A monitoring method for monitoring fetal movement

Abstract

The present invention provides a monitoring method for monitoring fetal movement, including: S1: real-time monitoring to obtain pulse wave signals and electrocardiogram signals at the radial artery of the pregnant woman's wrist and sound signals within a preset range around the pregnant woman's belly; S2: filtering the pulse wave signal, the electrocardiogram signal and the sound signal to obtain corresponding denoised pulse wave signals, denoised electrocardiograms and denoised sound signals; S3: performing multi-dimensional feature analysis on the denoised pulse wave signal, the denoised electrocardiogram and the denoised sound signal to obtain corresponding fetal movement results; filtering the obtained pulse wave signal and electrocardiogram signal of the pregnant woman and the measured sound signals within a preset range around the pregnant woman's belly and performing multi-dimensional feature analysis, thereby realizing the judgment of the actual number of fetal movements without the participation of the pregnant woman and reducing the requirements for the environment, saving the pregnant woman's precious time.

Description

Technical Field

[0001] The present invention relates to the technical field of fetal movement monitoring, and in particular to a monitoring method for monitoring fetal movement. Background Art

[0002] Currently, there are two methods for monitoring fetal movement. The first is to count fetal movements: pregnant mothers should choose a quiet environment and a comfortable position, either sitting in a chair or lying on their side in bed. They should place their hands lightly on their abdominal wall and calmly concentrate on feeling the fetal movement. One movement is counted from the start to the cessation of fetal movement. If the fetus moves several times in a row, it is only counted once. The movement should be counted again after it completely stops. Counting should be done three times a day, morning, noon, and evening, for one hour each time, to understand the pattern of fetal activity in the uterus. The second method is a Doppler fetal heart rate monitor / fetal movement monitor: This monitor uses the Doppler effect of ultrasound and the acoustic characteristics of human tissue structure to monitor fetal heartbeat and fetal movement information.

[0003] The problems with these two monitoring methods are: the measurement cycle is long. Fetal movement monitoring requires counting three times a day, morning, noon, and evening, each time for one hour, to understand the pattern of fetal intrauterine activity. This means that pregnant women have to spend at least three hours a day measuring fetal movement, which is quite time-consuming.

[0004] The environment requirements are high, and pregnant women need to maintain a quiet environment and a comfortable posture during the measurement process. Each measurement takes about 1 hour. Due to the particularity of the hospital environment, it is difficult to ensure that the environment does not change during the measurement process.

[0005] Therefore, the present invention provides a monitoring method for monitoring fetal movement. Summary of the Invention

[0006] The present invention provides a monitoring method for monitoring fetal movement, which is used to filter the pulse wave signal and electrocardiogram signal at the radial artery of the pregnant woman's wrist obtained based on a wearable smart watch and the sound signal within a preset range around the pregnant woman's belly measured based on a piezoelectric sensor, and perform multi-dimensional feature analysis on the filtered results. Then, the actual number of fetal movements can be determined without the participation of the pregnant woman and with reduced environmental requirements, saving the pregnant woman's valuable time.

[0007] The present invention provides a monitoring method for monitoring fetal movement, comprising:

[0008] S1: Used to monitor and obtain pulse wave signals and electrocardiogram signals at the radial artery of the pregnant woman's wrist in real time, as well as sound signals within a preset range around the pregnant woman's belly;

[0009] S2: filtering the pulse wave signal, the electrocardiogram signal, and the sound signal to obtain corresponding denoised pulse wave signal, denoised electrocardiogram, and denoised sound signal;

[0010] S3: Perform multi-dimensional feature analysis on the denoised pulse wave signal, the denoised electrocardiogram, and the denoised sound signal to obtain corresponding fetal movement results.

[0011] Preferably, the method for monitoring fetal movement, S1: for real-time monitoring of pulse wave signals and electrocardiogram signals at the radial artery of the pregnant woman's wrist and sound signals within a preset range around the pregnant woman's belly, comprises:

[0012] S101: Real-time monitoring of pulse wave signals and electrocardiogram signals at the radial artery of a pregnant woman's wrist based on a wearable smartwatch;

[0013] S102: Obtain a sound signal within a preset range around the pregnant woman's belly based on a piezoelectric sensor set at a preset position on the pregnant woman's belly.

[0014] Preferably, the method for monitoring fetal movement, S2: filtering the pulse wave signal, the electrocardiogram signal, and the sound signal to obtain corresponding denoised pulse wave signal, denoised electrocardiogram, and denoised sound signal, comprises:

[0015] S201: filtering out a baseline drift signal and a noise signal in the pulse wave signal and the electrocardiogram signal based on digital filtering to obtain a corresponding denoised pulse wave signal and denoised electrocardiogram;

[0016] S202: removing interference signals from the sound signal based on adaptive filtering to obtain a corresponding denoised sound signal.

[0017] Preferably, the method for monitoring fetal movement, S201: filtering out baseline drift signals and noise signals in the pulse wave signal and the electrocardiogram signal based on digital filtering to obtain corresponding denoised pulse wave signals and denoised electrocardiograms, includes:

[0018] filtering out the baseline drift signal in the pulse wave signal and the electrocardiogram signal based on digital filtering to obtain corresponding filtered pulse wave signal and filtered electrocardiogram signal;

[0019] Filtering out noise signals in the filtered pulse wave signal and the filtered electrocardiogram signal to obtain corresponding denoised pulse wave signal and denoised electrocardiogram signal;

[0020] A corresponding denoised electrocardiogram is fitted based on the denoised electrocardiogram signal.

[0021] Preferably, the method for monitoring fetal movement, based on digital filtering, filters out the baseline drift signal in the pulse wave signal and the electrocardiogram signal to obtain the corresponding filtered pulse wave signal and filtered electrocardiogram signal, includes:

[0022] Taking the pulse wave signal and the electrocardiogram signal as corresponding signals to be filtered, and determining the linear characteristics of the signals to be filtered;

[0023] When the linear characteristic of the signal to be filtered is linear, performing zero drift correction on the signal to be filtered to obtain a corresponding first corrected signal;

[0024] Performing baseline fitting based on the first corrected signal to obtain a corresponding first baseline signal;

[0025] Aligning the first corrected signal and the first baseline signal and performing subtraction processing thereon to obtain a corresponding filtered signal;

[0026] When the linear characteristic of the signal to be filtered is nonlinear, all maximum values contained in the signal to be filtered are determined;

[0027] Determine whether the maximum value difference between the maximum value and the adjacent maximum value is greater than a first maximum value difference threshold; if so, use the point corresponding to the corresponding maximum value as a segmentation point; otherwise, determine that the signal to be filtered does not contain a segmentation point;

