Remote monitoring method and system for physical signs of puerpera

Abstract

The invention provides a remote monitoring method and system for physical signs of a lying-in woman. The method comprises the following steps: remotely collecting fetal heart signals and motion data of the lying-in woman; separating from the fetal heart signal according to the correlation between the motion data and the fetal heart signal to obtain an artifact signal component when the puerpera moves; performing time domain alignment difference on the artifact signal component and the fetal heart signal to obtain a pure fetal heart signal of the puerpera, and further positioning in a joint time-frequency domain of the pure fetal heart signal according to the stripe characteristics of motion artifacts to obtain a data defect interval formed by artifact residues; performing trend reconstruction on the data defect interval based on the signal trend of the pure fetal heart signal in a preset neighborhood so as to generate a complete fetal heart monitoring waveform; and performing remote monitoring on intrauterine signs of a puerpera through the fetal heart monitoring waveform. By the adoption of the scheme, the clinical diagnosis features of the fetal heart rate can be completely reserved while motion artifacts are eliminated.

Description

Technical Field

[0001] This application relates to the field of vital sign monitoring technology, and more specifically, to a method and system for remote monitoring of maternal vital signs. Background Technology

[0002] Vital signs monitoring refers to the continuous collection and dynamic analysis of an individual's vital signs parameters (such as electrocardiogram, respiration, blood oxygen, body temperature, and blood pressure) through multiple sensing methods such as wearable sensors, non-contact monitoring devices, and medical-grade instruments. Combined with physiological signal processing, abnormal pattern recognition, and risk warning algorithms, discrete physiological data are transformed into clinically meaningful health status trend maps, thereby achieving a leap from isolated data points to systematic health assessment of the target object.

[0003] Remote monitoring of maternal vital signs refers to the continuous collection and cloud transmission of key physiological parameters of pregnant women, such as fetal heart rate, fetal movement, uterine contraction pressure, blood pressure, blood oxygenation, and heart rate, through IoT sensing systems such as smart wearable devices and home fetal heart rate monitors. This data is combined with a perinatal medicine knowledge base, abnormal waveform recognition algorithms, and risk stratification models to transform discrete physiological signals into clinically significant situational assessment methods. Traditional remote fetal heart rate monitoring typically uses fixed filtering parameters for global noise suppression, which struggles to effectively resolve the contradiction between eliminating motion artifacts and preserving fetal heart rate characteristics. Furthermore, it lacks adaptive recognition capabilities for the time-frequency characteristics of motion artifacts, resulting in insufficient reliability of monitoring data during maternal activity. For example, traditional methods may use filters with preset thresholds for static filtering of fetal heart rate signals. While this can suppress some motion interference, it also weakens the instantaneous variability and waveform details of fetal heart rate, making it difficult for subsequent analysis to distinguish between true heart rate changes and signal distortion caused by motion. Therefore, how to eliminate motion artifacts while fully preserving the clinical diagnostic characteristics of fetal heart rate has become a challenge for the industry. Summary of the Invention

[0004] This application provides a method and system for remote monitoring of maternal vital signs, which can completely preserve the clinical diagnostic characteristics of fetal heart rate while eliminating motion artifacts.

[0005] Firstly, this application provides a method for remote monitoring of maternal vital signs, comprising the following steps: Remotely collect fetal heart rate signals and movement data from pregnant women; Based on the correlation between the motion data and the fetal heart rate signal, the artifact signal component during maternal movement is separated from the fetal heart rate signal. The artifact signal components and the fetal heart signal are time-domain aligned and differentiated to obtain the mother's pure fetal heart signal. Then, based on the banding characteristics of motion artifacts, the data loss interval formed by artifact residue is located in the joint time-frequency domain of the pure fetal heart signal. Based on the signal trend of the pure fetal heart signal in a preset neighborhood, the data loss interval is reconstructed to generate a complete fetal heart monitoring waveform. The fetal heart rate monitoring waveform is used to remotely monitor the intrauterine vital signs of the parturient.

[0006] In some embodiments, separating the artifact signal component during maternal movement from the fetal heart signal based on the correlation between the motion data and the fetal heart signal specifically includes: Determine the correlation between the motion data and the fetal heart rate signal; By analyzing the correlation between the motion data and the fetal heart rate signal, the six-degree-of-freedom motion parameters in the motion data are fused to generate a comprehensive motion reference signal. An adaptive filter model is constructed based on the integrated motion reference signal and the fetal heart rate signal, and then the artifact signal component during maternal movement is obtained from the fetal heart rate signal. In some embodiments, performing time-domain aligned differential analysis between the artifact signal component and the fetal heart rate signal to obtain the mother's pure fetal heart rate signal specifically includes: Determine the transmission delay between the artifact signal component and the fetal heart rate signal; The artifact signal component is time-delayed by the transmission delay, so that the artifact signal component is aligned with the time domain of the fetal heart signal. The artifact signal components after time-domain alignment are differentially analyzed with the fetal heart signal to obtain the pure fetal heart signal of the mother.

