Big data intelligent healthy breath dynamic monitoring method, system and terminal

Abstract

The invention relates to the technical field of breath monitoring, in particular to a big data intelligent health breath dynamic monitoring method and system and a terminal. The method comprises the following steps: acquiring a time domain interval of each respiratory cycle according to change characteristics of a respiratory pressure signal; further obtaining work doing amount corresponding to each breathing cycle, and dividing breathing modes; the newest seasonal period of the respiratory pressure signal is further extracted, and a respiratory stability coefficient is obtained according to the change characteristics of work doing amounts of respiratory periods with adjacent time sequences in the seasonal period in combination with the number distribution characteristics of the respiratory periods of the respiratory mode; furthermore, when the breathing stability coefficient is larger than or equal to the preset stability threshold value, the predicted breathing pressure signal is obtained and verified, and finally the predicted breathing mode is obtained according to the final predicted breathing pressure signal, so that the hysteresis of real-time processing is reduced, the response speed of breathing support is increased, and the comfort is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of respiratory monitoring, in particular to a big data intelligent health respiratory dynamic monitoring method, system and terminal. BACKGROUND

[0002] In the prior art, respiratory behavior is usually achieved by monitoring key parameters such as respiratory frequency, tidal volume, airway pressure, minute ventilation, etc. Airflow pressure is generated during the respiratory process, which can reflect the respiratory attributes of the user such as respiratory state and respiratory pattern. Respiratory monitoring masks are usually equipped with gas pressure sensors, which can detect the airway pressure during the inhalation or exhalation process of the patient through the pressure probe installed in the airway.

[0003] In order to realize dynamic monitoring of respiratory behavior, real-time analysis of the current respiratory state of the user, and thus cater to the current respiratory state of the user and maintain the healthy respiration of the user, the prior art generally relies on physiological indicators of the patient and monitoring respiratory work to identify respiratory patterns, such as blood oxygen saturation. However, physiological indicators such as blood oxygen saturation often show corresponding changes only after a period of respiratory state changes, and monitoring respiratory work requires a certain amount of respiratory data to be accumulated to determine respiratory pattern conversion, resulting in a lag in the determination of respiratory patterns by the respiratory monitoring mask through monitoring respiratory work data. SUMMARY

[0004] In order to solve the technical problem of the lag in the determination of respiratory patterns by the respiratory monitoring mask through monitoring respiratory work data, the purpose of the present application is to provide a big data intelligent health respiratory dynamic monitoring method, system and terminal, and the technical solution adopted is as follows: A big data intelligent health respiratory dynamic monitoring method, the method comprising: obtaining a respiratory pressure signal and a respiratory pressure value of a current patient; obtaining a time domain interval of each respiratory cycle of the respiratory pressure signal according to the change characteristics of the respiratory pressure signal; based on all the pressure values in each of the time domain intervals, obtaining the work amount corresponding to each of the respiratory cycles; dividing the respiratory pattern of each of the respiratory cycles based on the work amount; extracting the latest seasonal cycle of the respiratory pressure signal based on the STL algorithm; obtaining a respiratory stability coefficient according to the change characteristics of the work amount of the time series adjacent respiratory cycles in the seasonal cycle, in combination with the number distribution characteristics of the respiratory cycles of the respiratory pattern; When the respiratory stability coefficient is greater than or equal to a preset stability threshold, a predicted respiratory pressure signal is obtained by using a preset prediction model; according to the difference characteristics of the respiratory pressure signal and the predicted respiratory pressure signal in the seasonal period, combined with the difference of the respiratory stability coefficient of the adjacent seasonal period, it is determined whether to re-predict; when it is determined not to re-predict, a predicted respiratory mode is obtained according to the final predicted respiratory pressure signal.

[0005] Further, the method for obtaining the time domain interval comprises: extracting a trend curve of the respiratory pressure signal; and dividing the time domain between two adjacent minimum points in the trend curve into a time domain interval of one respiratory cycle.

[0006] Further, the method for obtaining the respiratory mode comprises: When the work amount is in a first preset work amount interval, the corresponding respiratory cycle is marked as rapid shallow breathing, and the identification serial number is set as 01; when the work amount is in a second preset work amount interval, the corresponding respiratory cycle is marked as normal resting breathing, and the identification serial number is set as 02; when the work amount is in a third preset work amount interval, the corresponding respiratory cycle is marked as deep breathing, and the identification serial number is set as 03; when the work amount is in a fourth preset work amount interval, the corresponding respiratory cycle is marked as forced breathing, and the identification serial number is set as 04.

