Respiratory state monitoring method based on quantum magnetometer

Abstract

The present application relates to the technical field of respiratory state monitoring, and discloses a respiratory state monitoring method based on a quantum magnetometer, which is applied to a respiratory monitoring system comprising a plurality of quantum magnetometers and a respiratory state processing server. The method comprises: collecting respiratory-related magnetic field data by the quantum magnetometer to generate a data set; receiving real-time data by the server to construct a respiratory state matrix; predicting a respiratory mode by processing the matrix, extracting features, modeling analysis, combining magnetic field strength gradient and respiratory path information, outputting spatiotemporal distribution characteristics and updating the data set; and monitoring the respiratory state in real time according to the distribution characteristics, including mapping regional identification, controlling device action and dynamically allocating monitoring resources. The present method realizes non-contact monitoring by using the high sensitivity of the quantum magnetometer, acquires spatiotemporal characteristics of the respiratory mode through multi-node cooperation and multidimensional analysis, improves monitoring accuracy and real-time performance, and is suitable for medical, sports, safety and other fields, thereby providing an intelligent solution for respiratory state monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of respiratory state monitoring, in particular to a respiratory state monitoring method based on a quantum magnetometer. BACKGROUND

[0002] Respiratory state monitoring is of great significance in the fields of medical health, sports science, safety protection, etc. Traditional respiratory monitoring technologies such as chest strap type respiratory sensors, impedance type respiratory measurements, etc. have problems such as inconvenience to wear, easy to be disturbed by movement, limited monitoring range, etc. For example, chest strap type sensors need to directly contact the human body, and long-term wearing may cause user discomfort, and contact failure may occur when the patient is agitated or sweats a lot, affecting data accuracy; impedance type measurement detects chest electrical impedance changes through electrode sheets, which is easily disturbed by electromyographic signals generated by human body movement, and is limited in application in special groups such as infants and critically ill patients.

[0003] With the development of non-contact monitoring technology, optical and acoustic based respiratory monitoring methods have gradually emerged. Optical methods such as video respiratory monitoring capture chest movements through a camera, but are significantly affected by lighting conditions, and the monitoring effect decreases significantly in dark environments or under shielding, and there is a risk of privacy leakage; acoustic methods such as respiratory sound monitoring rely on microphones to collect respiratory sounds, which are easily disturbed by environmental noise, making it difficult to accurately extract respiratory features in noisy environments, and unable to monitor the spatial distribution of respiration.

[0004] In the prior art, there are still technical bottlenecks in the spatial distribution monitoring and multi-source data fusion analysis of respiratory state. Traditional single-point monitoring cannot obtain the distribution characteristics of respiratory patterns in different areas, making it difficult to meet the demand for group respiratory monitoring or respiratory state analysis in complex scenarios. For example, in sleep monitoring, traditional methods cannot distinguish the differences in respiratory patterns of different individuals in the same monitoring area; in public place safety monitoring, it is difficult to grasp the spatial distribution changes of the respiratory state of the crowd in real time, making it difficult to early warn abnormal situations. In addition, existing monitoring systems do not make full use of the magnetic field environment, and lack respiratory feature extraction and modeling methods based on magnetic field characteristics, resulting in the need to improve the accuracy and real-time performance of respiratory state analysis.

[0005] As a high-sensitivity magnetic field measurement device, quantum magnetometer has the advantages of high measurement accuracy, fast response speed, and non-contact measurement, providing a new technical idea for respiratory monitoring. However, how to combine the magnetic field measurement advantages of quantum magnetometer with respiratory state analysis, build a multi-node collaborative respiratory monitoring system, and realize the spatio-temporal distribution feature extraction and real-time monitoring of respiratory patterns, there is still no mature technical solution. The existing technology lacks a complete method system for feature fusion, modeling analysis and real-time monitoring of magnetic field data collected by quantum magnetometers, and cannot fully exploit the technical potential of quantum magnetometers in respiratory monitoring. SUMMARY

[0006] The present application aims to provide a respiratory state monitoring method based on a quantum magnetometer to solve the problems presented in the background.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solutions: a respiratory state monitoring method based on a quantum magnetometer, applied to a respiratory monitoring system, the system comprising a plurality of quantum magnetometers deployed in a monitoring area and a respiratory state processing server connected to the quantum magnetometers, two adjacent quantum magnetometers being apart by a set distance, the method comprising:

[0008] Collecting magnetic field data related to respiration in the area by the quantum magnetometer to generate a respiratory magnetic field data set;

[0009] Receiving real-time magnetic field data of the plurality of quantum magnetometers by the respiratory state processing server to construct a respiratory state matrix;

[0010] Determining the distribution characteristics of the respiratory pattern according to the respiratory magnetic field data set and the real-time data of each quantum magnetometer, wherein determining the distribution characteristics of the respiratory pattern comprises: processing the respiratory state matrix, extracting features in combination with the respiratory magnetic field data set, and predicting the respiratory pattern according to the magnetic field intensity gradient and the respiratory path information, outputting the spatiotemporal distribution characteristics of the respiratory pattern through a respiratory pattern analysis model, and updating the respiratory magnetic field data set according to the spatiotemporal distribution characteristics;

[0011] Real-time monitoring of the respiratory state according to the distribution characteristics.

[0012] Preferably, determining the distribution characteristics of the respiratory pattern comprises:

[0013] Processing the respiratory state matrix to extract the magnetic field intensity distribution, frequency response characteristics, and respiratory pattern trend;

[0014] Modeling the magnetic field intensity according to the magnetic field intensity distribution and frequency response characteristics, dividing the monitoring area into a plurality of sub-areas and marking the area identifier, associating and matching the magnetic field intensity of the sub-area with the respiratory magnetic field data set, and marking the area identifier in the respiratory magnetic field data set;

[0015] Calculating the magnetic field intensity gradient according to the position of the quantum magnetometer, predicting the distribution of the respiratory pattern according to the magnetic field intensity gradient and the respiratory pattern trend, and calculating the pattern prediction information of each sub-area;

[0016] Constructing a respiratory pattern analysis model, taking the pattern prediction information as the input parameter of the respiratory pattern analysis model, spatially correlating and modeling the pattern prediction information through the respiratory pattern analysis model, and outputting the spatiotemporal distribution characteristics of the respiratory pattern;

[0017] updating the respiratory magnetic field dataset according to the spatiotemporal distribution characteristics, and obtaining distribution characteristics of the respiratory pattern.