[0007] In some embodiments, the data loss interval formed by artifact residue is located in the joint time-frequency domain of the pure fetal heart signal based on the banding characteristics of motion artifacts, specifically including: Time-frequency analysis is performed on the pure fetal heart rate signal to generate a joint time-frequency distribution of the pure fetal heart rate signal; Identify anomalous energy regions in the joint time-frequency distribution that conform to the characteristics of motion artifact stripes; Based on the time-frequency characteristics of the abnormal energy region, the data loss interval formed by artifact residue is located.

[0008] In some embodiments, reconstructing the data loss interval based on the signal trend of the pure fetal heart signal in a preset neighborhood, thereby generating a complete fetal heart monitoring waveform, specifically includes: Determine the preset neighborhood range corresponding to the data loss interval in the pure fetal heart rate signal; Extract the signal trend features within the preset neighborhood range; Based on the signal trend characteristics, the signal is reconstructed in the data loss area to generate a complete fetal heart rate monitoring waveform.

[0009] In some embodiments, remote monitoring of the mother's intrauterine vital signs via the fetal heart rate monitoring waveform specifically includes: Clinical characteristic parameters of fetal heart rate were extracted from the complete fetal heart rate monitoring waveform. The intrauterine vital signs of the parturient are remotely monitored using the aforementioned clinical characteristic parameters.

[0010] In some embodiments, the clinical characteristic parameters include: mean fetal heart rate, fetal heart rate variability, number of fetal heart rate accelerations, and number of fetal heart rate decelerations. Secondly, this application provides a remote monitoring system for maternal vital signs, comprising: The acquisition module is used to remotely acquire fetal heart rate signals and movement data from mothers in labor. The processing module is used to separate the artifact signal component of the mother's movement from the fetal heart signal based on the correlation between the motion data and the fetal heart signal. The processing module is also used to perform time-domain alignment and differential analysis on the artifact signal component and the fetal heart signal to obtain the mother's pure fetal heart signal, and then locate the data loss interval formed by the artifact residue in the joint time-frequency domain of the pure fetal heart signal based on the strip characteristics of motion artifacts. The processing module is also used to reconstruct the data loss interval based on the signal trend of the pure fetal heart signal in a preset neighborhood, thereby generating a complete fetal heart monitoring waveform. The execution module is used to remotely monitor the intrauterine vital signs of the parturient through the fetal heart rate monitoring waveform.

[0011] Thirdly, this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method for remote monitoring of maternal vital signs.

[0012] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method for remote monitoring of maternal vital signs.

[0013] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The remote monitoring method and system for maternal vital signs provided in this application firstly separates the artifact signal components from the maternal fetal heart rate signal using time-domain aligned differential spectroscopy to obtain the pure fetal heart rate signal. This process constructs a dedicated motion artifact model based on the correlation between motion sensing data and the fetal heart rate signal, and uses an adaptive separation algorithm to achieve targeted removal of motion interference. Compared with traditional fixed-threshold filtering methods, this effectively avoids excessive filtering of clinically valuable transient variation features and morphological details in the fetal heart rate waveform, preserving complete physiological signal features for subsequent analysis. Subsequently, time-frequency analysis is performed on the pure fetal heart rate signal, based on the motion artifacts in the combined time-frequency... The high-energy bands characteristic of the region allow for precise localization of data loss regions caused by residual artifacts. This process uses time-frequency analysis tools to convert the signal into a two-dimensional time-frequency space for analysis, clearly distinguishing between continuous rhythmic subbands representing the true fetal heart rate and transient high-energy interference caused by motion. This enables precise localization of signal distortion areas, avoiding the misclassification and removal of clinically significant transient features such as fetal heart rate acceleration / deceleration as noise. It provides a clear target area for subsequent signal reconstruction, ensuring that only truly damaged data segments are repaired. In summary, this approach can completely preserve the clinical diagnostic characteristics of the fetal heart rate while eliminating motion artifacts. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating a method for remote monitoring of maternal vital signs according to some embodiments of this application; Figure 2 This is a flowchart illustrating the determination of artifact signal components according to some embodiments of this application; Figure 3 This is a schematic diagram of the process for determining transmission delay according to some embodiments of this application; Figure 4 This is a schematic diagram of the structure of a remote monitoring system for maternal vital signs according to some embodiments of this application; Figure 5 This is an internal structural diagram of a computer device for implementing a method for remote monitoring of maternal vital signs according to some embodiments of this application. Detailed Implementation

[0015] To better understand the technical solutions in this embodiment, the technical solutions in this embodiment will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0016] refer to Figure 1 The figure is a flowchart illustrating a method for remote monitoring of maternal vital signs according to some embodiments of this application. The method for remote monitoring of maternal vital signs mainly includes the following steps: In step 101, the fetal heart rate signal and movement data of the mother are collected remotely.