[0028] When it is determined that the segmentation point exists in the signal to be filtered, the signal to be filtered is segmented according to the segmentation point to obtain corresponding signal segments to be filtered;

[0029] determining whether the linear characteristic of the signal segment to be filtered is linear, and if so, performing zero drift correction on the signal segment to be filtered to obtain a corresponding first corrected signal segment;

[0030] Performing baseline fitting based on the first corrected signal segment to obtain a corresponding first baseline signal segment;

[0031] Aligning the first corrected signal segment and the first baseline signal segment and performing subtraction processing thereon to obtain a corresponding first filtered signal segment; and sorting and fitting all the first filtered signal segments based on the time sequence relationship between the signal segments to be filtered to obtain a corresponding filtered signal;

[0032] When it is determined that the segmentation point does not exist in the signal to be filtered, the longest unified linear characteristic signal segments are sequentially determined in the signal to be filtered;

[0033] Performing zero drift correction on the longest unified linear characteristic signal segment to obtain a corresponding third corrected signal segment;

[0034] Performing baseline fitting based on the third corrected signal segment to obtain a corresponding third baseline signal segment;

[0035] Aligning the third corrected signal and the third baseline signal and performing subtraction processing thereon to obtain a corresponding third filtered signal segment, and sorting and fitting all the third filtered signal segments based on the timing relationship between the signal segments to be filtered to obtain a corresponding filtered signal;

[0036] A corresponding filtered pulse wave signal and a filtered electrocardiogram signal are obtained based on the signal type corresponding to the filtered signal.

[0037] Preferably, the method for monitoring fetal movement, filtering out noise signals in the filtered pulse wave signal and the filtered electrocardiogram signal to obtain corresponding denoised pulse wave signals and denoised electrocardiogram signals, comprises:

[0038] Calculating a first signal-to-noise ratio corresponding to the filtered pulse wave signal and a second signal-to-noise ratio corresponding to the filtered electrocardiogram signal;

[0039] determining a first noise signal in the filtered pulse wave signal based on the first signal-to-noise ratio;

[0040] determining a second noise signal in the filtered electrocardiogram signal based on the second signal-to-noise ratio;

[0041] The first noise signal in the filtered pulse wave signal is filtered out to obtain a corresponding denoised pulse wave signal, and at the same time, the second noise signal in the filtered electrocardiogram signal is filtered out to obtain a corresponding denoised electrocardiogram signal.

[0042] Preferably, the method for monitoring fetal movement, S202: removing interference signals from the sound signal based on adaptive filtering to obtain a corresponding denoised sound signal, includes:

[0043] Determining a corresponding adaptive filtering algorithm based on a preset interference signal;

[0044] The sound signal is filtered based on a preset filtering order list and the adaptive filtering algorithm to obtain a corresponding denoised sound signal.

[0045] Preferably, the method for monitoring fetal movement, S3: performing multidimensional feature analysis on the denoised pulse wave signal, the denoised electrocardiogram, and the denoised sound signal to obtain corresponding fetal movement results, includes:

[0046] Performing feature extraction on the denoised pulse wave signal, the denoised electrocardiogram, and the denoised sound signal to obtain corresponding abnormal pulse features, abnormal electrocardiogram features, and abnormal sound features;

[0047] Performing multi-dimensional feature analysis on the abnormal pulse feature, the abnormal electrocardiogram feature, and the abnormal sound feature to determine the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle;

[0048] The fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle are used as the corresponding fetal movement results.

[0049] Preferably, the method for monitoring fetal movement, performing feature extraction on the denoised pulse wave signal, the denoised electrocardiogram, and the denoised sound signal to obtain corresponding abnormal pulse features, abnormal electrocardiogram features, and abnormal sound features, includes:

[0050] Acquire a first waveform in the denoised pulse wave signal, determine all first extreme value points contained in the first waveform, and obtain a corresponding first extreme value point set;

[0051] Determining a corresponding maximum first extreme value point in the first extreme value point set, and determining a corresponding main wave value range based on the maximum first extreme value point and a preset main wave error fluctuation value;

[0052] Screening out a second extreme point that satisfies the main wave value range from the first extreme point set;

[0053] Calculating a first mean between the second extreme value point and the corresponding previous adjacent first extreme value point and a second mean between the second extreme value point and the corresponding next adjacent first extreme value point;

[0054] The minimum value between the first mean and the second mean is used as the lower limit of the range corresponding to the second extreme point;

[0055] Determine the lower limit value of the first range based on the average value of the lower limit values of all second extreme value points;

[0056] Taking the maximum first extreme value point as the corresponding upper limit value of the main wave range, and taking the first range lower limit value as the corresponding lower limit value of the main wave range, to obtain the corresponding main wave range;

[0057] Determining a corresponding dicrotic wave value range based on the first range lower limit value and a preset dicrotic wave error fluctuation value;

[0058] Selecting a fourth extreme value point that satisfies the dicrotic wave value range from third extreme value points other than the second extreme value point in the first extreme value point set;

[0059] Calculating a third mean between the fourth extreme point and the corresponding previous adjacent first extreme point and a fourth mean between the fourth extreme point and the corresponding next adjacent first extreme point;

[0060] The minimum value between the third mean and the fourth mean is used as the lower limit of the range corresponding to the fourth extreme point;

[0061] The average value of the lower limits of the range of all fourth extreme value points is used as the lower limit of the corresponding dicrotic wave range, and the maximum fourth extreme value point is used as the upper limit of the dicrotic wave range to obtain the corresponding dicrotic wave range;

[0062] determining a first total number of fifth extreme value points within the main wave range in the first extreme value point set, and simultaneously determining a second total number of sixth extreme value points within the dicrotic wave range in the first extreme value point set;

[0063] determining whether the first total and the second total are consistent; if so, taking the seventh extreme value point in the first extreme value point set excluding the fifth extreme value point and the sixth extreme value point as the corresponding first abnormal extreme value point, and taking the first abnormal extreme value point as the corresponding abnormal pulse feature;

[0064] Otherwise, determining a corresponding second abnormal extreme value point based on the difference between the larger value and the smaller value in the first total and the second total, and taking the eighth extreme value point in the first extreme value point set excluding the fifth extreme value point and the sixth extreme value point as the corresponding third abnormal extreme value point, and taking the second abnormal extreme value point and the third abnormal extreme value point as the corresponding pulse abnormality features;

[0065] Screening out a corresponding fourth abnormal extreme value point in the second waveform corresponding to the denoised electrocardiogram, and using the fourth abnormal extreme value point as a corresponding electrocardiogram abnormal feature;

[0066] A corresponding fifth abnormal extreme value point is screened out from the third waveform corresponding to the denoised sound signal, and the fifth abnormal extreme value point is used as the corresponding sound abnormality feature.