[0018] Preferably, the processing of the respiratory state matrix comprises:

[0019] The respiratory state matrix is normalized, the respiratory hot spot area in the matrix is intercepted through a sliding window, the hot spot area is filtered for noise, and a respiratory pattern trend is calculated through a characteristic decomposition algorithm.

[0020] The spatial correlation characteristics of the respiratory state matrix are calculated, the interference intensity between regions, the signal stability coefficient, and the magnetic field blind area index are calculated according to the spatial correlation characteristics, a feature fusion network is constructed, and a frequency response characteristic is calculated through the feature fusion network.

[0021] The time domain characteristics and the frequency domain characteristics collected by each quantum magnetometer are extracted, a respiratory feature vector of the magnetometer is calculated according to the phase difference between the time domain characteristics and the frequency domain characteristics, the quantum magnetometers at different positions are matched according to the respiratory feature vector, and a respiratory pattern trend is calculated.

[0022] Preferably, the magnetic field intensity modeling of the respiratory state matrix comprises:

[0023] According to the magnetic field intensity distribution, a magnetic field intensity sampling point is extracted in each frame of data, the sampling point is associated and mapped with the frequency response characteristic, a magnetic field intensity atlas is generated, the magnetic field intensity atlases collected by multiple magnetometers are spatially aligned, and the magnetic field intensity distribution of a region is calculated.

[0024] An interference threshold value is set, an interference source is located according to the magnetic field intensity values of multiple respiratory state matrices, an interference intensity difference value is calculated, if the interference intensity difference value is greater than or equal to the interference threshold value, it indicates that there is a magnetic field blind area in the region, a propagation model constraint compensation is performed on the current region, the magnetic field intensity distribution of the current region is iteratively corrected according to the path loss model corresponding to the current region, and the intensity compensation value of the blind area magnetic field is calculated according to the correction result.

[0025] The magnetic field intensity distribution is used to model the respiratory state matrix, and the magnetic field model of a region is labeled in response to the frequency response characteristic.

[0026] Preferably, the calculation of the magnetic field intensity gradient according to the position of the quantum magnetometer comprises:

[0027] According to multiple sets of magnetic field monitoring data, a magnetic field intensity change point is extracted, and the change point is mapped to a unified geographic coordinate system according to the deployment position of the magnetometer, the change point is fitted through a spatial interpolation algorithm, and a magnetic field model of a region is generated.

[0028] Sampling at equal intervals along the respiratory path of the magnetic field model, calculating the magnetic field decay rate, interference fluctuation index and magnetic field change slope of the path according to the sampling results, and calculating the magnetic field change parameter according to the magnetic field decay rate, interference fluctuation index and magnetic field change slope;

[0029] According to the deployment parameters and acquisition accuracy of the quantum magnetometer, the distribution characteristics of the respiratory mode in each frame of data are projected to the magnetic field model, the respiratory direction of the magnetic field model is partitioned according to the number of magnetometers, the change law of the respiratory mode in the partition is analyzed, and the mode distribution characteristics are calculated according to the change law;

[0030] According to the magnetic field change parameter and the mode distribution characteristics, the magnetic field intensity gradient is calculated, and the calculation process of the magnetic field intensity gradient includes: selecting a spatial coordinate point in the deployment direction of the magnetometer based on the geographical position range from the first quantum magnetometer to the last quantum magnetometer, accumulating the product of the magnetic field field strength characteristic weight value and the respiratory mode distribution characteristic weight value in the spatial resolution range, and superimposing the influence value of the magnetometer acquisition frequency on the magnetic field intensity change rate.

[0031] Preferably, the calculation of the mode prediction information of each sub-region includes:

[0032] Taking the respiratory main path of the magnetic field model as a reference line and the peak position of the respiratory mode in each frame of data as a reference point, the mode offset is calculated, and the mode distribution curve is drawn according to the geographical coordinates;

[0033] According to the magnetic field intensity gradient, the change rate and direction in the respiratory mode trend are corrected;

[0034] Starting from the nearest mode distribution point, the distribution curve is continued to be drawn according to the correction results of the change rate and direction, the next period mode distribution point is generated, and the mode prediction information is generated until the distribution points cover the entire target area.

[0035] Preferably, the construction of the respiratory mode analysis model includes:

[0036] The input layer is used to organize the mode prediction information into spatial distribution data and perform standardization processing;

[0037] The feature fusion layer is used to extract the regional correlation characteristics of the respiratory mode by processing the spatial distribution data, and construct the dependency relationship between physiological units;

[0038] The state monitoring layer is used to integrate the correlation relationship of the respiratory mode in the spatial unit, and generate a respiratory state monitoring strategy.

[0039] Preferably, the acquisition of the distribution characteristics of the respiratory mode includes:

[0040] According to the spatio-temporal distribution characteristics of the breathing pattern output by the breathing pattern analysis model, the identification of the sub-region is corresponded to the spatio-temporal distribution characteristics;

[0041] According to the spatio-temporal characteristics of the region data in the respiratory magnetic field data set, the region distribution graph sorted by the intensity of the breathing pattern is generated by reorganizing the region data;

[0042] According to the reorganized region distribution graph, the optimized breathing pattern distribution characteristics are output.

[0043] Preferably, the real-time monitoring of the breathing state comprises:

[0044] The region identification is one-to-one mapped with the region of the distribution characteristics of the breathing pattern;

[0045] According to the spatio-temporal distribution characteristics of the breathing pattern, the execution action of the monitoring device is controlled, including the sensor sensitivity adjustment, the filter parameter setting and the sampling frequency operation;

[0046] According to the spatial distribution of the breathing pattern and the preset monitoring strategy, the monitoring resource is dynamically allocated to the corresponding physiological region.

[0047] Preferably, the control of the execution action of the monitoring device comprises:

[0048] When the breathing pattern reaches the preset intensity threshold in the target region, the sensitivity improvement instruction of the adjacent sensor is triggered;

[0049] According to the filter strategy, the available filter resources are dynamically combined to generate a sampling frequency vector;

[0050] Based on the sampling frequency vector, the acquisition parameters of the target region monitoring node are adjusted.