[0017] In practice, raw physiological data can be synchronously collected using a Doppler fetal heart rate probe deployed on the mother's abdomen and its integrated motion sensing module. Preferably, a continuous wave Doppler probe operating in a preset frequency band emits ultrasound waves into the abdominal cavity and receives echo signals containing fetal heart motion information. Simultaneously, a nine-axis inertial measurement unit fixed to the probe housing collects triaxial acceleration, triaxial angular velocity, and triaxial magnetic field strength data at a preset sampling rate. The echo signals are demodulated by a mixer and then the baseband fetal heart rate signal is extracted through a bandpass filter. The inertial data is transformed by coordinate transformation to generate six-degree-of-freedom motion parameters characterizing the probe's spatial pose. Both are strictly synchronized in time by a hardware clock source, together constituting the mother's fetal heart rate signal and motion data.

[0018] It should be noted that the fetal heart signal mentioned in this application can also refer to the low-frequency electrical signal of abdominal wall micro-motion information caused by fetal heart pulsation, and its physical meaning is the phase modulation amount generated by the mechanical motion of the fetal heart at the receiving probe per unit time; the motion data refers to the set of parameters describing the spatial motion state of the fetal heart probe directly collected by the inertial measurement unit; in addition, as a preferred embodiment, the preset frequency band can be set to 2.0-3.0MHz, and the preset sampling frequency is not lower than 200Hz.

[0019] In step 102, the artifact signal component during the mother's movement is separated from the fetal heart signal based on the correlation between the motion data and the fetal heart signal.

[0020] In some embodiments, reference Figure 2 As shown in the figure, this is a schematic flowchart illustrating the determination of artifact signal components according to some embodiments of this application. The artifact signal components during maternal movement can be separated from the fetal heart rate signal based on the correlation between the motion data and the fetal heart rate signal using the following steps: First, in step 1021, the correlation between the motion data and the fetal heart rate signal is determined; Then, in step 1022, the six-degree-of-freedom motion parameters in the motion data are fused by the correlation between the motion data and the fetal heart signal to generate a comprehensive motion reference signal; Finally, in step 1023, an adaptive filter model is constructed based on the integrated motion reference signal and the fetal heart signal, and then the artifact signal component during the mother's movement is constructed from the fetal heart signal.

[0021] In specific implementation, the correlation between the motion data and the fetal heart rate signal can be determined in the following ways: First, the six-degree-of-freedom motion parameter components in the motion data are time-series aligned with the fetal heart rate signal to ensure that the data points strictly correspond to the same time axis; then, for each motion parameter component, the cross-correlation function between each motion parameter component and the fetal heart rate signal within a preset time delay range is calculated, where the time delay range is preset based on the physical delay characteristics of physiological signal transmission; finally, the peak value of the cross-correlation function sequence corresponding to each motion parameter component is extracted, and the peak value is used as a quantitative index of the correlation strength between the motion dimension and the motion interference component in the fetal heart rate signal, thereby obtaining the correlation between the motion data and the fetal heart rate signal; preferably, the calculation of the cross-correlation function can be performed using a frequency domain method based on fast Fourier transform to improve computational efficiency, and the quantitative index of correlation strength will serve as a direct basis for weight allocation in the subsequent feature fusion step, which is not limited in this application.

[0022] It should be noted that the correlation between the motion data and the fetal heart signal mentioned in this application refers to a quantitative index used to characterize the degree of linear correlation between each motion parameter component in the motion data and the motion interference component in the fetal heart signal. It is used to reflect the interference intensity of different motion modes on the quality of the fetal heart signal, thereby providing a basis for weight allocation for the separation of motion artifacts.

[0023] In specific implementation, the six-degree-of-freedom motion parameters in the motion data are fused based on the correlation between the motion data and the fetal heart rate signal to generate a comprehensive motion reference signal. This can be achieved in the following way: First, the correlation strength quantification index of each motion parameter component is normalized so that its sum is one, thus forming a set of weighting coefficients; then, the time series of each component of each six-degree-of-freedom motion parameter is multiplied by the corresponding weighting coefficient to obtain a set of weighted motion parameter component sequences; finally, all weighted motion parameter component sequences are arithmetically summed at each sampling time point to generate a one-dimensional comprehensive motion reference signal. Among them, the motion parameter component with a larger weighting coefficient has a higher contribution to the fused signal, thus ensuring that the comprehensive motion reference signal can prominently represent the motion pattern most related to motion artifacts in the fetal heart rate signal. In other embodiments, the weight allocation strategy of each dimension can also be dynamically adjusted according to the principal component analysis results of the motion type, which is not limited in this application.

[0024] It should be noted that the composite motion reference signal mentioned in this application refers to a one-dimensional time series signal generated by weighted fusion of six-degree-of-freedom motion parameters, which is used to centrally characterize the composite motion pattern most related to motion artifacts in the fetal heart rate signal.