[0007] Further, the method for obtaining the respiratory stability coefficient comprises: In the latest seasonal period, when the number of respiratory cycles corresponding to a certain respiratory mode is the largest, the respiratory mode is marked as a main respiratory mode, and other respiratory modes are marked as secondary respiratory modes. According to the number of respiratory cycles corresponding to all the secondary respiratory modes in the seasonal period, combined with the absolute value of the difference of the work amount of the time-sequentially adjacent respiratory cycles, a respiratory stability coefficient of the latest seasonal period is obtained; the number of respiratory cycles corresponding to all the secondary respiratory modes and the absolute value of the difference of the work amount of the time-sequentially adjacent respiratory cycles are negatively correlated with the respiratory stability coefficient.

[0008] Further, the method for determining whether to re-predict comprises: obtaining a time domain interval of each respiratory cycle of the predicted respiratory pressure signal; taking the number of respiratory cycles in the latest seasonal period as a comparison cycle number; taking the predicted respiratory pressure signal of the comparison cycle number of respiratory cycles in the time domain in front as a predicted comparison signal, and obtaining the respiratory mode of each respiratory cycle in the predicted comparison signal; extracting a preset number of comparison pairs of the same time sequence number of breathing cycles from the prediction comparison signal and the seasonal period of the respiratory pressure signal; obtaining a prediction credibility according to a difference of the identification number of the breathing mode of the two breathing cycles in each of the comparison pairs, in combination with a difference of the breathing stability coefficient of adjacent seasonal periods; the difference of the identification number is negatively correlated with the prediction credibility; when the prediction credibility is less than a preset credibility threshold, determining to re-predict and re-determining after re-predicting; when the prediction credibility is greater than or equal to the preset credibility threshold, determining not to re-predict.

[0009] Further, the method for obtaining the predicted breathing mode comprises: In the prediction comparison signal of the final predicted respiratory pressure signal, the proportion of the number of breathing cycles of each breathing mode in the total number of all breathing cycles is taken as the proportion of each breathing mode; the breathing mode with the largest proportion is selected as the predicted breathing mode.

[0010] Further, the method for obtaining the work amount comprises: Integrating the pressure values in each of the time domain intervals, and taking the integral value as the work amount corresponding to the breathing cycle.

[0011] Further, the preset prediction model is an ARIMA model.

[0012] The present application also provides a big data intelligent health respiratory dynamic monitoring system, which comprises: a data acquisition module: acquiring the respiratory pressure signal and the pressure value of the current patient; obtaining the time domain interval of each breathing cycle of the respiratory pressure signal according to the change characteristics of the respiratory pressure signal; a breathing analysis module: obtaining the work amount corresponding to each breathing cycle based on all the pressure values in each of the time domain intervals; dividing the breathing mode of each breathing cycle based on the work amount; extracting the latest seasonal period of the respiratory pressure signal based on the STL algorithm; obtaining the breathing stability coefficient according to the change characteristics of the work amount of the time sequence adjacent breathing cycles in the seasonal period, in combination with the number distribution characteristics of the breathing cycles of the breathing mode; a breathing prediction module: when the breathing stability coefficient is greater than or equal to a preset stability threshold, obtaining a predicted respiratory pressure signal by using a preset prediction model; determining whether to re-predict according to the difference characteristics of the respiratory pressure signal and the predicted respiratory pressure signal in the seasonal period, in combination with the difference of the breathing stability coefficient of adjacent seasonal periods; when it is determined not to re-predict, obtaining a predicted breathing mode according to the final predicted respiratory pressure signal.

[0013] The application further provides a big data intelligent health respiratory dynamic monitoring terminal, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the steps of any one of the big data intelligent health respiratory dynamic monitoring methods when executing the computer program.

[0014] The application has the following advantages: The application firstly acquires the respiratory pressure signal and the respiratory pressure value of the current patient, provides a data basis, further acquires the time domain interval of each respiratory cycle of the respiratory pressure signal, divides the respiratory pressure signal into different respiratory cycles, is more conducive to analyzing the local characteristics of the signal and identifying the respiratory mode, further acquires the work amount corresponding to each respiratory cycle, provides a basis for dividing the respiratory mode, further divides the respiratory mode of each respiratory cycle based on the work amount, is convenient for comprehensively understanding the respiratory state of the patient, further extracts the latest seasonal cycle of the respiratory pressure signal based on the STL algorithm, can reflect the current state of the respiratory signal in real time, provides a basis for subsequent respiratory mode analysis and stability evaluation, and at the same time prepares for signal prediction, further acquires the respiratory stability coefficient, represents the respiratory stability characteristics of the patient in the seasonal cycle, provides a judgment basis for whether the respiratory prediction can be performed subsequently, further uses the preset prediction model to acquire the predicted respiratory pressure signal when the respiratory stability coefficient is greater than or equal to the preset stability threshold, provides a data basis for reducing the hysteresis of the respiratory mode judgment of the respiratory monitoring mask, and at the same time guarantees the nursing effect, further judges whether to predict again, ensures that the model can adapt to the actual state at any time, and improves the accuracy and reliability of the prediction, and finally acquires the predicted respiratory mode according to the final predicted respiratory pressure signal when it is judged that the prediction is not performed again. The application identifies the respiratory mode of a single respiratory cycle, analyzes the stability characteristics of the patient's respiration in the seasonal cycle, further acquires the predicted respiratory pressure signal and the predicted respiratory mode, reduces the hysteresis of real-time processing, improves the response speed of respiratory support, and improves the comfort. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0016] Figure 1 A flowchart of a big data intelligent health respiratory dynamic monitoring method provided by one embodiment of the application; Figure 2 A pressure value waveform provided by one embodiment of the application; Figure 3 A flow chart of a method for determining whether to re-predict according to an embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to further clarify the technical means and effects taken by the present application to achieve the predetermined object of the application, the following describes in detail the specific implementation, structure, features and effects of a big data smart health respiratory dynamic monitoring method, system and terminal according to the present application, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0019] The specific scheme of the big data smart health respiratory dynamic monitoring method, system and terminal provided by the present application is described below in combination with the accompanying drawings.