[0051] Compared with the prior art, the beneficial effects of the present application are:

[0052] The non-contact respiratory monitoring is carried out by using the quantum magnetometer, which avoids the problems of traditional contact type sensors, such as inconvenience to wear and easy to be disturbed by movement, and improves the comfort and adaptability of monitoring. The quantum magnetometer does not need to directly contact the human body, and can be applied to special groups such as infants, patients and respiratory monitoring in sports scenes, which widens the application scenarios. Through the cooperative acquisition of magnetic field data by multiple quantum magnetometers, the spatial distribution characteristics of the breathing pattern can be obtained, and the problem that the traditional single-point monitoring cannot reflect the spatial difference of the breathing state is solved. For example, in sleep monitoring, the breathing patterns of different bed personnel in the same room can be distinguished, and data support is provided for respiratory health analysis of multiple groups; in public places, the spatial distribution of the breathing state of the crowd can be monitored in real time, and abnormal areas can be found in time to provide technical support for public safety management.

[0053] By standardizing the respiratory state matrix, filtering noise, and feature decomposition, key information such as magnetic field strength distribution, frequency response characteristics, and respiratory pattern trends is effectively extracted. Combined with magnetic field strength modeling and interference source positioning, the sub-regional division and magnetic field strength correction of the monitoring area are realized, improving the accuracy and reliability of the data. For example, by setting the interference threshold and path loss model, the magnetic field blind area is compensated and corrected, ensuring the effectiveness of the respiratory data in complex magnetic field environments and improving the anti-interference ability of the system.

[0054] Based on the quantum magnetometer position calculation of the magnetic field strength gradient, combined with the respiratory pattern trend for prediction, a respiratory pattern analysis model is constructed to realize the dynamic prediction and update of the spatio-temporal distribution characteristics of the respiratory pattern. This model can deeply analyze the correlation between respiratory patterns in different sub-regions through spatial correlation modeling, providing a scientific basis for real-time monitoring of respiratory status. For example, by generating a pattern distribution curve based on pattern prediction information, the trend of respiratory pattern changes can be predicted in advance, providing a time window for medical intervention or safety warning.

[0055] According to the spatio-temporal distribution characteristics of the respiratory pattern, the execution actions of the monitoring equipment are dynamically adjusted, such as sensor sensitivity, filtering parameters, and sampling frequency, to realize the optimal allocation of monitoring resources. When the respiratory pattern reaches the preset intensity threshold in the target area, the sensitivity of the adjacent sensor is automatically improved to ensure accurate monitoring of the key area. By dynamically combining filtering resources and adjusting the sampling frequency, the efficiency and quality of data acquisition are improved, and the system energy consumption is reduced. At the same time, based on the spatial distribution of the respiratory pattern, the monitoring resources are dynamically allocated to the corresponding physiological regions, realizing the intelligentization and refinement of monitoring, and improving the monitoring efficiency and response speed of the system.

[0056] The respiratory state monitoring method based on quantum magnetometer constructed by the present application forms a complete technical chain from data acquisition, feature extraction, modeling analysis to real-time monitoring, providing a high-precision, high-reliability, and intelligent solution for respiratory monitoring. This solution not only improves the accuracy and real-time performance of respiratory state analysis, but also provides new technical means for medical diagnosis, sports health management, public safety monitoring, and other fields, with significant social and economic benefits and broad application prospects. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 The working principle diagram of the respiratory state monitoring method based on quantum magnetometer described in the present application;

[0058] Figure 2 The working principle diagram for determining the respiratory pattern distribution characteristics;

[0059] Figure 3 The working principle diagram for modeling the magnetic field strength of the respiratory state matrix;

[0060] Figure 4 A schematic diagram of the working principle of calculating the magnetic field strength gradient according to the quantum magnetometer position. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0062] Please refer to Figures 1-4 The present application provides a respiratory state monitoring method based on quantum magnetometer, applied to a respiratory monitoring system, which includes multiple quantum magnetometers deployed in a monitoring area and a respiratory state processing server connected to the quantum magnetometers, and two adjacent quantum magnetometers are apart by a set distance. The specific implementation steps are as follows:

[0063] The quantum magnetometer collects the magnetic field data related to respiration in the area to generate a respiratory magnetic field data set. In actual application, the quantum magnetometer samples the magnetic field in the monitoring area in real time according to a predetermined sampling frequency, preliminarily preprocesses the collected raw magnetic field data, such as denoising and filtering, and then stores the processed data to form a respiratory magnetic field data set.

[0064] The respiratory state processing server receives the real-time magnetic field data of multiple quantum magnetometers and constructs a respiratory state matrix. The respiratory state processing server establishes a communication connection with each quantum magnetometer, acquires the magnetic field data collected by each quantum magnetometer in real time, and organizes these data into a multi-dimensional respiratory state matrix according to the position information and collection time sequence of the quantum magnetometer, which can reflect the magnetic field state of different positions in the monitoring area at different time points.

[0065] According to the respiratory magnetic field data set and the real-time data of each quantum magnetometer, the distribution characteristics of the respiratory pattern are determined. This process includes processing the respiratory state matrix, extracting features in combination with the respiratory magnetic field data set, predicting the respiratory pattern according to the magnetic field strength gradient and the respiratory path information, outputting the spatio-temporal distribution characteristics of the respiratory pattern through the respiratory pattern analysis model, and updating the respiratory magnetic field data set according to the spatio-temporal distribution characteristics.

[0066] According to the distribution characteristics, the respiratory state is monitored in real time.

[0067] Embodiment 1:

[0068] In determining the distribution characteristics of the breathing pattern, the breathing state matrix needs to be processed first to extract the magnetic field intensity distribution, frequency response characteristics and breathing pattern trend. In actual operation, the breathing state matrix is constructed by the real-time magnetic field data of multiple quantum magnetometers received by the breathing state processing server, and the matrix contains the magnetic field information at different positions and different time points in the monitoring area. Due to the large amount of data and non-uniform format of the original breathing state matrix, it needs to be standardized first. Through standardization, the data collected by different quantum magnetometers can be compared, laying a foundation for subsequent analysis.

[0069] After standardization, the sliding window technique is used to intercept the breathing hot spot area in the matrix. The size and moving step of the sliding window can be adjusted according to the actual situation, and the purpose is to accurately capture the area with significant magnetic field changes related to breathing. After intercepting the breathing hot spot area, noise filtering is needed for this area, because in the actual data collection process, it is inevitable to be disturbed by various external factors, producing noise, which will affect the subsequent analysis results. Noise filtering can use various filtering algorithms such as Gaussian filtering and median filtering, which can remove high-frequency noise and random noise in the data and retain the true breathing magnetic field signal.