[0025] In specific implementation, an adaptive filter model is constructed based on the integrated motion reference signal and the fetal heart rate signal. The artifact signal component during maternal movement can then be derived from the fetal heart rate signal using the following method: The integrated motion reference signal is used as the reference input signal of the filter, and the fetal heart rate signal is used as the desired response signal. A weight vector for a finite-length unit impulse response filter is initialized. Subsequently, the filter enters an iterative update process. At each sampling time, the current weights are used to convolve the reference input signal to obtain an estimate of the artifact signal component at the current time. This estimate is then subtracted from the desired response signal to obtain the instantaneous error. The signal is then processed; then, based on the normalized ratio of the instantaneous error signal to the current reference input signal power, the filter weight vector is updated according to a preset step size factor; after continuous iteration, when the average power of the instantaneous error signal decreases and stabilizes below a preset convergence threshold, it is determined that the filter has reached a stable state. At this time, the output signal of the filter, i.e., the estimated value of the artifact signal component, is used as the finally constructed artifact signal component; wherein, the order of the filter is preset according to the typical duration of motion artifacts, and the step size factor and convergence threshold are set according to the system's requirements for convergence speed and steady-state accuracy, which are not limited in this application.

[0026] It should be noted that the artifact signal component mentioned in this application refers to the interference signal component related to the pregnant woman's movement separated from the original fetal heart signal, which is used to characterize the signal distortion introduced by external factors such as probe displacement and body movement.

[0027] In step 103, the artifact signal component and the fetal heart signal are time-domain aligned and differentiated to obtain the mother's pure fetal heart signal. Then, based on the banding characteristics of the motion artifact, the data loss interval formed by the artifact residue is located in the joint time-frequency domain of the pure fetal heart signal.

[0028] In some embodiments, the process of performing time-domain aligned differential analysis between the artifact signal component and the fetal heart rate signal to obtain the mother's pure fetal heart rate signal can be achieved through the following steps: Determine the transmission delay between the artifact signal component and the fetal heart rate signal; The artifact signal component is time-delayed by the transmission delay, so that the artifact signal component is aligned with the time domain of the fetal heart signal. The artifact signal components after time-domain alignment are differentially analyzed with the fetal heart signal to obtain the pure fetal heart signal of the mother.

[0029] For specific implementation, refer to Figure 3As shown in the figure, this is a schematic flowchart illustrating the process of determining the transmission delay in some embodiments of this application. The determination of the transmission delay between the artifact signal component and the fetal heart rate signal can be achieved in the following ways: First, the artifact signal component and the fetal heart rate signal are time-series aligned to ensure they are based on the same time reference; then, the cross-correlation function between the artifact signal component and the fetal heart rate signal within a preset time delay range is calculated, where the preset time delay range is set according to the inherent delay characteristics of the signal processing link; finally, by locating the peak point with the largest amplitude in the cross-correlation function sequence, the specific time offset corresponding to this peak point is determined as the transmission delay. The calculation of the cross-correlation function can employ a frequency domain correlation method based on Fast Fourier Transform to improve computational efficiency. The upper limit of the preset time delay range is determined based on the signal sampling rate and the system's maximum expected delay, which is not limited in this application.

[0030] In specific implementation, the artifact signal component is time-delay compensated by the transmission delay to align the artifact signal component with the fetal heart rate signal in the time domain. This can be achieved in the following way: First, the transmission delay is decomposed into two parts: an integer sampling point offset and a fractional sampling point offset. Then, the artifact signal component is first subjected to a cyclic shift operation of integer sampling points, and then a fractional delay filter based on the Lagrange interpolation algorithm is used to accurately compensate for the fractional offset. Finally, the alignment effect is verified by calculating the timing deviation between the time-delay compensated artifact signal component and the fetal heart rate signal at feature points. When the timing deviation is less than one sampling interval, the time domain alignment requirement is determined to be met. The order of the fractional delay filter is preset according to the system's requirements for time delay compensation accuracy. In scenarios requiring high-precision compensation, a higher-order filter structure can be selected, but this application does not limit this.

[0031] In specific implementation, the time-domain aligned artifact signal component and the fetal heart rate signal are differentially processed to obtain the mother's pure fetal heart rate signal. This can be achieved in the following way: First, the time-domain aligned artifact signal component and the fetal heart rate signal are subtracted point by point to obtain an initial differential signal; then, the initial differential signal is bandpass filtered, and the passband frequency range of the filter used is consistent with the passband of the bandpass filter used when extracting the baseband fetal heart rate signal; finally, the filtered signal is amplitude normalized so that its amplitude range conforms to the amplitude standard preset based on the statistical characteristics of typical fetal heart rate signals, and the final pure fetal heart rate signal is output. The design of the bandpass filter must ensure that the phase response within the passband is linear to avoid waveform distortion. In application scenarios with high signal fidelity requirements, a finite-length unit impulse response filter with linear phase characteristics can be used, but this application does not limit this.

[0032] It should be noted that the pure fetal heart signal mentioned in this application refers to the physiological signal generated by the fetal heartbeat that is retained after motion artifact separation processing, which is used to reflect the true fetal heart rate information after removing motion interference.

[0033] In some embodiments, the data loss region formed by artifact residue can be located in the joint time-frequency domain of the pure fetal heart signal based on the banding characteristics of motion artifacts by the following steps: Time-frequency analysis is performed on the pure fetal heart rate signal to generate a joint time-frequency distribution of the pure fetal heart rate signal; Identify anomalous energy regions in the joint time-frequency distribution that conform to the characteristics of motion artifact stripes; Based on the time-frequency characteristics of the abnormal energy region, the data loss interval formed by artifact residue is located.