[0020] Please refer to Figure 1 which shows a flow chart of a big data smart health respiratory dynamic monitoring method according to an embodiment of the present application, specifically including: In the embodiment of the present application, the respiratory pressure signal of the patient is obtained in real time by the respiratory monitoring mask equipped with the piezoelectric sensor, and then the respiratory pressure signal is divided into respiratory cycles and respiratory patterns. The existing respiratory pressure signal is predicted to provide the predicted respiratory pressure signal for the respiratory monitoring mask to predict the respiratory pattern, avoiding the hysteresis of the existing determination method and timely adapting to the respiratory changes of the patient.

[0021] Step S1: Obtain the respiratory pressure signal and the respiratory pressure value of the current patient; according to the change characteristics of the respiratory pressure signal, obtain the time domain interval of each respiratory cycle of the respiratory pressure signal.

[0022] In an embodiment of the present application, the piezoelectric sensor is usually installed in the pipeline connecting the airway and the respiratory monitoring mask to accurately monitor the pressure fluctuation in the airway; the sensitivity parameter of the sensor is: 1-10pC / Pa (picocoulomb / pascal); the working pressure range is: 0-100cmH2O (centimeter of water column); the frequency response is: 0.1-500Hz, the real-time detection of the pressure change in the airway is converted into an electrical signal to obtain the respiratory pressure signal and the respiratory pressure value of the current patient, providing a data basis.

[0023] Please refer to Figure 2 which shows a pressure value waveform diagram according to an embodiment of the present application,Figure 2 The horizontal axis is a time axis, and the vertical axis is a pressure value axis.

[0024] Considering that human respiration has obvious periodic characteristics, the respiration pressure signal has obvious periodic changes, and therefore, according to the change characteristics of the respiration pressure signal, the time domain interval of each respiration period of the respiration pressure signal is obtained, the respiration pressure signal is divided into different respiration periods, and the local characteristics of the signal are more conducive to analysis and identification of the respiration mode.

[0025] Preferably, in an embodiment of the present application, considering that the original pressure value waveform curve has a local sawtooth shape, which is not conducive to analysis of the periodic characteristics of the respiration pressure signal, the trend curve of the respiration pressure signal is extracted first, and the change characteristics of the respiration pressure signal are analyzed by means of the trend curve; considering that each respiration causes the trend curve to rise and fall once, the time domain between two adjacent minimum points in the trend curve is divided into a time domain interval of one respiration period, one respiration period contains one expiration and one inspiration, the change characteristics of the trend curve are reflected by the minimum value, and the change characteristics of the respiration pressure signal are reflected.

[0026] As an example, the trend curve of the respiration pressure signal is extracted by a singular spectrum analysis (SSA) algorithm, and then the minimum points are obtained to divide the time domain interval.

[0027] It should be noted that in other embodiments of the present application, the implementer can also obtain the trend curve by means of polynomial fitting, which is prior art and will not be described in detail.

[0028] Step S2: based on all pressure values in each time domain interval, the work amount corresponding to each respiration period is obtained; based on the work amount, each respiration period is divided into a respiration mode; based on an STL algorithm, the latest seasonal period of the respiration pressure signal is extracted; according to the change characteristics of the work amount of the time-sequentially adjacent respiration periods in the seasonal period, in combination with the number distribution characteristics of the respiration periods of the respiration mode, a respiration stability coefficient is obtained.

[0029] Considering that the respiration characteristics of different respiration periods are different, and a more comprehensive understanding of the respiration state of the patient is required, the respiration periods are divided into respiration modes; the work amount of all pressure values in one respiration period represents the work done by the respiration monitoring mask to maintain normal ventilation of the patient, and therefore, based on all pressure values in each time domain interval, the work amount corresponding to each respiration period is obtained, which provides a basis for division of the respiration mode.