[0067] Preferably, the method for monitoring fetal movement, performing multi-dimensional feature analysis on the abnormal pulse characteristics, the abnormal electrocardiogram characteristics, and the abnormal sound characteristics to determine the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle, includes:

[0068] Aligning the abnormal pulse feature, the abnormal electrocardiogram feature, and the abnormal sound feature in time sequence to obtain a corresponding alignment result;

[0069] The time point corresponding to the aligned abnormal extreme value point in the alignment result is used as the corresponding fetal movement time point;

[0070] Based on all fetal movement time points, the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle are determined.

[0071] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purposes and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0072] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0074] Figure 1 This is a flow chart of a monitoring method for monitoring fetal movement according to an embodiment of the present invention;

[0075] Figure 2 4 is a flow chart of another monitoring method for monitoring fetal movement according to an embodiment of the present invention;

[0076] Figure 3 The following is a flow chart of another monitoring method for monitoring fetal movement in an embodiment of the present invention. DETAILED DESCRIPTION

[0077] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0078] Example 1:

[0079] The present invention provides a monitoring method for monitoring fetal movement, referring to Figure 1 ,include:

[0080] S1: Used to monitor and obtain pulse wave signals and electrocardiogram signals at the radial artery of the pregnant woman's wrist in real time, as well as sound signals within a preset range around the pregnant woman's belly;

[0081] S2: filtering the pulse wave signal, the electrocardiogram signal, and the sound signal to obtain corresponding denoised pulse wave signal, denoised electrocardiogram, and denoised sound signal;

[0082] S3: Perform multi-dimensional feature analysis on the denoised pulse wave signal, the denoised electrocardiogram, and the denoised sound signal to obtain corresponding fetal movement results.

[0083] In this embodiment, the pulse wave signal is a wave signal representing the pulse condition at the radial artery of the pregnant woman's wrist.

[0084] In this embodiment, the electrocardiogram signal is a signal representing the heart activity status of the pregnant woman.

[0085] In this embodiment, the sound signal is a sound signal collected from around the belly of the pregnant woman within a preset range.

[0086] In this embodiment, the denoised pulse wave signal is a signal obtained by filtering the pulse wave signal.

[0087] In this embodiment, the denoised electrocardiogram is a signal obtained by filtering the electrocardiogram signal.

[0088] In this embodiment, the denoised sound signal is a signal obtained by filtering the sound signal.

[0089] In this embodiment, the fetal movement result is a result representing the fetal movement obtained by performing multi-dimensional feature analysis on the denoised pulse wave signal, the denoised electrocardiogram signal, and the denoised sound signal.

[0090] The beneficial effects of the above technology are: filtering the pulse wave signal and electrocardiogram signal at the radial artery of the pregnant woman's wrist obtained based on the wearable smart watch and the sound signal within a preset range around the pregnant woman's belly measured based on the piezoelectric sensor, and performing multi-dimensional feature analysis on the filtered results, thereby realizing the judgment of the actual number of fetal movements without the participation of the pregnant woman and reducing the requirements for the environment, saving the pregnant woman's precious time.

[0091] Example 2:

[0092] Based on Example 1, the method for monitoring fetal movement, S1: for real-time monitoring of the pulse wave signal and electrocardiogram signal at the radial artery of the pregnant woman's wrist and the sound signal within a preset range around the pregnant woman's belly, Figure 2 ,include:

[0093] S101: Real-time monitoring of pulse wave signals and electrocardiogram signals at the radial artery of a pregnant woman's wrist based on a wearable smartwatch;

[0094] S102: Obtain a sound signal within a preset range around the pregnant woman's belly based on a piezoelectric sensor set at a preset position on the pregnant woman's belly.

[0095] The beneficial effects of the above technology are: based on wearable smart watches and piezoelectric sensors, the pulse wave signal and electrocardiogram signal at the radial artery of the pregnant woman's wrist and the sound signal within a preset range around the pregnant woman's belly are obtained. This not only provides a data basis for subsequent fetal movement results, but also overcomes the problems of long measurement cycle and high environmental requirements of traditional fetal movement measurement methods, and also saves time for pregnant women.

[0096] Example 3:

[0097] Based on Example 2, the method for monitoring fetal movement, S2: filtering the pulse wave signal, the electrocardiogram signal and the sound signal to obtain corresponding denoised pulse wave signal, denoised electrocardiogram and denoised sound signal, refer to Figure 3 ,include:

[0098] S201: filtering out a baseline drift signal and a noise signal in the pulse wave signal and the electrocardiogram signal based on digital filtering to obtain a corresponding denoised pulse wave signal and denoised electrocardiogram;

[0099] S202: removing interference signals from the sound signal based on adaptive filtering to obtain a corresponding denoised sound signal.

[0100] In this embodiment, the baseline drift signal is a drift signal contained in the pulse wave signal and the electrocardiogram signal and caused by baseline interference.

[0101] In this embodiment, the noise signal is the noise signal caused by power frequency interference and respiratory jitter contained in the pulse wave signal and the electrocardiogram signal, which affects the accuracy of the pulse wave signal and the electrocardiogram signal.

[0102] In this embodiment, the interference signal is an interference sound signal such as the bowel sounds of a pregnant woman and the sounds of abdominal blood flow.

[0103] The beneficial effects of the above technology are: based on digital filtering, baseline drift signals and noise signals in pulse wave signals and electrocardiogram signals are filtered out. At the same time, based on adaptive filtering, interference signals in sound signals are removed. Clean pulse wave signals, electrocardiograms and fetal movement sound signals without noise and interference can be obtained, which provides an important basis for the subsequent accurate judgment of fetal movement results.

[0104] Example 4:

[0105] Based on Example 3, the method for monitoring fetal movement, S201: filtering out baseline drift signals and noise signals in the pulse wave signal and the electrocardiogram signal based on digital filtering to obtain corresponding denoised pulse wave signals and denoised electrocardiograms, includes:

[0106] filtering out the baseline drift signal in the pulse wave signal and the electrocardiogram signal based on digital filtering to obtain corresponding filtered pulse wave signal and filtered electrocardiogram signal;

[0107] Filtering out noise signals in the filtered pulse wave signal and the filtered electrocardiogram signal to obtain corresponding denoised pulse wave signal and denoised electrocardiogram signal;

[0108] A corresponding denoised electrocardiogram is fitted based on the denoised electrocardiogram signal.