[0034] In specific implementation, time-frequency analysis is performed on the pure fetal heart rate signal to generate the joint time-frequency distribution of the pure fetal heart rate signal. This can be achieved in the following ways: First, the pure fetal heart rate signal is preprocessed to eliminate baseline drift and unify signal amplitude; then, continuous wavelet transform is used as the core time-frequency analysis tool, and Morlet wavelet is selected as the mother wavelet function, based on its Gaussian envelope characteristics and its balanced advantage in time-frequency focusing; during the transform process, the scale parameter conversion is performed to convert the scale sequence into the corresponding physical frequency sequence through a preset pseudo-frequency mapping relationship. The range of this frequency sequence covers and is slightly larger than the complete physiological range of fetal heart rate; the step of the translation parameter is set to a sampling time interval to retain the highest time resolution; finally, by calculating the wavelet coefficient amplitude at each time-frequency grid point, a two-dimensional time-frequency energy distribution map with time as the horizontal axis, physical frequency as the vertical axis, and coefficient amplitude as the intensity is generated as the joint time-frequency distribution of the pure fetal heart rate signal; in other embodiments, if it is necessary to reduce computational complexity, short-time Fourier transform can be used and the window function length can be determined according to the highest frequency of fetal heart rate change. This application does not limit this.

[0035] It should be noted that the joint time-frequency distribution mentioned in this application refers to the energy distribution map obtained by mapping the pure fetal heart signal to a two-dimensional time-frequency space using time-frequency analysis tools. This map is used to simultaneously display the dynamic change characteristics of the signal in the time and frequency domains, thereby providing a visual analysis basis for the identification of motion artifact residues.

[0036] In specific implementation, identifying anomalous energy regions that conform to the characteristics of motion artifact stripes in the joint time-frequency distribution can be achieved in the following way: First, on the time-frequency energy distribution map, the time-frequency texture is quantified by calculating the energy entropy in the local frequency band near each time point and the gradient features along the frequency direction; then, based on the characteristics of motion artifacts as high-energy, narrow-band, and long-duration stripes in the time-frequency domain, specific identification criteria are formulated. These criteria require that candidate regions simultaneously meet the following conditions: local energy exceeds the dynamic threshold obtained from the statistical analysis of historical segments of pure fetal heart signals, the bandwidth of the main energy concentration is less than the preset maximum bandwidth, and the continuous length on the time axis is greater than the typical duration of a single fetal heart cycle; finally, a two-dimensional detection window that slides on the time-frequency plane traverses the entire distribution map, and the regions within the window are matched and verified according to the above criteria, marking all regions that simultaneously meet all conditions as anomalous energy regions; wherein, the dynamic threshold, maximum bandwidth, and other parameters can be initialized based on statistical analysis of known artifact-free training data, and this application does not limit this.

[0037] It should be noted that the abnormal energy region mentioned in this application refers to a local energy concentration region that conforms to the characteristics of motion artifacts in the joint time-frequency distribution, and is used to identify the signal distortion period and frequency band caused by motion artifact residue.

[0038] In specific implementation, based on the time-frequency characteristics of the abnormal energy region, the location of the data loss interval formed by artifact residue can be achieved in the following way, for example: First, perform morphological closing operation on the marked abnormal energy region to close the small holes caused by noise in the region and smooth its boundaries; then, project each processed connected region along the time axis, and convert its two-dimensional position information on the time-frequency map into one-dimensional time period information, that is, determine the start time point and end time point corresponding to each abnormal region; finally, merge the intervals projected onto the time axis according to the principle of time proximity. If the time interval between two intervals is less than a preset merging threshold (this threshold can be set according to the periodicity of the fetal heart rate), then merge them into a continuous data loss interval, thereby locating the data loss interval formed by artifact residue; wherein, the size of the structuring element of the morphological closing operation can be preset according to the resolution of the time-frequency map to achieve a balance between maintaining the region shape and eliminating noise, which is not limited in this application.

[0039] It should be noted that the data loss interval mentioned in this application refers to the time segment in the pure fetal heart signal that is unreliable due to residual motion artifacts, and is used to mark the specific time period that needs signal reconstruction.

[0040] In step 104, the data loss interval is reconstructed based on the signal trend of the pure fetal heart signal in a preset neighborhood, thereby generating a complete fetal heart monitoring waveform.

[0041] In some embodiments, the process of reconstructing the data loss interval based on the signal trend of the pure fetal heart signal in a preset neighborhood, thereby generating a complete fetal heart monitoring waveform, can be achieved through the following steps: Determine the preset neighborhood range corresponding to the data loss interval in the pure fetal heart rate signal; Extract the signal trend features within the preset neighborhood range; Based on the signal trend characteristics, the signal is reconstructed in the data loss area to generate a complete fetal heart rate monitoring waveform.