[0030] Preferably, in an embodiment of the present application, the pressure values in each time domain interval are integrated, and the integral value is taken as the work amount of the corresponding respiration period.

[0031] It should be noted that the integral method is prior art and will not be described again.

[0032] The increase in work usually relates to the body needing more oxygen, increased airway resistance or lung dysfunction, while the decrease in work indicates shallow and rapid breathing, which is often seen in cases of anxiety or mild hypoxia, so the respiratory pattern can be divided based on the amount of work for each respiratory cycle.

[0033] Preferably, in one embodiment of the application, when the amount of work is in the first preset work interval, the corresponding respiratory cycle is marked as rapid shallow breathing, and the identification serial number is set to 01; when the amount of work is in the second preset work interval, the corresponding respiratory cycle is marked as normal resting breathing, and the identification serial number is set to 02; when the amount of work is in the third preset work interval, the corresponding respiratory cycle is marked as deep breathing, and the identification serial number is set to 03; when the amount of work is in the fourth preset work interval, the corresponding respiratory cycle is marked as forced breathing, and the identification serial number is set to 04.

[0034] As an example, set the average work value under normal breathing to 0.5 to obtain the normalized reference value, linearly normalize the amount of work for each respiratory cycle, and set the work value exceeding 2 times the average work value under normal breathing to 1; set the first preset work interval to [0, 0.4], the second preset work interval to (0.4, 0.6], the third preset work interval to (0.6, 0.8], and the fourth preset work interval to (0.8, 1]. The identification serial number is the identifier of the respiratory pattern, which facilitates subsequent analysis of changes in the respiratory pattern.

[0035] Among them, normal breathing refers to the breathing of a patient under normal resting breathing, and relevant personnel determine and measure a plurality of respiratory cycles under normal breathing in advance, such as 100 (which can involve multiple patients, such as 10 for each patient, involving 10 patients), and calculate the average work value measured for each respiratory cycle.

[0036] The respiratory pressure signal can exhibit non-stationary characteristics due to factors such as patient status, external environment, device changes, etc. The STL (Seasonal-Trend decomposition using Loess) algorithm has good tolerance to noise and outliers, and can extract stable seasonal cycles in the signal, avoiding being affected by short-term interference data.

[0037] Based on the STL algorithm, the latest seasonal cycle of the respiratory pressure signal can be extracted, which can reflect the current state of the respiratory signal in real time, providing a basis for subsequent respiratory pattern analysis and stability evaluation, and also preparing for signal prediction.

[0038] In another embodiment of the present application, the implementer can also set the length of the seasonal cycle in a preset prediction time interval, for example, the preset prediction time interval is 10 minutes, and the length of the seasonal cycle is fixed at 10 minutes.

[0039] Considering that the work amount change feature of the adjacent respiratory cycles can accurately reflect short-term dynamic fluctuations, and combining the number distribution feature of the respiratory patterns, long-term trends and respiratory regularities can be comprehensively evaluated, therefore, according to the work amount change feature of the adjacent respiratory cycles in the time sequence in the seasonal cycle, and combining the number distribution feature of the respiratory cycles of the respiratory patterns, the respiratory stability coefficient is obtained, which represents the respiratory stability feature of the patient in the seasonal cycle, and provides a basis for determining whether to perform respiratory prediction subsequently.

[0040] Preferably, in an embodiment of the present application, considering that the respiratory pattern that appears most frequently in the latest seasonal cycle usually represents the normal respiratory state or the most common respiratory behavior of the patient, therefore, in the latest seasonal cycle, when the number of respiratory cycles corresponding to a certain respiratory pattern is the most, this respiratory pattern is marked as the main respiratory pattern, and other respiratory patterns are marked as secondary respiratory patterns; According to the number of respiratory cycles corresponding to all secondary respiratory patterns in the seasonal cycle, and combining the absolute value of the difference of the work amounts of the adjacent respiratory cycles in the time sequence, the respiratory stability coefficient of the latest seasonal cycle is obtained; Considering that the number of secondary respiratory patterns directly reflects the degree of unstable respiration, and the difference of the work amounts between the respiratory cycles reflects the dynamic fluctuations of the respiratory behavior and also reflects the degree of unstable respiration, therefore, the number of respiratory cycles corresponding to all secondary respiratory patterns, and the absolute value of the difference of the work amounts of the adjacent respiratory cycles in the time sequence are all negatively correlated with the respiratory stability coefficient.