[0070] After noise filtering, the breathing pattern trend is calculated by the characteristic decomposition algorithm. The characteristic decomposition algorithm can decompose the complex breathing state matrix into several characteristic vectors and eigenvalues. Through the analysis of these characteristic vectors and eigenvalues, the trend of the breathing pattern changing with time can be extracted. For example, the main component of the breathing pattern can be determined by analyzing the size of the eigenvalue, and the change direction of the breathing pattern can be determined by the direction of the characteristic vector.

[0071] At the same time, the spatial correlation characteristics of the breathing state matrix also need to be calculated. Spatial correlation characteristics reflect the correlation of magnetic field changes between different positions in the monitoring area. By calculating the spatial correlation characteristics, the distribution law and propagation characteristics of the magnetic field in space can be understood. According to the spatial correlation characteristics, the interference intensity between regions, the signal stability coefficient and the magnetic field blind area index are further calculated. The interference intensity is used to measure the degree of magnetic field interference between different regions, the signal stability coefficient is used to evaluate the stability of the magnetic field signal, and the magnetic field blind area index is used to judge whether there is a region that cannot be effectively monitored by the magnetic field.

[0072] After obtaining these parameters, a feature fusion network is constructed to fuse these parameters through the network, so as to calculate the frequency response characteristics. The feature fusion network can use neural network algorithm, through the comprehensive analysis of multiple features, to extract the frequency response characteristics that can better reflect the essence of the breathing pattern.

[0073] Furthermore, the time-domain features and frequency-domain features collected by each quantum magnetometer need to be extracted. Time-domain features can reflect the law of magnetic field change over time, such as waveform, amplitude, etc.; frequency-domain features can reflect the distribution of magnetic field at different frequency components, such as frequency peak, bandwidth, etc. According to the phase difference between time-domain features and frequency-domain features, the respiratory feature vector of quantum magnetometer is calculated. The respiratory feature vector is a vector that integrates time-domain and frequency-domain information, which can more comprehensively describe the respiratory magnetic field features collected by quantum magnetometer.

[0074] According to the respiratory feature vector, the quantum magnetometers at different positions are matched in features. Through feature matching, the correlation and consistency between different quantum magnetometers can be determined, so as to calculate the respiratory pattern trend. For example, the similarity between respiratory feature vectors can be calculated to determine whether the respiratory patterns at different positions are consistent, and then the propagation trend of respiratory pattern in space is inferred.

[0075] After completing the processing of the respiratory state matrix and extracting the magnetic field intensity distribution, frequency response features and respiratory pattern trend, the magnetic field intensity modeling of the respiratory state matrix needs to be performed according to these features. The monitoring area is divided into multiple sub-areas, and each sub-area is labeled with a unique area identifier. The division of the area can be reasonably planned according to the deployment position of the quantum magnetometer, the size and shape of the monitoring area, etc.

[0076] According to the magnetic field intensity of the sub-area and the respiratory magnetic field data set, the correlation matching is performed. The respiratory magnetic field data set is generated by collecting the respiratory-related magnetic field data in the area by the quantum magnetometer, which contains a large amount of historical respiratory magnetic field information. Through correlation matching, the current magnetic field intensity data of each sub-area can be compared and analyzed with the historical data, so as to mark the corresponding area identifier in the respiratory magnetic field data set for subsequent query and analysis.

[0077] According to the position of the quantum magnetometer, the magnetic field intensity gradient is calculated, which reflects the rate of change of the magnetic field intensity in space. According to the magnetic field intensity gradient and the previously extracted respiratory pattern trend, the distribution of the respiratory pattern is predicted. In the prediction process, factors such as the propagation characteristics of the magnetic field and the change law of the respiratory pattern need to be considered, so as to calculate the pattern prediction information of each sub-area.

[0078] After obtaining the pattern prediction information, the respiratory pattern analysis model is constructed. The input parameter of the respiratory pattern analysis model is the pattern prediction information, and the spatial correlation modeling of the pattern prediction information is performed through the model to analyze the correlation relationship and spatial distribution law of the respiratory pattern between different sub-areas, so as to output the spatiotemporal distribution characteristics of the respiratory pattern. The spatiotemporal distribution characteristics not only reflect the distribution of the respiratory pattern in space, but also reflect its trend of change over time.

[0079] Embodiment 2:

[0080] When modeling the magnetic field strength of the respiratory state matrix, a series of operations need to be carried out based on the magnetic field strength distribution and frequency response characteristics. In each frame of respiratory state matrix data, the effective sampling points are extracted according to the preset magnetic field strength threshold, which can reflect the key change position of the magnetic field strength in the current frame. These sampling points are associated and mapped with the frequency response characteristics calculated by the feature fusion network, which contains the response characteristics of the magnetic field signal at different frequency bands. Through the mapping, the corresponding relationship between the spatial position and the frequency characteristics is established, and then the magnetic field strength atlas is generated. The atlas presents the spatial distribution and frequency characteristics of the magnetic field strength in the monitoring area in a visual way. Then, the magnetic field strength atlases collected by multiple quantum magnetometers are spatially aligned, and through coordinate transformation and interpolation methods, the atlases of different magnetometers are placed in the same geographic coordinate system, so as to accurately calculate the magnetic field strength distribution of the region.

[0081] The interference threshold value is set, which can be set according to the historical interference data of the monitoring environment and the actual monitoring requirements. Based on the magnetic field strength values of multiple frames of respiratory state matrix, the interference source is located by analyzing the abnormal fluctuations of the magnetic field strength. Specifically, by comparing the magnetic field strength changes at the same position in each frame of data, if the magnetic field strength of a certain region appears continuous abnormal fluctuations in multiple frames of data, it is judged that there may be an interference source in this region. After locating the interference source, the interference intensity difference is calculated, that is, the difference between the magnetic field strength of the interference region and the normal region. If the interference intensity difference is greater than or equal to the preset interference threshold value, it indicates that there is a magnetic field blind area in this region, that is, the magnetic field signal cannot be effectively propagated or accurately collected in this region due to interference.

[0082] When it is determined that there is a magnetic field blind area, the propagation model constraint compensation needs to be carried out for the current region. First, according to the geographic location, surrounding environment and other factors of the current region, the corresponding path loss model is determined, which describes the attenuation law of the magnetic field signal when propagating in this region. Then, based on the path loss model, the magnetic field strength distribution of the current region is iteratively corrected. The specific process is as follows: according to the path loss model, the theoretical magnetic field strength distribution is calculated, which is compared with the actually collected magnetic field strength distribution to find the difference points. After adjusting the parameters of the difference points, the calculation is performed again, and the iteration is repeated until the error between the theoretical distribution and the actual distribution is within an acceptable range. According to the final correction result, the strength compensation value of the blind area magnetic field is calculated, which is used for the subsequent correction of the blind area magnetic field data.