[0042] In specific implementation, determining the preset neighborhood range corresponding to the data defect interval in the pure fetal heart rate signal can be achieved in the following ways, for example: First, obtain the start and end time points of the data defect interval and calculate the time length of the interval; then, based on the available signal segments adjacent to the data defect interval in the pure fetal heart rate signal, estimate the instantaneous fetal heart rate in the vicinity of the data defect interval by detecting the R-wave peak value or calculating the autocorrelation function, and calculate the corresponding fetal heart rate cycle duration accordingly; finally, take the start and end points of the data defect interval as the center, extend forward and backward by several complete fetal heart rate cycles as the preset neighborhood range, wherein the number of extended cycles is dynamically determined according to the ratio of the time length of the data defect interval to the fetal heart rate cycle duration to ensure that the neighborhood range can fully reflect the changing trend of the signal at the boundary of the defect interval; when the data defect interval is located at the start or end of the signal and causes insufficient signal on one side, only the available complete cycle signal segments on one side are used as the neighborhood, and this application does not limit this.

[0043] In specific implementation, the signal trend features extracted within the preset neighborhood range can be achieved in the following ways: First, the pure fetal heart rate signal within the preset neighborhood range is detrended, and its mean, variance, skewness, and kurtosis are calculated as time-domain statistical features. At the same time, its power spectral density is calculated using Fast Fourier Transform to extract the dominant frequency components and frequency band energy distribution as frequency-domain features. Then, the Burg algorithm is used to solve the autoregressive model coefficients of the neighborhood signal. This algorithm estimates the model parameters by minimizing the sum of the forward and backward prediction error powers, thereby stably capturing the short-term linear prediction trend of the signal. Finally, the autoregressive model coefficients, time-domain statistical features, and frequency-domain features are combined to form a multidimensional signal trend feature vector. The order of the autoregressive model can be automatically determined from the neighborhood signal using model selection criteria such as the Akaike information criterion or the Bayesian information criterion, which is not limited in this application.

[0044] It should be noted that the signal trend characteristics described in this application refer to the time-frequency statistical characteristics and prediction model parameters of the pure fetal heart signal extracted from the preset neighborhood range, which are used to characterize the change pattern of the signal at the boundary of the defect interval, thereby providing a basis for trend extrapolation for data reconstruction.

[0045] In specific implementation, the signal reconstruction of the data-deficient interval based on the signal trend characteristics to generate a complete fetal heart rate monitoring waveform can be achieved in the following way: First, a prediction filter is constructed using the autoregressive model coefficients in the signal trend feature vector. Taking the sampling points at the end of the neighboring signal as the initial state, a point-by-point forward prediction is performed on the data-deficient interval to generate an initial reconstructed signal segment. Then, the spectrum of the initial reconstructed signal is constrained according to the frequency domain characteristics (dominant frequency components) in the feature vector to ensure that its dominant frequency falls within the physiological range of the fetal heart rate, and based on the time domain statistical characteristics... The amplitude information in the model is used to scale the amplitude of the reconstructed signal to make it consistent with the amplitude characteristics of the neighboring signals. Finally, the reconstructed signal segment is spliced ​​with the original pure fetal heart rate signal using a weighted overlapping addition method. At the splicing point, a raised cosine window function is used to window the ends of the reconstructed signal and the original signal, and cross-gradual superposition is performed within a window length to eliminate splicing transients. Finally, a complete fetal heart rate monitoring waveform that is continuous in the time domain and physiologically reasonable in morphology is output. The length of the window function can be adjusted according to the variability of the fetal heart rate, and this application does not limit it.

[0046] It should be noted that the fetal heart rate monitoring waveform described in this application refers to a complete and continuous fetal heart rate waveform obtained after signal reconstruction processing, used to provide reliable data for monitoring fetal physiological status. In step 105, the intrauterine vital signs of the parturient are remotely monitored using the fetal heart rate monitoring waveform.

[0047] In some embodiments, remote monitoring of the mother's intrauterine vital signs using the fetal heart rate monitoring waveform can be achieved through the following steps: Clinical characteristic parameters of fetal heart rate were extracted from the complete fetal heart rate monitoring waveform. The intrauterine vital signs of the parturient are remotely monitored using the aforementioned clinical characteristic parameters.

[0048] In specific implementation, extracting clinical feature parameters of fetal heart rate from the complete fetal heart rate monitoring waveform can be achieved in the following ways: First, digital filtering is performed on the complete fetal heart rate monitoring waveform to eliminate baseline drift; then, a peak detection algorithm based on slope threshold and amplitude threshold is used to identify the peak points of fetal heartbeats in the waveform, and the instantaneous fetal heart rate sequence is obtained by calculating the time interval sequence between adjacent peak points; next, feature parameters including average fetal heart rate, fetal heart rate variability, number of fetal heart rate accelerations, and number of fetal heart rate decelerations are extracted from the instantaneous fetal heart rate sequence, wherein the average fetal heart rate... The heart rate is obtained by taking the arithmetic mean of the instantaneous fetal heart rate sequence; the fetal heart rate variability is obtained by calculating the root mean square difference between adjacent fetal heartbeat intervals; the number of fetal heart rate accelerations and decelerations is obtained by detecting fetal heart rate change events in the instantaneous fetal heart rate sequence that exceed a preset clinical threshold and last for an appropriate period of time; finally, all extracted feature parameters are normalized to form standard format clinical feature parameters; wherein, the slope threshold and amplitude threshold in the peak detection algorithm can be adaptively adjusted according to the waveform signal-to-noise ratio, and the detection threshold can be appropriately reduced to improve sensitivity when the signal quality is poor, which is not limited in this application.