[0109] In this embodiment, the filtered pulse wave signal is a signal obtained by removing a baseline drift signal from the pulse wave signal through digital filtering.

[0110] In this embodiment, the filtered ECG signal is a signal obtained by removing a baseline drift signal from the ECG signal through digital filtering.

[0111] In this embodiment, the denoised pulse wave signal is a signal obtained by filtering the filtered pulse wave signal.

[0112] In this embodiment, the denoised ECG signal is to remove the noise signal in the filtered ECG signal.

[0113] The beneficial effect of the above technology is: by filtering the pulse wave signal and the electrocardiogram signal in batches, it is possible to filter out the extreme drift signals in the pulse wave signal and the electrocardiogram signal, as well as the noise signals in the pulse wave signal and the electrocardiogram signal, thereby achieving a greater degree of removal of all interference signals in the pulse wave signal and the electrocardiogram signal that may interfere with the fetal movement judgment results.

[0114] Example 5:

[0115] Based on Example 4, the method for monitoring fetal movement, which filters out baseline drift signals in the pulse wave signal and the electrocardiogram signal based on digital filtering to obtain corresponding filtered pulse wave signals and filtered electrocardiogram signals, includes:

[0116] Taking the pulse wave signal and the electrocardiogram signal as corresponding signals to be filtered, and determining the linear characteristics of the signals to be filtered;

[0117] When the linear characteristic of the signal to be filtered is linear, performing zero drift correction on the signal to be filtered to obtain a corresponding first corrected signal;

[0118] Performing baseline fitting based on the first corrected signal to obtain a corresponding first baseline signal;

[0119] Aligning the first corrected signal and the first baseline signal and performing subtraction processing thereon to obtain a corresponding filtered signal;

[0120] When the linear characteristic of the signal to be filtered is nonlinear, all maximum values contained in the signal to be filtered are determined;

[0121] Determine whether the maximum value difference between the maximum value and the adjacent maximum value is greater than a first maximum value difference threshold; if so, use the point corresponding to the corresponding maximum value as a segmentation point; otherwise, determine that the signal to be filtered does not contain a segmentation point;

[0122] When it is determined that the segmentation point exists in the signal to be filtered, the signal to be filtered is segmented according to the segmentation point to obtain corresponding signal segments to be filtered;

[0123] determining whether the linear characteristic of the signal segment to be filtered is linear, and if so, performing zero drift correction on the signal segment to be filtered to obtain a corresponding first corrected signal segment;

[0124] Performing baseline fitting based on the first corrected signal segment to obtain a corresponding first baseline signal segment;

[0125] Aligning the first corrected signal segment and the first baseline signal segment and performing subtraction processing thereon to obtain a corresponding first filtered signal segment; and sorting and fitting all the first filtered signal segments based on the time sequence relationship between the signal segments to be filtered to obtain a corresponding filtered signal;

[0126] When it is determined that the segmentation point does not exist in the signal to be filtered, the longest unified linear characteristic signal segments are sequentially determined in the signal to be filtered;

[0127] Performing zero drift correction on the longest unified linear characteristic signal segment to obtain a corresponding third corrected signal segment;

[0128] Performing baseline fitting based on the third corrected signal segment to obtain a corresponding third baseline signal segment;

[0129] Aligning the third corrected signal and the third baseline signal and performing subtraction processing thereon to obtain a corresponding third filtered signal segment, and sorting and fitting all the third filtered signal segments based on the timing relationship between the signal segments to be filtered to obtain a corresponding filtered signal;

[0130] A corresponding filtered pulse wave signal and a filtered electrocardiogram signal are obtained based on the signal type corresponding to the filtered signal.

[0131] In this embodiment, the signals to be filtered are the pulse wave signal and the electrocardiogram signal.

[0132] In this embodiment, the linear feature is the result of determining whether the signal to be filtered is linear.

[0133] In this embodiment, the first correction signal is a corresponding correction signal obtained by performing zero-drift correction on the signal to be filtered when the linear characteristic of the signal to be filtered is determined to be linear.

[0134] In this embodiment, the first baseline signal is a baseline signal obtained after baseline fitting is performed based on the first correction signal.

[0135] In this embodiment, the filtered signal is a signal obtained by removing the baseline drift signal in the signal to be filtered based on digital filtering.

[0136] In this embodiment, the first maximum value difference threshold is the minimum value of the maximum value difference between the point corresponding to the maximum value and the adjacent maximum value when the point corresponding to the maximum value is determined as the segmentation point.

[0137] In this embodiment, the maximum value difference is the difference between a maximum value and an adjacent maximum value.

[0138] In this embodiment, the segmentation point is a point corresponding to a maximum value where the maximum value difference between adjacent maximum values is greater than a first maximum value difference threshold.

[0139] In this embodiment, the signal segment to be filtered is a signal segment obtained by segmenting the signal to be filtered according to the segmentation point.

[0140] In this embodiment, the first corrected signal segment is a signal segment obtained after performing zero-drift correction on the signal segment to be filtered when the linear characteristic of the signal segment to be filtered is determined to be linear.

[0141] In this embodiment, the first baseline signal segment is a baseline signal segment obtained after baseline fitting is performed based on the first corrected signal segment.

[0142] In this embodiment, the first filtered signal segment is a signal segment obtained by aligning the first corrected signal segment and the first baseline signal segment and performing a subtraction process thereon.

[0143] In this embodiment, the longest unified linear characteristic signal segment is a continuous signal segment with a unified linear trend that is determined in sequence in the signal to be filtered when it is determined that the segmentation point does not exist in the signal to be filtered.

[0144] In this embodiment, the third corrected signal segment is a signal segment obtained by performing zero-drift correction on the longest unified linear characteristic signal segment.

[0145] In this embodiment, the third baseline signal segment is a baseline signal segment obtained after baseline fitting is performed based on the third corrected signal segment.

[0146] In this embodiment, the third filtered signal segment is a signal segment obtained by aligning the third correction signal and the third baseline signal and performing subtraction processing.

[0147] The beneficial effects of the above technology are: by determining the corresponding digital filtering method based on the linear trend of the waveforms corresponding to the pulse wave signal and the electrocardiogram signal, and when it is determined that the waveforms corresponding to the pulse wave signal and the electrocardiogram signal are nonlinear, the corresponding waveforms are segmented and then baseline fitting is performed, so that the baseline signals in the pulse wave signal and the electrocardiogram signal can be accurately fitted regardless of whether the pulse wave signal and the electrocardiogram signal are linear or nonlinear, thereby filtering out the baseline interference signals contained in the pulse wave signal and the electrocardiogram signal.