[0083] After the above interference processing and correction are completed, the respiratory state matrix is modeled according to the finally determined magnetic field strength distribution. During modeling, the magnetic field strength data, frequency response characteristics, and corrected compensation values of each sub-region are integrated into the model, so that the model can accurately reflect the magnetic field distribution in the monitoring area. At the same time, the magnetic field model of the region is labeled by the frequency response characteristics, that is, the response characteristics of different regions to different frequency magnetic field signals are marked in the model, such as high frequency response region, low frequency response region, etc., so as to facilitate subsequent analysis and prediction of the respiratory pattern.

[0084] During the entire magnetic field strength modeling process, data needs to be processed and optimized continuously. For example, when extracting sampling points, the threshold needs to be adjusted according to the actual situation to ensure the effectiveness of the sampling points; when performing spatial alignment, the deployment position error and measurement error of the quantum magnetometer need to be considered, and a suitable algorithm is used to improve the alignment accuracy; when calculating the interference strength difference, accidental factors causing fluctuations need to be excluded to ensure the accuracy of the judgment of the magnetic field blind area. Through the above series of meticulous operations, an accurate and reliable magnetic field strength model can be established, providing a solid data foundation for subsequent respiratory pattern analysis and respiratory state monitoring. The modeling process fully considers various interference factors and magnetic field propagation characteristics in the actual monitoring environment, improves the accuracy and adaptability of the model through scientific processing methods and processes, and makes the respiratory state matrix more truly reflect the respiratory-related magnetic field changes, thereby laying a foundation for accurate respiratory state monitoring.

[0085] Embodiment 3:

[0086] When calculating the magnetic field strength gradient according to the position of the quantum magnetometer, multiple sets of magnetic field monitoring data need to be processed. From these data, the magnetic field strength change points, i.e., the position points where the magnetic field strength changes significantly, are extracted. The extraction of these change points can be realized by setting a threshold for the change of the magnetic field strength, and when the magnetic field strength at a certain position changes beyond the threshold, it is marked as a change point. According to the deployment position of the quantum magnetometer, these change points are mapped to a unified geographic coordinate system. The unified geographic coordinate system provides a reference for subsequent spatial analysis, ensuring that change points at different positions can be processed in the same spatial framework. The change points are fitted by a spatial interpolation algorithm, which can estimate the magnetic field strength value at unknown positions according to the known position and magnetic field strength value of the change points, thereby generating a magnetic field model of the region. This model can more completely describe the distribution of the magnetic field in the monitoring area.

[0087] After the magnetic field field model is generated, equal interval sampling is performed along the breathing path of the magnetic field field model. The breathing path refers to a path where the magnetic field change related to breathing is relatively obvious, and the equal interval sampling can ensure that the obtained data is regular and representative. According to the sampling results, the magnetic field attenuation rate, the interference fluctuation index and the magnetic field change slope of the path are calculated. The magnetic field attenuation rate is used to describe the attenuation speed of the magnetic field strength along the breathing path, the interference fluctuation index reflects the degree of disturbance of the magnetic field in the sampling process, and the magnetic field change slope reflects the speed of change of the magnetic field strength on the path. The three parameters reflect the change characteristics of the magnetic field on the breathing path from different angles, and the magnetic field change parameter can be calculated according to them, which comprehensively reflects the information of the above three parameters and more comprehensively describes the change of the magnetic field.

[0088] According to the deployment parameters and acquisition accuracy of the quantum magnetometer, the distribution characteristics of the breathing pattern in each frame of data are projected to the magnetic field field model. The deployment parameters include the position, spacing and other information of the quantum magnetometer, and the acquisition accuracy determines the accuracy of the data. The projection process combines the distribution characteristics of the breathing pattern with the magnetic field field model, so that the distribution of the breathing pattern can be reflected in the magnetic field field model. According to the number of quantum magnetometers, the breathing direction of the magnetic field field model is divided into zones. The number and size of the zones can be determined according to the number of magnetometers and the length of the breathing direction, to ensure that there are a reasonable number of quantum magnetometers in each zone for data acquisition. The change rule of the breathing pattern in each zone is analyzed, the time series data of the breathing pattern in each zone is analyzed, the change characteristics and trends of the breathing pattern in different zones are understood, and the pattern distribution characteristics are calculated according to the change rule. The pattern distribution characteristics describe the distribution and change rule of the breathing pattern in space.

[0089] According to the magnetic field change parameter and the pattern distribution characteristics, the magnetic field strength gradient is calculated. The calculation process of the magnetic field strength gradient is as follows: based on the geographical position range from the first quantum magnetometer to the last quantum magnetometer, spatial coordinate points are selected in the deployment direction of the quantum magnetometer. For each selected spatial coordinate point, the product of the magnetic field field strength characteristic weight value and the breathing pattern distribution characteristic weight value in the spatial resolution range is calculated, wherein the magnetic field field strength characteristic weight value reflects the importance of the magnetic field field strength near the spatial coordinate point, and the breathing pattern distribution characteristic weight value reflects the influence degree of the breathing pattern distribution on the point. Then, the influence value of the magnetometer acquisition frequency on the magnetic field strength change rate is superimposed, and the higher the magnetometer acquisition frequency, the more accurate the measurement of the magnetic field strength change rate, and the greater the influence value. It is expressed by the formula as follows:

[0090]

[0091] wherein G represents the magnetic field strength gradient; n is the number of selected spatial coordinate points; w E,iThe magnetic field strength characteristic weight value of the i-th spatial coordinate point, which is related to the change of the magnetic field strength near the point, the more significant the change of the magnetic field strength, the greater the weight value; w P,i The respiratory pattern distribution characteristic weight value of the i-th spatial coordinate point, which depends on the typicality and representativeness of the respiratory pattern in the region where the point is located; f is the acquisition frequency of the quantum magnetometer, with the unit of hertz (Hz), representing the acquisition times per unit time; k is an influence coefficient, which is a dimensionless constant, used to adjust the influence degree of the acquisition frequency on the change rate of the magnetic field strength, and its value can be determined according to the actual monitoring environment and the performance of the quantum magnetometer.