[0049] It should be noted that the clinical characteristic parameters mentioned in this application refer to a set of indicators that quantify fetal heart rate characteristics extracted from complete fetal heart rate monitoring waveforms. These indicators are used to objectively assess the intrauterine physiological state of the fetus, thereby providing data support for the automatic analysis and early warning of the remote monitoring system.

[0050] In specific implementation, remote monitoring of intrauterine vital signs of the parturient through the aforementioned clinical characteristic parameters can be achieved in the following ways: First, the clinical characteristic parameters are compared and analyzed with the reference range of normal physiological parameters based on gestational age stratification. When the characteristic parameters exceed the reference range for the corresponding gestational age, different levels of warning information are generated according to the degree of deviation. Then, the complete fetal heart rate monitoring waveform data, clinical characteristic parameter set, and warning information are packaged into a medical data packet conforming to the HL7FHIR standard through an integrated wireless communication module and sent to the remote monitoring center server via 4G / 5G mobile network or Wi-Fi transmission protocol. Finally, in the data receiving and processing system deployed in the remote monitoring center, the incoming data packet is parsed and verified, and the fetal heart rate monitoring waveform and characteristic parameters are displayed in real time on the monitoring interface. When a warning information is received, medical staff are notified for professional interpretation through audio-visual prompts and message push. At the same time, all monitoring data is encrypted and stored in the patient's personal health record. The reference range of normal physiological parameters is established based on large-scale obstetric clinical statistical data, and corresponding parameter boundaries are set for different gestational age segments. In the late stage of pregnancy, the normal range of each parameter is appropriately adjusted according to the fetal development characteristics. This application does not limit this.

[0051] In another aspect, in some embodiments, this application provides a remote monitoring system for maternal vital signs, as described above. Figure 4 The figure is a schematic diagram of the structure of a remote monitoring system for maternal vital signs according to some embodiments of this application. The remote monitoring system 200 for maternal vital signs includes: a data acquisition module 201, a processing module 202, and an execution module 203, which are described below: The acquisition module 201 in this application is mainly used for remotely acquiring the fetal heart rate signal and motion data of the mother. Processing module 202, in this application, is mainly used to separate the artifact signal component of the mother's movement from the fetal heart signal based on the correlation between the motion data and the fetal heart signal; In addition, the processing module 202 in this application is also used to perform time-domain alignment and differential analysis on the artifact signal component and the fetal heart signal to obtain the pure fetal heart signal of the mother, and then locate the data loss interval formed by the artifact residue in the joint time-frequency domain of the pure fetal heart signal based on the strip characteristics of motion artifacts. In addition, the processing module 202 in this application is also used to reconstruct the data loss interval based on the signal trend of the pure fetal heart signal in a preset neighborhood, thereby generating a complete fetal heart monitoring waveform. The execution module 203 in this application is mainly used to remotely monitor the intrauterine vital signs of the parturient through the fetal heart monitoring waveform.

[0052] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described method for remote monitoring of maternal vital signs.

[0053] In some embodiments, reference Figure 5 This figure is an internal structural diagram of a computer device for implementing a method for remote monitoring of maternal vital signs according to some embodiments of this application. The remote monitoring method for maternal vital signs in the above embodiments can be implemented through... Figure 5 The computer device shown is used to implement this, and the computer device 300 includes at least one processor 301, a communication bus 302, a memory 303, and at least one communication interface 304.

[0054] The processor 301 may be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more devices used to control the execution of the remote monitoring method for maternal vital signs in this application.

[0055] The communication bus 302 is used to transmit information between the aforementioned components.

[0056] Memory 303 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CDROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 303 may exist independently and be connected to processor 301 via communication bus 302. Memory 303 may also be integrated with processor 301.

[0057] The memory 303 stores program code for executing the scheme of this application, and its execution is controlled by the processor 301. The processor 301 executes the program code stored in the memory 303. The program code may include one or more software modules. In the above embodiments, the remote monitoring method for maternal vital signs can be implemented by the processor 301 and one or more software modules in the program code in the memory 303.

[0058] Communication interface 304 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0059] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core processor or a multi-core processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0060] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device may be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0061] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for remote monitoring of maternal vital signs.

