LF) and high-frequency (HF) components of the maternal heart rate signal and calculates the maternal heart rate. The LF / HF ratio is used to determine the maternal anxiety level. When the LF / HP ratio is greater than 1.5, the mother is considered to be experiencing anxiety. The wearable fetal heart rate monitor includes several MEMS sensor chips on flexible substrates. Each sensor chip is equipped with pressure-sensitive monocrystalline silicon. After the mother wears the monitor, the sensor chip's position is automatically corrected. The correction process is as follows: each flexible bandage of the fetal heart rate monitor is equipped with two centrally symmetrical sensor chips. The signal intensities A1 and A2 measured by the two sensor chips in each flexible bandage are collected. The distances d1 and d2 between the two sensor chips and the fetal heart rate are calculated according to the signal attenuation formula. The formula is: A1... = A0·e^(-α·d1), A2= A0·e^(-α·d2); where A0 is the initial intensity of the fetal heart signal, e is the natural constant, and α is the human body medium attenuation coefficient. When the absolute value of the difference between d1 and d2 is greater than the preset threshold, the mother is reminded to adjust the position of the flexible bandage. The specific adjustment process is as follows: based on the curvature change of the flexible bandage, the approximate arc formed by the flexible bandage is fitted to generate an ideal arc. The center position and diameter length of the ideal circle where the ideal arc is located are calculated, and the positions of distances d1 and d2 in the ideal circle are marked. A diameter line segment is drawn through the center position and the fetal heart position. The arc length L1 corresponding to the angle formed by distance d1 and the diameter line segment, and the arc length L2 corresponding to the angle formed by distance d2 and the diameter line segment are calculated respectively. When the absolute value of the difference between d1 and d2 is greater than the preset threshold, the position of the flexible bandage is abnormal. The value of |L1-L2| / 2 is calculated and marked as the adjustment distance value and sent to the fetal heart monitor display page.

2. The fetal heart rate monitoring system based on data analysis according to claim 1, characterized in that, The maternal assessment module uses the open-source Python toolbox pyHRV to analyze maternal heart rate signals with heart rate indicators, including calculating heart rate parameters through electrocardiogram, SpO2, blood volume, and pulse.

3. The fetal heart rate monitoring system based on data analysis according to claim 1, characterized in that, The specific process of the signal separation module to extract the fetal signal is as follows: singular value decomposition is performed on the mixed signal to obtain singular value vectors; the singular value vectors are arranged in descending order, and the left and right singular vectors corresponding to the first k singular values ​​are selected as spatial filters; the spatial filters are applied to the mixed signal to obtain the filtered fetal signal.

4. The fetal heart rate monitoring system based on data analysis according to claim 1, characterized in that, It also includes a wireless transmission module for receiving and displaying monitoring results, including fetal heart rate, maternal mood, and the position of the flexible bandage, and transmitting the monitoring results to the devices of preset medical staff and family members.

Images