[0092] During the entire calculation process, attention should be paid to the accuracy and reasonableness of each step. For example, when extracting the change points, the threshold should be set reasonably to avoid missing important change points or introducing too many noise points; when performing spatial interpolation, a suitable interpolation algorithm should be selected to ensure that the generated magnetic field field model can accurately reflect the actual magnetic field distribution; when calculating the parameters and weight values, the actual situation should be fully considered to ensure the reasonableness of the parameters and the accuracy of the weight values.

[0093] Example 4:

[0094] When calculating the pattern prediction information of each sub-region, the respiratory main path of the magnetic field field model is taken as the reference line. The magnetic field field model is generated by processing and fitting the magnetic field data collected by multiple quantum magnetometers, which describes the distribution of the magnetic field in the monitoring region, and the respiratory main path is the main path related to the respiratory activity in the magnetic field field model, which is usually the path with the most significant change in magnetic field strength. Taking this main path as the reference line can provide a unified reference framework for subsequent pattern prediction.

[0095] The peak position of the respiratory pattern in each frame of data is taken as the reference point. The peak position of the respiratory pattern refers to the position where the magnetic field strength reaches the maximum value in each frame of data, which reflects the main action area of the respiratory activity in the current frame. By determining these peak positions, the spatial distribution of the respiratory pattern at different time points can be understood. Then, the pattern offset is calculated according to these reference points, which is the deviation degree of the peak position of the current frame relative to the reference line or the peak position of the previous frame, which can be represented by the difference of the spatial coordinates. By calculating the pattern offset, the trend of the change of the respiratory pattern in space can be analyzed.

[0096] After obtaining the mode offset, draw the mode distribution curve according to the geographical coordinates. Geographical coordinates are unified spatial coordinate systems that can accurately represent the spatial position of each location in the monitoring area. When drawing the mode distribution curve, take time as the horizontal axis and spatial coordinates as the vertical axis. The peak position and mode offset of each time point are plotted in the coordinate system to form a curve reflecting the change of the breathing pattern over time. This curve can intuitively show the movement and change of the breathing pattern in space.

[0097] According to the previously calculated magnetic field intensity gradient, the change rate and direction of the breathing pattern trend are corrected. The magnetic field intensity gradient reflects the change rate of the magnetic field intensity in space, which is closely related to the propagation and change of the breathing pattern. Through the magnetic field intensity gradient, the influence of the spatial change of the magnetic field on the breathing pattern can be understood, so as to adjust the change rate and direction of the breathing pattern to make it more consistent with the actual physical law. For example, if the magnetic field intensity gradient is large, it means that the magnetic field changes rapidly in space, and the breathing pattern may be affected by the change of the magnetic field and speed up the propagation speed or change the propagation direction. At this time, the change rate and direction need to be corrected accordingly.

[0098] After completing the correction of the change rate and direction, start from the nearest mode distribution point and continue to draw the distribution curve according to the corrected change rate and direction. The nearest mode distribution point refers to the last mode distribution point before the current time point, which is the starting point of subsequent prediction. According to the corrected change rate and direction, gradually move forward on the time axis, calculate the mode distribution point of the next time point, and plot it on the distribution curve.

[0099] Repeat this process until the distribution points cover the entire target area, thereby generating mode prediction information. The target area refers to the entire area that needs to be monitored for breathing status, and covering the entire target area means that the mode prediction information can reflect the future distribution of the breathing pattern in the entire monitoring area. In this process, new distribution points need to be calculated continuously according to the corrected change rate and direction to ensure the accuracy and reliability of the prediction.

[0100] For example, assume that the monitoring area is a rectangular room, and multiple quantum magnetometers are deployed inside the room. The main axis of the respiratory pattern is the central axis of the room. At a certain moment, the peak position of the respiratory pattern in the first frame of data is located at the front end of the central axis of the room, and the pattern offset is 0. Over time, in the second frame of data, the peak position may have moved a certain distance along the central axis, and the pattern offset is positive. According to the calculation of the magnetic field intensity gradient, it is found that the magnetic field intensity gradient at the back end of the room is larger, which means that the respiratory pattern may speed up when it propagates backward due to the influence of the change of the magnetic field. Therefore, when correcting the change rate and direction of the respiratory pattern trend, the change rate needs to be increased to make the predicted peak position move backward faster. Then, starting from the peak position of the second frame, the peak positions of the third frame, the fourth frame, etc. are predicted according to the corrected change rate and direction, until these peak positions cover the entire room, generating complete pattern prediction information.

[0101] During the entire calculation process, the following points need to be noted: first, the determination of the reference line needs to be accurate, which directly affects the calculation of the pattern offset and the drawing of the distribution curve; second, the determination of the peak position needs to be accurate, which requires careful analysis and processing of each frame of data; third, the calculation of the magnetic field intensity gradient needs to be accurate, which is the key to correcting the respiratory pattern trend; finally, when drawing the distribution curve and generating the pattern prediction information, various possible interference factors and uncertainties need to be considered to ensure the reliability and practicality of the prediction results.

[0102] Example 5:

[0103] When monitoring the respiratory state in real time, the region identifier needs to be mapped one-to-one with the region of the distribution characteristics of the respiratory pattern. For example, in a rectangular monitoring area, the quantum magnetometer is divided into four sub-regions according to the deployment position, marked as region A, region B, region C and region D. Each sub-region corresponds to a specific geographical range, and the spatio-temporal distribution characteristics output by the respiratory pattern analysis model show the intensity and distribution of different respiratory patterns in these regions. At this time, the identifiers of regions A, B, C and D are associated with the respiratory pattern distribution characteristics in their respective regions to establish a clear correspondence, so that the respiratory state information of the region can be quickly obtained according to the region identifier in the future.

[0104] After the mapping is completed, the execution action of the monitoring device needs to be controlled according to the spatiotemporal distribution characteristics of the breathing pattern, which includes sensor sensitivity adjustment, filter parameter setting, and sampling frequency operation. Taking sensor sensitivity adjustment as an example, when the breathing pattern in the target area (such as the above-mentioned area A) reaches the preset intensity threshold, the system will trigger the sensitivity improvement instruction of the adjacent sensor. Assuming that the preset intensity threshold is 80 magnetic field intensity units, when the breathing pattern intensity calculation value in area A reaches 85, the system immediately sends an instruction to the quantum magnetometer deployed in area A and the surrounding area to increase its sensitivity from the default 10% to 30%, so as to more accurately capture the subtle magnetic field changes in this area and obtain more detailed breathing-related data.