[0041] As an example, the calculation formula of the respiratory stability coefficient includes: ; Wherein, Sta represents the respiratory stability coefficient of the latest seasonal cycle; norm{} represents a linear normalization function; r represents the number of respiratory cycles corresponding to all secondary respiratory patterns in the seasonal cycle; C0 represents a positive parameter except zero, and in this example, C0=0.001; M represents the number of respiratory cycles in the seasonal cycle; m represents the serial number of the respiratory cycle in the seasonal cycle; w m represents the work amount of the mth respiratory cycle; w m+1 represents the work amount of the m+1th respiratory cycle; || represents taking the absolute value.

[0042] In the calculation formula of the respiratory stability coefficient, by taking the reciprocal, r and |w m -w m+1| is a negative correlation mapping; the number of respiratory cycles of the respiratory mode is represented by the number of respiratory cycles corresponding to all secondary respiratory modes, the larger r is, the more respiratory cycles of the secondary respiratory mode, the more switching times of the respiratory mode, the stronger the respiratory instability, and the smaller Sta is; the change characteristics of the work amount are represented by the absolute value of the difference, |w m -w m+1 | is larger, the larger the change of the work amount of the adjacent respiratory cycle, the more unstable the respiratory work, the stronger the respiratory instability, and the smaller Sta is.

[0043] At the same time, the patient's respiratory work in multiple cycles actually reflects the level of the patient's respiratory compensation capacity, |w m -w m+1 | is smaller, the higher the level of the patient's respiratory compensation capacity, |w m -w m+1 | is larger, the lower the level of the patient's respiratory compensation capacity.

[0044] It should be noted that in other embodiments of the present application, the implementer can also perform negative correlation mapping through a negative correlation mapping function such as an exponential function with natural constant e as the base number exp(-x), x represents the independent variable; when the number of respiratory cycles of two or more respiratory modes is the most at the same time, the respiratory mode with the most time proportion is selected as the main respiratory mode.

[0045] Step S3: When the respiratory stability coefficient is greater than or equal to the preset stability threshold, a predicted respiratory pressure signal is obtained by using a preset prediction model; according to the difference characteristics of the respiratory pressure signal and the predicted respiratory pressure signal in the seasonal cycle, combined with the difference of the respiratory stability coefficient of the adjacent seasonal cycle, it is determined whether to re-predict; when it is determined not to re-predict, a predicted respiratory mode is obtained according to the final predicted respiratory pressure signal.

[0046] It is considered that the larger the respiratory stability coefficient is, the more stable the respiratory state is, and the change of the respiratory pressure signal has strong regularity, and the prediction model can accurately fit these rules and output reliable prediction results, so when the respiratory stability coefficient is greater than or equal to the preset stability threshold, the predicted respiratory pressure signal is obtained by using the preset prediction model.

[0047] As an example, the preset stability threshold is 0.8, the preset prediction model is an ARIMA model, and the existing respiratory pressure signal data of the respiratory cycle of [1, M] is input into the ARIMA model to output the predicted respiratory pressure signal.

[0048] It should be noted that when the respiratory stability coefficient is less than the preset stability threshold, it indicates that the current respiratory state of the patient is variable, and timely professional monitoring and management of medical personnel are required, so when the respiratory stability coefficient is less than 0.8, feedback is sent to relevant personnel, and prediction is not performed, so as to ensure safety and treatment effect.

[0049] It should be noted that the embodiment of the present application does not directly predict the work value of respiration, because the work value is obtained by integrating and calculating the pressure signal in each respiratory cycle, and the respiratory pressure signal contains more details about the respiratory behavior, and the respiratory pressure signal usually has obvious periodicity and regularity, and is more suitable for modeling and prediction by using a time series analysis method.

[0050] Among them, using an ARIMA model to predict a time series signal is prior art, and in other embodiments of the present application, the implementer can also train other methods such as a long short-term memory network to perform prediction, which will not be described in detail.

[0051] Considering that the respiratory state in the seasonal cycle is relatively stable, the predicted respiratory pressure signal should be similar to the actual respiratory pressure signal, so the difference between the respiratory pressure signal in the seasonal cycle and the predicted respiratory pressure signal reflects the prediction accuracy; the difference between the respiratory stability coefficients of adjacent seasonal cycles also reflects the change in the respiratory compensation capacity level of the patient, and also reflects the prediction accuracy, so according to the difference between the respiratory pressure signal in the seasonal cycle and the predicted respiratory pressure signal, combined with the difference between the respiratory stability coefficients of adjacent seasonal cycles, it is determined whether to re-predict, so as to ensure that the model adapts to the current actual state at any time, avoid accumulation of errors, and improve the accuracy and reliability of prediction.

[0052] Preferably, in an embodiment of the present application, the method for determining whether to re-predict comprises: Please refer to Figure 3 which shows a flowchart of a method for determining whether to re-predict provided by an embodiment of the present application, and specifically comprises: Step S301: Obtain the time domain interval of each respiratory cycle of the predicted respiratory pressure signal; take the number of respiratory cycles in the latest seasonal cycle as the number of comparison cycles; cut off the predicted respiratory pressure signal of the first comparison cycle number of respiratory cycles in the time domain, as a predicted comparison signal, and obtain the respiratory mode of each respiratory cycle in the predicted comparison signal.