[0148] Example 6:

[0149] Based on Example 5, the method for monitoring fetal movement, filtering out noise signals in the filtered pulse wave signal and the filtered electrocardiogram signal to obtain corresponding denoised pulse wave signals and denoised electrocardiogram signals, includes:

[0150] Calculating a first signal-to-noise ratio corresponding to the filtered pulse wave signal and a second signal-to-noise ratio corresponding to the filtered electrocardiogram signal;

[0151] determining a first noise signal in the filtered pulse wave signal based on the first signal-to-noise ratio;

[0152] determining a second noise signal in the filtered electrocardiogram signal based on the second signal-to-noise ratio;

[0153] The first noise signal in the filtered pulse wave signal is filtered out to obtain a corresponding denoised pulse wave signal, and at the same time, the second noise signal in the filtered electrocardiogram signal is filtered out to obtain a corresponding denoised electrocardiogram signal.

[0154] In this embodiment, calculating the first signal-to-noise ratio corresponding to the filtered pulse wave signal and the second signal-to-noise ratio corresponding to the filtered electrocardiogram signal includes:

[0155]

[0156]

[0157] Where α1 is the first signal-to-noise ratio corresponding to the filtered pulse wave signal, lg is the logarithmic function with base 10, i is the current calculated maximum value in the filtered pulse wave signal, n is the total number of maximum values contained in the filtered pulse wave signal, and V imax is the i-th maximum value contained in the filtered pulse wave signal, α2 is the second signal-to-noise ratio corresponding to the filtered ECG signal, j is the current calculated maximum value in the filtered ECG signal, m is the total number of maximum values contained in the filtered ECG signal, V jmax is the j-th maximum value contained in the filtered ECG signal;

[0158] For example, if the filtered pulse wave signal contains three maximum values, which are 10, 11, and 9, then α1 is 202; if the filtered electrocardiogram signal contains three maximum values, which are 10, 11, and 9, then α2 is 202.

[0159] In this embodiment, the first signal-to-noise ratio is the signal-to-noise ratio corresponding to the filtered pulse wave signal.

[0160] In this embodiment, the second signal-to-noise ratio is the signal-to-noise ratio corresponding to the filtered electrocardiogram signal.

[0161] In this embodiment, the first noise signal is the noise signal contained in the filtered pulse wave signal determined based on the first signal-to-noise ratio.

[0162] In this embodiment, the second noise signal is the noise signal contained in the filtered ECG signal determined based on the second signal-to-noise ratio.

[0163] The beneficial effect of the above technology is: by calculating the signal-to-noise ratio corresponding to the filtered pulse wave signal and the filtered ECG signal, the noise signal in the filtered pulse wave signal and the filtered ECG signal can be accurately determined, and then the filtered pulse wave signal and the filtered ECG signal can be accurately denoised.

[0164] Example 7:

[0165] Based on Example 6, the method for monitoring fetal movement, S202: removing interference signals from the sound signal based on adaptive filtering to obtain a corresponding denoised sound signal, includes:

[0166] Determining a corresponding adaptive filtering algorithm based on a preset interference signal;

[0167] The sound signal is filtered based on a preset filtering order list and the adaptive filtering algorithm to obtain a corresponding denoised sound signal.

[0168] In this embodiment, the preset interference signal includes, for example, bowel sounds of pregnant women, abdominal blood flow sounds, etc.

[0169] In this embodiment, the adaptive filtering algorithm, for example, the least mean square error algorithm (LMS) is used for pregnant women's bowel sounds; and the recursive least squares algorithm (RLS) is used for abdominal blood flow sounds.

[0170] In this embodiment, the preset filtering order list is a list containing the order in which filtering is performed on each preset interference signal.

[0171] The beneficial effect of the above technology is: based on the preset adaptive filtering algorithm corresponding to different types of interfering sound signals, the corresponding types of interfering sound signals in the sound signal are removed in turn, and then the noise-removed sound signal that accurately represents the fetal movement can be obtained.

[0172] Example 8:

[0173] Based on Example 7, the method for monitoring fetal movement, S3: performing multidimensional feature analysis on the denoised pulse wave signal, the denoised electrocardiogram, and the denoised sound signal to obtain corresponding fetal movement results, includes:

[0174] Performing feature extraction on the denoised pulse wave signal, the denoised electrocardiogram, and the denoised sound signal to obtain corresponding abnormal pulse features, abnormal electrocardiogram features, and abnormal sound features;

[0175] Performing multi-dimensional feature analysis on the abnormal pulse feature, the abnormal electrocardiogram feature, and the abnormal sound feature to determine the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle;

[0176] The fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle are used as the corresponding fetal movement results.

[0177] In this embodiment, the abnormal pulse feature is the pulse wave feature representing the fetal movement obtained by extracting features from the de-noised pulse wave signal.

[0178] In this embodiment, the abnormal electrocardiogram feature is the electrocardiogram signal feature representing the fetal movement obtained by extracting features from the denoised electrocardiogram.

[0179] In this embodiment, the abnormal sound feature is the sound feature representing the fetal movement obtained by extracting features from the de-noised sound signal.

[0180] In this embodiment, the fetal movement cycle is the total duration of multiple fetal movements in which the interval between adjacent fetal movements is less than the interval threshold.

[0181] The beneficial effects of the above technology are: by performing multi-dimensional analysis on abnormal features representing fetal movement in the denoised pulse wave signal, denoised electrocardiogram, and denoised sound signal, it is possible to identify fetal movement based on multi-dimensional features such as the electrocardiogram signal and the sound signal, accurately determine the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle, and thus obtain the fetal movement results.