[0105] In terms of filter parameter setting, the system dynamically combines available filter resources according to the filter strategy to generate a sampling frequency vector. For example, there are multiple filter methods such as low-pass filtering and band-pass filtering in the system, and each filter method corresponds to different filter parameters. When it is monitored that the breathing pattern in area B has high-frequency interference, the system will automatically select the band-pass filtering method and combine the corresponding filter parameters, such as setting the passband frequency to 5-15 Hz, and at the same time generating the corresponding sampling frequency vector, such as [100 Hz, 150 Hz, 200 Hz], to adapt to different filtering needs. Based on this sampling frequency vector, the acquisition parameters of the monitoring nodes in the target area (area B) are adjusted, so that the monitoring nodes collect data according to the new sampling frequency, improving the quality and effectiveness of the data.

[0106] In addition, according to the spatial distribution of the breathing pattern and the preset monitoring strategy, the monitoring resources are dynamically allocated to the corresponding physiological regions. For example, through the breathing pattern analysis model, it is found that the breathing pattern in the lung corresponding physiological region of area C is more concentrated and has higher intensity, at this time the system will allocate more monitoring resources, such as the sampling frequency of the quantum magnetometer, the data processing priority, etc., to area C according to the preset monitoring strategy. The specific operation may include increasing the sampling frequency of the quantum magnetometer in area C from the original 100 Hz to 200 Hz, and at the same time setting the data transmission priority to the highest, ensuring that the breathing data of this area can be processed and analyzed in priority.

[0107] For another example, when it is monitored that the breathing pattern intensity in area D is low and the distribution is relatively dispersed, the system will judge that this area may be in a state of weak respiratory activity. At this time, in order to reasonably utilize the monitoring resources, the system will reduce the sampling frequency of the quantum magnetometer in area D, such as from 100 Hz to 50 Hz, and at the same time reduce the allocation of processing resources to the data of this area, and concentrate more computing resources to areas with strong respiratory activity, such as area A or area C, to realize the optimal configuration of monitoring resources.

[0108] During the entire real-time monitoring process, it is also necessary to continuously adjust the monitoring strategy according to the spatiotemporal distribution characteristics of the respiratory pattern. For example, when the respiratory pattern moves from region A to region B, the system will track this change in real time and timely shift the allocation focus of the monitoring resources from region A to region B. Specifically, when it is detected that the respiratory pattern intensity in region A begins to decline while the respiratory pattern intensity in region B gradually rises and approaches a preset threshold, the system will gradually reduce the sensitivity of the sensors in region A, while increasing the sensitivity of the sensors in region B, and adjust the sampling frequency and filtering parameters, to ensure that the monitoring device can always accurately monitor the changes in the respiratory pattern.

[0109] In addition, when dynamically allocating monitoring resources, the deployment location and acquisition capacity of the quantum magnetometer need to be considered. For example, there may be 3 quantum magnetometers deployed in region C and 2 quantum magnetometers deployed in region D. When allocating resources, the number and location of the magnetometers in each region need to be considered to reasonably adjust their sampling parameters, to ensure that the data acquisition in the entire monitoring region is comprehensive and focused. At the same time, the bandwidth limitations of data transmission and processing need to be considered to avoid data transmission congestion or processing delay caused by unreasonable resource allocation.

[0110] Through the above series of operations, real-time monitoring of the respiratory state is achieved. This monitoring method can dynamically adjust the working parameters of the monitoring device and the allocation of monitoring resources according to the actual distribution of the respiratory pattern, improve the accuracy and efficiency of monitoring, and reasonably utilize resources to reduce the operating cost of the system. For example, in actual application scenarios, when the patient is in different respiratory states, such as calm breathing, deep breathing, or respiratory abnormalities, the system can monitor the changes in the respiratory pattern in real time and timely adjust the monitoring strategy to provide accurate and real-time data support for medical diagnosis and monitoring.

[0111] It should be noted that in this paper, relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device.

[0112] Although embodiments of the present application have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A respiratory status monitoring method based on quantum magnetometers, applied to a respiratory monitoring system, wherein the system comprises a plurality of quantum magnetometers deployed in a monitoring area and a respiratory status processing server connected to the quantum magnetometers, wherein two adjacent quantum magnetometers are separated by a set distance, and wherein: The method comprises: The quantum magnetometer is used to collect respiration-related magnetic field data in the region to generate a respiratory magnetic field dataset; The respiratory state processing server receives real-time magnetic field data from multiple quantum magnetometers and constructs a respiratory state matrix; The respiratory state matrix is ​​constructed by receiving real-time magnetic field data from multiple quantum magnetometers by the respiratory state processing server. The matrix contains magnetic field information at different locations and time points within the monitoring area. Determining distribution characteristics of a respiratory pattern based on the respiratory magnetic field dataset and real-time data from each quantum magnetometer, wherein determining the distribution characteristics of the respiratory pattern comprises: processing a respiratory state matrix, performing feature extraction in combination with the respiratory magnetic field dataset, predicting the respiratory pattern based on a magnetic field intensity gradient and respiratory path information, outputting spatiotemporal distribution characteristics of the respiratory pattern through a respiratory pattern analysis model, and updating the respiratory magnetic field dataset based on the spatiotemporal distribution characteristics; According to the distribution characteristics, the respiratory state is monitored in real time; Determining the distribution characteristics of the breathing pattern includes: processing the respiratory state matrix to extract magnetic field intensity distribution, frequency response characteristics, and respiratory pattern trends; Performing magnetic field intensity modeling on the respiratory state matrix according to the magnetic field intensity distribution and frequency response characteristics, dividing the monitoring area into multiple sub-areas and marking the area identifiers, correlating and matching the magnetic field intensity of the sub-areas with the respiratory magnetic field dataset, and marking the area identifiers in the respiratory magnetic field dataset; Calculating a magnetic field intensity gradient according to the position of the quantum magnetometer, predicting a distribution of a breathing pattern according to the magnetic field intensity gradient and the breathing pattern trend, and calculating pattern prediction information for each sub-region; Constructing a breathing pattern analysis model, using the pattern prediction information as an input parameter of the breathing pattern analysis model, performing spatial correlation modeling on the pattern prediction information through the breathing pattern analysis model, and outputting spatiotemporal distribution characteristics of the breathing pattern; The respiratory magnetic field dataset is updated according to the spatiotemporal distribution characteristics to obtain the distribution characteristics of the respiratory pattern.