[0053] Considering that the difference between the respiratory pressure signal in the seasonal period and the predicted respiratory pressure signal is small, the respiratory patterns of the respiratory cycles with the same time sequence are consistent, therefore, the predicted respiratory pressure signals corresponding to the respiratory cycles in the comparison period number of the time domain are extracted from the predicted respiratory pressure signal, and the respiratory patterns of each respiratory cycle are obtained, thus the predicted respiratory pressure signals of the respiratory cycles with the time sequence numbers [1, M] are obtained according to the respiratory pressure signals of the respiratory cycles with the time sequence numbers [1, M], which is convenient for subsequent comparison.

[0054] Meanwhile, the predicted respiratory pressure signal is the current predicted signal, so the predicted comparison signal can also be regarded as the respiratory pressure signal of the respiratory cycles with the current [M, 2M].

[0055] It should be noted that the time domain interval of the predicted respiratory pressure signal and the acquisition method of the respiratory pattern are the same as the time domain interval of the actually collected respiratory pressure signal and the acquisition method of the respiratory pattern, and will not be described here.

[0056] Step S302: Extracting a preset number of respiratory cycles with the same time sequence to form a comparison pair from the predicted comparison signal and the respiratory pressure signal in the seasonal period; obtaining the prediction credibility according to the difference between the identification numbers of the respiratory patterns of the two respiratory cycles in each comparison pair, and combining the difference between the respiratory stability coefficients of adjacent seasonal periods.

[0057] Considering that the calculation amount of the respiratory patterns of all respiratory cycles with the same time sequence is large, in the embodiment of the present application, part of them is selected for comparison, as an example, the preset number is 10, and 10 respiratory cycles are randomly extracted from the respiratory pressure signal of the latest seasonal period by random sampling, at the same time, 10 respiratory cycles are extracted from the predicted comparison signal according to the time sequence numbers of the extracted respiratory cycles, the respiratory cycles with the same time sequence are formed into a comparison pair, and 10 comparison pairs are obtained.

[0058] At the same time of dividing the respiratory patterns of the respiratory cycles, the identification numbers are given, so that the difference between the respiratory patterns can be directly compared; it is also considered that the respiratory function of the patient has certain fluctuation, the greater the difference between the respiratory stability coefficient of the latest seasonal period and the respiratory stability coefficient of the adjacent historical seasonal period, the greater the fluctuation of the respiratory stability of the patient, the greater the fluctuation of the respiratory compensation ability level, and the lower the prediction credibility, so as to obtain the prediction credibility.

[0059] As an example, the calculation formula of the prediction credibility includes: ; wherein, Tr represents the prediction confidence of the latest prediction; norm{} represents a linear normalization function; K represents a preset number, which is also the number of contrast binary tuples; k is the serial number of the contrast binary tuple; c k represents the serial number of the respiratory pattern of the respiratory cycle of the respiratory pressure signal in the kth contrast binary tuple; represents the serial number of the respiratory pattern of the respiratory cycle of the predicted contrast signal in the kth contrast binary tuple; || represents the absolute value; and ΔSta represents the absolute value of the difference between the respiratory stability coefficient of the latest seasonal cycle and the respiratory stability coefficient of the adjacent historical seasonal cycle.

[0060] In the calculation formula of the prediction confidence, the difference between the serial numbers of the respiratory patterns of the two respiratory cycles in the contrast binary tuple is represented by the absolute value of the difference, and the differences corresponding to all the contrast binary tuples are represented by summation, reflecting the difference characteristics of the respiratory pressure signal and the predicted respiratory pressure signal, The greater the difference characteristics, the greater the difference between the respiratory pressure signal and the predicted respiratory pressure signal, the less ideal the prediction effect, and the smaller the prediction confidence. The greater ΔSta, the greater the respiratory fluctuation of the patient, and the smaller the prediction confidence.

[0061] It should be noted that the random sampling method is already a prior art, and in other embodiments of the present application, the implementer can also use equidistant sampling or other sampling methods; and all respiratory cycles with the same time sequence number can also be compared, which will not be described in detail.

[0062] Step S303: When the prediction confidence is less than the preset confidence threshold, it is determined to re-predict and re-determine after re-prediction; when the prediction confidence is greater than or equal to the preset confidence threshold, it is determined not to re-predict.

[0063] As an example, the preset confidence threshold is 0.3, and the repeated iterative prediction is performed until the prediction confidence corresponding to the latest prediction is greater than 0.3, and it is determined not to re-predict, at which time the final predicted respiratory pressure signal is obtained, providing a basis for the respiratory mode of the respiratory monitoring mask in advance, and reducing the hysteresis of the respiratory monitoring mask in determining the respiratory mode.