[0182] Example 9:

[0183] Based on Example 8, the method for monitoring fetal movement, performing feature extraction on the denoised pulse wave signal, the denoised electrocardiogram, and the denoised sound signal to obtain corresponding abnormal pulse features, abnormal electrocardiogram features, and abnormal sound features, includes:

[0184] Acquire a first waveform in the denoised pulse wave signal, determine all first extreme value points contained in the first waveform, and obtain a corresponding first extreme value point set;

[0185] Determining a corresponding maximum first extreme value point in the first extreme value point set, and determining a corresponding main wave value range based on the maximum first extreme value point and a preset main wave error fluctuation value;

[0186] Screening out a second extreme point that satisfies the main wave value range from the first extreme point set;

[0187] Calculating a first mean between the second extreme value point and the corresponding previous adjacent first extreme value point and a second mean between the second extreme value point and the corresponding next adjacent first extreme value point;

[0188] The minimum value between the first mean and the second mean is used as the lower limit of the range corresponding to the second extreme point;

[0189] Determine the lower limit value of the first range based on the average value of the lower limit values of all second extreme value points;

[0190] Taking the maximum first extreme value point as the corresponding upper limit value of the main wave range, and taking the first range lower limit value as the corresponding lower limit value of the main wave range, to obtain the corresponding main wave range;

[0191] Determining a corresponding dicrotic wave value range based on the first range lower limit value and a preset dicrotic wave error fluctuation value;

[0192] Selecting a fourth extreme value point that satisfies the dicrotic wave value range from third extreme value points other than the second extreme value point in the first extreme value point set;

[0193] Calculating a third mean between the fourth extreme point and the corresponding previous adjacent first extreme point and a fourth mean between the fourth extreme point and the corresponding next adjacent first extreme point;

[0194] The minimum value between the third mean and the fourth mean is used as the lower limit of the range corresponding to the fourth extreme point;

[0195] The average value of the lower limits of the range of all fourth extreme value points is used as the lower limit of the corresponding dicrotic wave range, and the maximum fourth extreme value point is used as the upper limit of the dicrotic wave range to obtain the corresponding dicrotic wave range;

[0196] determining a first total number of fifth extreme value points within the main wave range in the first extreme value point set, and simultaneously determining a second total number of sixth extreme value points within the dicrotic wave range in the first extreme value point set;

[0197] determining whether the first total and the second total are consistent; if so, taking the seventh extreme value point in the first extreme value point set excluding the fifth extreme value point and the sixth extreme value point as the corresponding first abnormal extreme value point, and taking the first abnormal extreme value point as the corresponding abnormal pulse feature;

[0198] Otherwise, determining a corresponding second abnormal extreme value point based on the difference between the larger value and the smaller value in the first total and the second total, and taking the eighth extreme value point in the first extreme value point set excluding the fifth extreme value point and the sixth extreme value point as the corresponding third abnormal extreme value point, and taking the second abnormal extreme value point and the third abnormal extreme value point as the corresponding pulse abnormality features;

[0199] Screening out a corresponding fourth abnormal extreme value point in the second waveform corresponding to the denoised electrocardiogram, and using the fourth abnormal extreme value point as a corresponding electrocardiogram abnormal feature;

[0200] A corresponding fifth abnormal extreme value point is screened out from the third waveform corresponding to the denoised sound signal, and the fifth abnormal extreme value point is used as the corresponding sound abnormality feature.

[0201] In this embodiment, the first waveform is the waveform included in the denoised pulse wave signal.

[0202] In this embodiment, the first extreme value points are all extreme value points included in the first waveform.

[0203] In this embodiment, the first extreme value point set is a set including all first extreme value points.

[0204] In this embodiment, the maximum first extreme value point is the maximum extreme value point in the first extreme value point set.

[0205] In this embodiment, the preset main wave error fluctuation value is a preset error fluctuation value of the main wave corresponding to the de-noised pulse wave signal.

[0206] In this embodiment, the main wave value range is the corresponding value range obtained by taking the maximum first extreme value point as the upper limit value of the main wave value range, and at the same time, taking the difference between the maximum first extreme value point and the preset main wave error fluctuation value as the lower limit value of the main wave value range.

[0207] In this embodiment, the second extreme value point is an extreme value point selected from the first extreme value point set that satisfies the value range of the main wave.

[0208] In this embodiment, the first mean value is the mean value between the second extreme value point and the corresponding previous adjacent first extreme value point.

[0209] In this embodiment, the second mean is the mean between the two extreme value points and the corresponding next adjacent first extreme value point.

[0210] In this embodiment, the lower limit of the range of the second extreme point is the minimum value between the first mean value and the second mean value.

[0211] In this embodiment, the lower limit value of the first range is the average value of the lower limits of all the second extreme value points.

[0212] In this embodiment, the upper limit value of the main wave range is the maximum first extreme value point.

[0213] In this embodiment, the lower limit value of the main wave range is the lower limit value of the first range.

[0214] In this embodiment, the main wave range is a value range determined based on the upper limit value of the main wave range and the lower limit value of the main wave range.

[0215] In this embodiment, the dicrotic wave value range is determined by taking the lower limit value of the first range as the upper limit value of the dicrotic wave value range, and taking the difference between the lower limit value of the first range and the preset dicrotic wave error fluctuation value as the lower limit value of the dicrotic wave value range.

[0216] In this embodiment, the preset dicrotic wave error fluctuation value is a preset dicrotic wave error fluctuation value corresponding to the de-noised pulse wave signal.

[0217] In this embodiment, the third extreme value is the point in the first extreme value point set excluding the extreme value point in the first extreme value point set.

[0218] In this embodiment, the fourth extreme value point is an extreme value point selected from the third extreme value points that meets the dicrotic wave value range.

[0219] In this embodiment, the third mean is the mean between the fourth extreme point and the corresponding previous adjacent first extreme point.

[0220] In this embodiment, the fourth mean is the mean between the fourth extreme point and the corresponding next adjacent first extreme point.

[0221] In this embodiment, the lower limit of the range of the fourth extreme point is the minimum value between the third mean value and the fourth mean value.

[0222] In this embodiment, the lower limit value of the dicrotic wave range is the average value of the lower limit values of all fourth extreme value points.

[0223] In this embodiment, the upper limit of the dicrotic wave range is the maximum fourth extreme point.

[0224] In this embodiment, the dicrotic wave range is a value range obtained based on the dicrotic wave range lower limit value and the dicrotic wave range upper limit value.

[0225] In this embodiment, the first total number is the total number of fifth extreme value points in the first extreme value point set within the main wave range.

[0226] In this embodiment, the fifth extreme point is the extreme point in the first extreme point set within the main wave range.

[0227] In this embodiment, the sixth extreme point is an extreme point in the first extreme point set within the dicrotic wave range.

[0228] In this embodiment, the second total number is the total number of sixth extreme value points in the first extreme value point set within the dicrotic wave range.

[0229] In this embodiment, the seventh extreme value point is an extreme value point in the first extreme value point set other than the sixth extreme value point in the first extreme value point set when it is determined that the first total number and the second total number are consistent.

[0230] In this embodiment, the first abnormal extreme value point is the seventh extreme value point.