2. The respiratory state monitoring method based on quantum magnetometer according to claim 1, characterized in that: The processing of the respiratory state matrix includes: Normalizing the respiratory state matrix, intercepting the respiratory hotspot area in the matrix through a sliding window, filtering the hotspot area for noise, and calculating the respiratory pattern trend through an eigendecomposition algorithm; Calculating the spatial correlation characteristics of the respiratory state matrix, calculating the interference intensity, signal stability coefficient and magnetic field blind zone index between regions based on the spatial correlation characteristics, constructing a feature fusion network, and calculating the frequency response characteristics through the feature fusion network; The time domain features and frequency domain features collected by each quantum magnetometer are extracted, and the breathing feature vector of the magnetometer is calculated based on the phase difference between the time domain features and the frequency domain features. The quantum magnetometers at different positions are feature matched based on the breathing feature vector, and the breathing pattern trend is calculated.

3. The respiratory state monitoring method based on quantum magnetometer according to claim 2, characterized in that: The magnetic field intensity modeling of the respiratory state matrix includes: According to the magnetic field intensity distribution, magnetic field intensity sampling points are extracted from each frame of data, and the sampling points are correlated and mapped with the frequency response characteristics to generate a magnetic field intensity spectrum. The magnetic field intensity spectrum collected by multiple intensity meters is spatially aligned to calculate the magnetic field intensity distribution of the region; Set the interference threshold value, locate the interference source according to the magnetic field strength value of the multi-frame breathing state matrix, calculate the interference strength difference, if the interference strength difference is greater than or equal to the interference threshold value, it means that there is a magnetic field blind spot in the area, perform propagation model constraint compensation on the current area, iteratively correct the magnetic field strength distribution of the current area according to the path loss model corresponding to the current area, and calculate the intensity compensation value of the blind spot magnetic field based on the correction result; The magnetic field intensity modeling is performed on the respiratory state matrix according to the magnetic field intensity distribution, and the response annotation of the magnetic field model of the region is performed through the frequency response characteristics.

4. The respiratory state monitoring method based on quantum magnetometer according to claim 3, characterized in that: The step of calculating the magnetic field intensity gradient according to the position of the quantum magnetometer comprises: Based on multiple sets of magnetic field monitoring data, the magnetic field intensity change points are extracted and mapped to a unified geographic coordinate system according to the deployment location of the intensity meter. The change points are fitted using a spatial interpolation algorithm to generate a regional magnetic field model. Performing equal-interval sampling along the respiratory path of the magnetic field model, calculating the magnetic field attenuation rate, interference fluctuation index, and magnetic field change slope of the path based on the sampling results, and calculating the magnetic field change parameter based on the magnetic field attenuation rate, interference fluctuation index, and magnetic field change slope; Based on the deployment parameters and acquisition accuracy of the quantum magnetometer, the distribution characteristics of the breathing pattern in each frame of data are projected onto the magnetic field model. The magnetic field model is partitioned along the breathing direction according to the number of magnetometers. The variation pattern of the breathing pattern in the partition is analyzed, and the pattern distribution characteristics are calculated based on the variation pattern. The magnetic field intensity gradient is calculated based on the magnetic field change parameter and the pattern distribution characteristics. The calculation process of the magnetic field intensity gradient includes: based on the geographical location range from the first quantum magnetometer to the last quantum magnetometer, selecting spatial coordinate points in the magnetometer deployment direction, cumulatively calculating the product of the magnetic field intensity characteristic weight value and the breathing pattern distribution characteristic weight value within the spatial resolution range, and superimposing the influence value of the magnetometer acquisition frequency on the magnetic field intensity change rate.

5. The respiratory state monitoring method based on quantum magnetometer according to claim 4, characterized in that: The calculating of the pattern prediction information of each sub-region includes: Using the main breathing diameter of the magnetic field model as a baseline and the peak position of the breathing pattern in each frame of data as a reference point, the pattern offset is calculated and a pattern distribution curve is drawn according to geographic coordinates; Correcting the rate and direction of change in the breathing pattern trend based on the magnetic field intensity gradient; Starting from the nearest pattern distribution point, the distribution curve is continuously drawn according to the correction results of the change rate and direction to generate the pattern distribution points for the next period until the distribution points cover the entire target area and the pattern prediction information is generated.

6. The respiratory state monitoring method based on quantum magnetometer according to claim 1, characterized in that: The constructing of the breathing pattern analysis model comprises: The input layer is used to organize the pattern prediction information into spatial distribution data and perform normalization processing; The feature fusion layer is used to extract regional correlation features of respiratory patterns by processing spatial distribution data and construct dependency relationships between physiological units; The state monitoring layer is used to integrate the association between breathing patterns in spatial units and generate a breathing state monitoring strategy.

7. The respiratory state monitoring method based on quantum magnetometer according to claim 1, characterized in that: The obtaining of distribution characteristics of the breathing pattern includes: According to the spatiotemporal distribution characteristics of the breathing pattern output by the breathing pattern analysis model, the identifiers of the sub-regions are matched with the spatiotemporal distribution characteristics; The regional data in the respiratory magnetic field dataset are reorganized according to the spatiotemporal characteristics to generate a regional distribution map sorted by the intensity of the respiratory pattern; According to the reorganized regional distribution map, the optimized breathing pattern distribution features are output.

8. The respiratory state monitoring method based on quantum magnetometer according to claim 1, characterized in that: The real-time monitoring of the respiratory state includes: Mapping the region identifiers to the regions of the distribution characteristics of the breathing pattern one by one; According to the spatiotemporal distribution characteristics of the breathing pattern, the monitoring equipment is controlled to perform actions, including sensor sensitivity adjustment, filter parameter setting and sampling frequency operation; Based on the spatial distribution of breathing patterns and preset monitoring strategies, monitoring resources are dynamically allocated to corresponding physiological areas.

9. The respiratory state monitoring method based on quantum magnetometer according to claim 8, characterized in that: The execution action of the control monitoring device includes: When the breathing pattern reaches a preset intensity threshold in the target area, it triggers a sensitivity increase instruction for the adjacent sensors; Dynamically combine available filtering resources according to filtering strategies to generate sampling frequency vectors; The acquisition parameters of the target area monitoring nodes are adjusted based on the sampling frequency vector.

Images