[0064] It should be noted that when the prediction is continuously performed, the current prediction interval is the time domain length of the prediction contrast signal of the final predicted respiratory pressure signal; for example, when the first prediction is performed, the time domain length of the corresponding prediction contrast signal is 5 minutes, and the second prediction is performed after 5 minutes; when the second prediction is performed, the time domain length of the corresponding prediction contrast signal is 4 minutes, and the third prediction is performed after 4 minutes. When the prediction is interrupted, i.e., the respiratory stability coefficient is less than the preset stability threshold, the prediction interval is set to 10 minutes, and the latest seasonal cycle is re-acquired after 10 minutes, and the respiratory stability coefficient is acquired to determine whether to perform prediction.

[0065] Finally, when it is determined not to re-predict, the respiratory monitoring mask determines the respiratory pattern according to the final predicted respiratory pressure signal, obtains a predicted respiratory pattern, and reduces the hysteresis of the respiratory monitoring mask in determining the respiratory pattern.

[0066] Preferably, in one embodiment of the present application, in the prediction comparison signal of the final predicted respiratory pressure signal, the proportion of the number of respiratory cycles of each respiratory pattern in the total number of all respiratory cycles is taken as the proportion of each respiratory pattern; the respiratory pattern with the largest proportion is selected as the predicted respiratory pattern.

[0067] In another embodiment of the present application, after obtaining the predicted respiratory pattern, the predicted comparison signal is also used to feedback adjust the ventilator, specifically comprising: In the prediction comparison signal of the final predicted respiratory pressure signal, the proportion of the number of respiratory cycles of each respiratory pattern in the total number of all respiratory cycles is taken as the proportion of each respiratory pattern; When the main respiratory pattern is rapid shallow breathing or normal resting breathing, and the proportion and value of deep breathing and forced breathing are less than a first preset proportion threshold, the ventilator is set to S mode; When the main respiratory pattern is rapid shallow breathing or normal resting breathing, and the proportion and value of deep breathing and forced breathing are greater than or equal to the first preset proportion threshold, the ventilator is set to S / T mode; When the main respiratory pattern is deep breathing, and the proportion of forced breathing is less than a second preset threshold, the ventilator is set to S / T mode; When the main respiratory pattern is deep breathing, and the proportion of forced breathing is greater than or equal to the second preset threshold, the ventilator is set to CACP mode; When the main respiratory pattern is forced breathing, and the proportion of forced breathing is less than a third preset proportion threshold, the ventilator is set to CACP mode; When the main respiratory pattern is forced breathing, and the proportion of forced breathing is greater than or equal to the third preset proportion threshold, the ventilator is set to T mode.

[0068] In other embodiments of the present application, the implementer can also discard the third preset proportion threshold, and when the main respiratory pattern is forced breathing, the ventilator is set to T mode.

[0069] As an example, the first preset proportion threshold is 15%, the second preset threshold is 30%, and the third preset proportion threshold is 50%.

[0070] It should be noted that the S mode, S / T mode, CACP mode and T mode of the ventilator are prior art, and the implementer can adjust the first preset proportion threshold, the second preset proportion threshold and the third preset proportion threshold.

[0071] It should be noted that when the respirator with the working mode that can be adaptively adjusted according to the respiratory pressure signal is adopted, the final predicted respiratory pressure signal can also be inputted, so that the respirator can obtain the respiratory pressure signal in advance and make adjustment, reduce the hysteresis of real-time processing, improve the response speed of respiratory support, and improve the comfort.

[0072] An embodiment of the present application also provides a big data intelligent health respiratory dynamic monitoring system, which specifically comprises: A data acquisition module is configured to acquire the respiratory pressure signal and the respiratory pressure value of a current patient, and acquire the time domain interval of each respiratory cycle of the respiratory pressure signal according to the change characteristic of the respiratory pressure signal; A respiratory analysis module is configured to acquire the work amount corresponding to each respiratory cycle based on all the pressure values in each time domain interval, divide the respiratory mode of each respiratory cycle based on the work amount, extract the latest seasonal cycle of the respiratory pressure signal based on the STL algorithm, and acquire the respiratory stability coefficient according to the change characteristic of the work amount of the time-sequentially adjacent respiratory cycles in the seasonal cycle and the number distribution characteristic of the respiratory cycles of the respiratory mode. A respiratory prediction module is configured to acquire a predicted respiratory pressure signal by using a preset prediction model when the respiratory stability coefficient is greater than or equal to a preset stability threshold, determine whether to re-predict according to the difference characteristic of the respiratory pressure signal and the predicted respiratory pressure signal in the seasonal cycle and the difference of the respiratory stability coefficients of adjacent seasonal cycles, and acquire a predicted respiratory mode according to the final predicted respiratory pressure signal when it is determined not to re-predict.