[0231] In this embodiment, determining the corresponding second abnormal extreme point based on the difference between the larger value and the smaller value in the first total and the second total includes:

[0232] When the first total is greater than the second total, the difference outlier points are selected from the fifth extreme value points as the corresponding second abnormal extreme value points;

[0233] When the first total is less than the second total, outliers corresponding to the number of the difference are selected from the sixth extreme value points as the corresponding second abnormal extreme value points.

[0234] In this embodiment, the third abnormal extreme value point is the eighth extreme value point in the first extreme value point set except the fifth extreme value point and the sixth extreme value point when it is determined that the first total number is inconsistent with the second total number.

[0235] In this embodiment, the eighth extreme value point is the extreme value point in the first extreme value point set except the fifth extreme value point and the sixth extreme value point.

[0236] In this embodiment, the fourth abnormal extreme value point is the corresponding abnormal extreme value point screened out from the second waveform corresponding to the denoised electrocardiogram.

[0237] In this embodiment, the second waveform is the waveform corresponding to the denoised electrocardiogram.

[0238] In this embodiment, the third waveform is the waveform corresponding to the denoised audio signal.

[0239] In this embodiment, the fifth abnormal extreme value point is a corresponding abnormal extreme value point screened out from the third waveform corresponding to the noise-removing sound signal.

[0240] The beneficial effects of the above technology are: by identifying and judging the extreme points corresponding to the main wave and the dicrotic wave in the pulse wave, it is possible to accurately identify the extreme points corresponding to the main wave and the dicrotic wave in the pulse wave, and then screen out other abnormal extreme points caused by fetal movement. At the same time, the extreme points in the denoised electrocardiogram and the denoised sound signal are determined, which provides a basis for accurately determining the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle.

[0241] Example 10:

[0242] Based on Example 9, the method for monitoring fetal movement, performing multidimensional feature analysis on the abnormal pulse characteristics, the abnormal electrocardiogram characteristics, and the abnormal sound characteristics to determine the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle, includes:

[0243] Aligning the abnormal pulse feature, the abnormal electrocardiogram feature, and the abnormal sound feature in time sequence to obtain a corresponding alignment result;

[0244] The time point corresponding to the aligned abnormal extreme value point in the alignment result is used as the corresponding fetal movement time point;

[0245] Based on all fetal movement time points, the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle are determined.

[0246] In this embodiment, the alignment result is a result obtained by aligning the abnormal pulse feature, the abnormal electrocardiogram feature, and the abnormal sound feature in time sequence.

[0247] In this embodiment, the fetal movement time point is the time point corresponding to the aligned abnormal extreme value point in the alignment result.

[0248] In this embodiment, based on all fetal movement time points, the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle are determined, including:

[0249] A group of adjacent fetal movement time points with a time interval less than a preset time interval are regarded as the same fetal movement cycle, thereby dividing all fetal movement time points into cycles, determining the corresponding fetal movement cycles, and determining the actual number of fetal movements corresponding to the corresponding cycle based on the total number of fetal movement time points contained in each fetal movement cycle.

[0250] The beneficial effects of the above technology are: based on the aligned statistics of abnormal pulse characteristics, abnormal electrocardiogram characteristics and abnormal sound characteristics, the accuracy of fetal movement recognition can be controlled in the final step, and the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle can be accurately determined, thereby obtaining the fetal movement results.

[0251] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for monitoring fetal movement, characterized in that: include: S1: Real-time monitoring to obtain pulse wave signals and electrocardiogram signals at the radial artery of the pregnant woman's wrist, as well as sound signals within a preset range around the pregnant woman's belly; S2: filtering the pulse wave signal, the electrocardiogram signal, and the sound signal to obtain corresponding denoised pulse wave signal, denoised electrocardiogram, and denoised sound signal; S3: performing multi-dimensional feature analysis on the denoised pulse wave signal, the denoised electrocardiogram, and the denoised sound signal to obtain corresponding fetal movement results, including: Performing feature extraction on the denoised pulse wave signal, the denoised electrocardiogram, and the denoised sound signal to obtain corresponding abnormal pulse features, abnormal electrocardiogram features, and abnormal sound features; Performing multi-dimensional feature analysis on the abnormal pulse feature, the abnormal electrocardiogram feature, and the abnormal sound feature to determine the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle, including: Aligning the abnormal pulse feature, the abnormal electrocardiogram feature, and the abnormal sound feature in time sequence to obtain a corresponding alignment result; The time point corresponding to the aligned abnormal extreme value point in the alignment result is used as the corresponding fetal movement time point; Based on all fetal movement time points, determine the corresponding fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle; The fetal movement cycle and the actual number of fetal movements corresponding to each fetal movement cycle are used as the corresponding fetal movement results.

2. A method for monitoring fetal movement according to claim 1, characterized in that: S1: Used to monitor and obtain pulse wave signals and electrocardiogram signals at the radial artery of the pregnant woman's wrist in real time, as well as sound signals within a preset range around the pregnant woman's belly, including: S101: Real-time monitoring of pulse wave signals and electrocardiogram signals at the radial artery of a pregnant woman's wrist based on a wearable smartwatch; S102: Obtain a sound signal within a preset range around the pregnant woman's belly based on a piezoelectric sensor set at a preset position on the pregnant woman's belly.

3. A method for monitoring fetal movement according to claim 2, characterized in that: S2: Filtering the pulse wave signal, the electrocardiogram signal, and the sound signal to obtain corresponding denoised pulse wave signal, denoised electrocardiogram, and denoised sound signal, including: S201: filtering out a baseline drift signal and a noise signal in the pulse wave signal and the electrocardiogram signal based on digital filtering to obtain a corresponding denoised pulse wave signal and denoised electrocardiogram; S202: removing interference signals from the sound signal based on adaptive filtering to obtain a corresponding denoised sound signal.

4. A method for monitoring fetal movement according to claim 3, characterized in that: S201: filtering out the baseline drift signal and the noise signal in the pulse wave signal and the electrocardiogram signal based on digital filtering to obtain a corresponding denoised pulse wave signal and denoised electrocardiogram, including: filtering out the baseline drift signal in the pulse wave signal and the electrocardiogram signal based on digital filtering to obtain corresponding filtered pulse wave signal and filtered electrocardiogram signal; Filtering out noise signals in the filtered pulse wave signal and the filtered electrocardiogram signal to obtain corresponding denoised pulse wave signal and denoised electrocardiogram signal; A corresponding denoised electrocardiogram is fitted based on the denoised electrocardiogram signal.