[0062] In summary, the remote monitoring method and system for maternal vital signs disclosed in this application remotely collects fetal heart rate signals and motion data of the mother; based on the correlation between the motion data and the fetal heart rate signal, the artifact signal component during the mother's movement is separated from the fetal heart rate signal; the artifact signal component and the fetal heart rate signal are time-domain aligned and differentiated to obtain the mother's pure fetal heart rate signal; then, based on the banding characteristics of the motion artifact, the data loss interval formed by the artifact residue is located in the joint time-frequency domain of the pure fetal heart rate signal; based on the signal trend of the pure fetal heart rate signal in a preset neighborhood, the data loss interval is reconstructed to generate a complete fetal heart rate monitoring waveform; the intrauterine vital signs of the mother are remotely monitored through the fetal heart rate monitoring waveform; and the clinical diagnostic characteristics of the fetal heart rate can be completely preserved while eliminating motion artifacts.

[0063] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0064] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for remote monitoring of maternal vital signs, characterized in that, Includes the following steps: Remotely collect fetal heart rate signals and movement data from pregnant women; Based on the correlation between the motion data and the fetal heart rate signal, the artifact signal component during maternal movement is separated from the fetal heart rate signal. The artifact signal components and the fetal heart signal are time-domain aligned and differentiated to obtain the mother's pure fetal heart signal. Then, based on the banding characteristics of motion artifacts, the data loss interval formed by artifact residue is located in the joint time-frequency domain of the pure fetal heart signal. Based on the signal trend of the pure fetal heart signal in a preset neighborhood, the data loss interval is reconstructed to generate a complete fetal heart monitoring waveform. The fetal heart rate monitoring waveform is used to remotely monitor the intrauterine vital signs of the parturient.

2. The method as described in claim 1, characterized in that, The artifact signal component during maternal movement is specifically separated from the fetal heart signal based on the correlation between the motion data and the fetal heart signal, including: Determine the correlation between the motion data and the fetal heart rate signal; By analyzing the correlation between the motion data and the fetal heart rate signal, the six-degree-of-freedom motion parameters in the motion data are fused to generate a comprehensive motion reference signal. An adaptive filter model is constructed based on the integrated motion reference signal and the fetal heart signal, and then the artifact signal component during the mother's movement is obtained from the fetal heart signal.

3. The method as described in claim 1, characterized in that, The process of performing time-domain aligned differential analysis between the artifact signal components and the fetal heart rate signal to obtain the mother's pure fetal heart rate signal specifically includes: Determine the transmission delay between the artifact signal component and the fetal heart rate signal; The artifact signal component is time-delayed by the transmission delay, so that the artifact signal component is aligned with the time domain of the fetal heart signal. The artifact signal components after time-domain alignment are differentially analyzed with the fetal heart signal to obtain the pure fetal heart signal of the mother.

4. The method as described in claim 1, characterized in that, Based on the banding characteristics of motion artifacts, the data loss intervals caused by artifact residues are located in the joint time-frequency domain of the pure fetal heart rate signal, specifically including: Time-frequency analysis is performed on the pure fetal heart rate signal to generate a joint time-frequency distribution of the pure fetal heart rate signal; Identify anomalous energy regions in the joint time-frequency distribution that conform to the characteristics of motion artifact stripes; Based on the time-frequency characteristics of the abnormal energy region, the data loss interval formed by artifact residue is located.

5. The method as described in claim 1, characterized in that, Based on the signal trend of the pure fetal heart rate signal in a preset neighborhood, trend reconstruction is performed on the data defect interval to generate a complete fetal heart rate monitoring waveform. Specifically, this includes: Determine the preset neighborhood range corresponding to the data loss interval in the pure fetal heart rate signal; Extract the signal trend features within the preset neighborhood range; Based on the signal trend characteristics, the signal is reconstructed in the data loss area to generate a complete fetal heart rate monitoring waveform.

6. The method as described in claim 1, characterized in that, Remote monitoring of the mother's intrauterine vital signs using the aforementioned fetal heart rate monitoring waveform specifically includes: Clinical characteristic parameters of fetal heart rate were extracted from the complete fetal heart rate monitoring waveform. The intrauterine vital signs of the parturient are remotely monitored using the aforementioned clinical characteristic parameters.

7. The method as described in claim 1, characterized in that, The clinical characteristic parameters include: mean fetal heart rate, fetal heart rate variability, number of fetal heart rate accelerations, and number of fetal heart rate decelerations.

8. A remote monitoring system for maternal vital signs, characterized in that, include: The acquisition module is used to remotely acquire fetal heart rate signals and movement data from mothers in labor. The processing module is used to separate the artifact signal component of the mother's movement from the fetal heart signal based on the correlation between the motion data and the fetal heart signal. The processing module is also used to perform time-domain alignment and differential analysis on the artifact signal component and the fetal heart signal to obtain the mother's pure fetal heart signal, and then locate the data loss interval formed by the artifact residue in the joint time-frequency domain of the pure fetal heart signal based on the band characteristics of the motion artifact. The processing module is also used to reconstruct the data loss interval based on the signal trend of the pure fetal heart signal in a preset neighborhood, thereby generating a complete fetal heart monitoring waveform. The execution module is used to remotely monitor the intrauterine vital signs of the parturient through the fetal heart rate monitoring waveform.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the remote monitoring method for maternal vital signs as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the remote monitoring method for maternal vital signs as described in any one of claims 1 to 7.

Images