5. A method for monitoring fetal movement according to claim 4, characterized in that: Filtering out the baseline drift signal in the pulse wave signal and the electrocardiogram signal based on digital filtering to obtain corresponding filtered pulse wave signal and filtered electrocardiogram signal, including: Taking the pulse wave signal and the electrocardiogram signal as corresponding signals to be filtered, and determining the linear characteristics of the signals to be filtered; When the linear characteristic of the signal to be filtered is linear, performing zero drift correction on the signal to be filtered to obtain a corresponding first corrected signal; Performing baseline fitting based on the first corrected signal to obtain a corresponding first baseline signal; Aligning the first corrected signal and the first baseline signal and performing subtraction processing thereon to obtain a corresponding filtered signal; When the linear characteristic of the signal to be filtered is nonlinear, all maximum values contained in the signal to be filtered are determined; Determine whether the maximum value difference between the maximum value and the adjacent maximum value is greater than a first maximum value difference threshold; if so, use the point corresponding to the corresponding maximum value as a segmentation point; otherwise, determine that the signal to be filtered does not contain a segmentation point; When it is determined that the segmentation point exists in the signal to be filtered, the signal to be filtered is segmented according to the segmentation point to obtain corresponding signal segments to be filtered; determining whether the linear characteristic of the signal segment to be filtered is linear, and if so, performing zero drift correction on the signal segment to be filtered to obtain a corresponding first corrected signal segment; Performing baseline fitting based on the first corrected signal segment to obtain a corresponding first baseline signal segment; Aligning the first corrected signal segment and the first baseline signal segment and performing subtraction processing thereon to obtain a corresponding first filtered signal segment; and sorting and fitting all the first filtered signal segments based on the time sequence relationship between the signal segments to be filtered to obtain a corresponding filtered signal; When it is determined that the segmentation point does not exist in the signal to be filtered, the longest unified linear characteristic signal segments are sequentially determined in the signal to be filtered; Performing zero drift correction on the longest unified linear characteristic signal segment to obtain a corresponding third corrected signal segment; Performing baseline fitting based on the third corrected signal segment to obtain a corresponding third baseline signal segment; Aligning the third corrected signal and the third baseline signal and performing subtraction processing thereon to obtain a corresponding third filtered signal segment, and sorting and fitting all the third filtered signal segments based on the timing relationship between the signal segments to be filtered to obtain a corresponding filtered signal; A corresponding filtered pulse wave signal and a filtered electrocardiogram signal are obtained based on the signal type corresponding to the filtered signal.

6. A method for monitoring fetal movement according to claim 5, characterized in that: Filtering out noise signals in the filtered pulse wave signal and the filtered electrocardiogram signal to obtain corresponding denoised pulse wave signals and denoised electrocardiogram signals, including: Calculating a first signal-to-noise ratio corresponding to the filtered pulse wave signal and a second signal-to-noise ratio corresponding to the filtered electrocardiogram signal; determining a first noise signal in the filtered pulse wave signal based on the first signal-to-noise ratio; determining a second noise signal in the filtered electrocardiogram signal based on the second signal-to-noise ratio; The first noise signal in the filtered pulse wave signal is filtered out to obtain a corresponding denoised pulse wave signal, and at the same time, the second noise signal in the filtered electrocardiogram signal is filtered out to obtain a corresponding denoised electrocardiogram signal.

7. A method for monitoring fetal movement according to claim 6, characterized in that: S202: removing interference signals from the sound signal based on adaptive filtering to obtain a corresponding denoised sound signal, including: Determining a corresponding adaptive filtering algorithm based on a preset interference signal; The sound signal is filtered based on a preset filtering order list and the adaptive filtering algorithm to obtain a corresponding denoised sound signal.

8. The method for monitoring fetal movement according to claim 1, wherein: Performing feature extraction on the denoised pulse wave signal, the denoised electrocardiogram, and the denoised sound signal to obtain corresponding abnormal pulse features, abnormal electrocardiogram features, and abnormal sound features, including: Acquire a first waveform in the denoised pulse wave signal, determine all first extreme value points contained in the first waveform, and obtain a corresponding first extreme value point set; Determining a corresponding maximum first extreme value point in the first extreme value point set, and determining a corresponding main wave value range based on the maximum first extreme value point and a preset main wave error fluctuation value; Screening out a second extreme point that satisfies the main wave value range from the first extreme point set; Calculating a first mean between the second extreme value point and the corresponding previous adjacent first extreme value point and a second mean between the second extreme value point and the corresponding next adjacent first extreme value point; The minimum value between the first mean and the second mean is used as the lower limit of the range corresponding to the second extreme point; Determine the lower limit value of the first range based on the average value of the lower limit values of all second extreme value points; Taking the maximum first extreme value point as the corresponding upper limit value of the main wave range, and taking the first range lower limit value as the corresponding lower limit value of the main wave range, to obtain the corresponding main wave range; Determining a corresponding dicrotic wave value range based on the first range lower limit value and a preset dicrotic wave error fluctuation value; Selecting a fourth extreme value point that satisfies the dicrotic wave value range from third extreme value points other than the second extreme value point in the first extreme value point set; Calculating a third mean between the fourth extreme point and the corresponding previous adjacent first extreme point and a fourth mean between the fourth extreme point and the corresponding next adjacent first extreme point; The minimum value between the third mean and the fourth mean is used as the lower limit of the range corresponding to the fourth extreme point; The average of the lower limits of the range of all fourth extreme value points is used as the lower limit of the corresponding dicrotic wave range, and the maximum fourth extreme value point is used as the upper limit of the dicrotic wave range to obtain the corresponding dicrotic wave range; determining a first total number of fifth extreme value points within the main wave range in the first extreme value point set, and simultaneously determining a second total number of sixth extreme value points within the dicrotic wave range in the first extreme value point set; determining whether the first total and the second total are consistent; if so, taking the seventh extreme value point in the first extreme value point set excluding the fifth extreme value point and the sixth extreme value point as the corresponding first abnormal extreme value point, and taking the first abnormal extreme value point as the corresponding abnormal pulse feature; Otherwise, determining a corresponding second abnormal extreme value point based on the difference between the larger value and the smaller value in the first total and the second total, and taking the eighth extreme value point in the first extreme value point set excluding the fifth extreme value point and the sixth extreme value point as the corresponding third abnormal extreme value point, and taking the second abnormal extreme value point and the third abnormal extreme value point as the corresponding pulse abnormality features; Screening out a corresponding fourth abnormal extreme value point in the second waveform corresponding to the denoised electrocardiogram, and using the fourth abnormal extreme value point as a corresponding electrocardiogram abnormal feature; A corresponding fifth abnormal extreme value point is screened out from the third waveform corresponding to the denoised sound signal, and the fifth abnormal extreme value point is used as the corresponding sound abnormality feature.

Images