[0073] An embodiment of the present application also provides a big data intelligent health respiratory dynamic monitoring terminal, which comprises a memory, a processor and a computer program, wherein the memory is used for storing the corresponding computer program, the processor is used for running the corresponding computer program, and the computer program can realize the big data intelligent health respiratory dynamic monitoring method described in steps S1-S3 when running in the processor.

[0074] To sum up, in view of the technical problem of the existing breathing monitoring mask that is lagging in judging the breathing mode by monitoring the breathing work data, the present application acquires the time domain interval of each breathing cycle according to the change characteristics of the breathing pressure signal; further acquires the work amount corresponding to each breathing cycle, divides the breathing mode; further extracts the latest seasonal cycle of the breathing pressure signal, acquires the breathing stability coefficient according to the change characteristics of the work amount of the time series adjacent breathing cycles in the seasonal cycle, and combines the number distribution characteristics of the breathing cycle of the breathing mode; further, when the breathing stability coefficient is greater than or equal to the preset stability threshold, the predicted breathing pressure signal is acquired by using the preset prediction model; according to the difference characteristics of the breathing pressure signal and the predicted breathing pressure signal in the seasonal cycle, it is determined whether to re-predict; finally, when it is determined not to re-predict, the predicted breathing mode is acquired according to the final predicted breathing pressure signal.

[0075] It should be noted that the above-mentioned embodiment sequence of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or may be advantageous.

[0076] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment mainly describes the difference from other embodiments.

Claims

1. A big data-driven intelligent health respiratory dynamic monitoring method, characterized in that, The method comprises: obtaining a respiratory pressure signal and a pressure value of respiration of a current patient; obtaining a time domain interval of each respiratory cycle of the respiratory pressure signal according to a variation feature of the respiratory pressure signal; obtaining a work amount corresponding to each respiratory cycle based on all the pressure values in each time domain interval; dividing a respiratory mode for each respiratory cycle based on the work amount; extracting a latest seasonal cycle of the respiratory pressure signal based on an STL algorithm; obtaining a respiratory stability coefficient according to a variation feature of the work amount of the respiratory cycle adjacent in time sequence in the seasonal cycle and a number distribution feature of the respiratory cycle of the respiratory mode; when the respiratory stability coefficient is greater than or equal to a preset stability threshold, obtaining a predicted respiratory pressure signal by using a preset prediction model; determining whether to re-predict according to a difference feature of the respiratory pressure signal and the predicted respiratory pressure signal in the seasonal cycle and a difference of the respiratory stability coefficient of adjacent seasonal cycles; when it is determined not to re-predict, obtaining a predicted respiratory mode according to the final predicted respiratory pressure signal. 2.The respiratory dynamic monitoring method of big data smart health according to claim 1, characterized in that, The method for obtaining the time domain interval comprises: extracting a trend curve of the respiratory pressure signal; dividing a time domain between two adjacent minimum points in the trend curve into a time domain interval of one respiratory cycle. 3.The respiratory dynamic monitoring method of big data smart health according to claim 1, characterized in that, The method for obtaining the respiratory mode comprises: when the work amount is in a first preset work interval, marking the corresponding respiratory cycle as rapid shallow breathing and setting an identification serial number as 01; when the work amount is in a second preset work interval, marking the corresponding respiratory cycle as normal resting breathing and setting an identification serial number as 02; when the work amount is in a third preset work interval, marking the corresponding respiratory cycle as deep breathing and setting an identification serial number as 03; when the work amount is in a fourth preset work interval, marking the corresponding respiratory cycle as forced breathing and setting an identification serial number as 04. 4.The respiratory dynamic monitoring method of big data smart health according to claim 3, characterized in that, The method for obtaining the respiratory stability coefficient comprises: in the latest seasonal cycle, when the number of the respiratory cycle corresponding to a certain respiratory mode is the largest, marking this respiratory mode as a main respiratory mode and marking other respiratory modes as secondary respiratory modes; obtaining a respiratory stability coefficient of the latest seasonal cycle according to the number of the respiratory cycle corresponding to all the secondary respiratory modes in the seasonal cycle and an absolute value of a difference of the work amount of the respiratory cycle adjacent in time sequence; the number of the respiratory cycle corresponding to all the secondary respiratory modes and the absolute value of the difference of the work amount of the respiratory cycle adjacent in time sequence are negatively correlated with the respiratory stability coefficient.

5. The big data smart health respiratory dynamic monitoring method according to claim 4, characterized in that, The method for determining whether to re-predict comprises: obtaining a time domain interval of each respiratory cycle of the predicted respiratory pressure signal; taking the number of respiratory cycles in the latest seasonal cycle as a comparison cycle number; taking the predicted respiratory pressure signal of the comparison cycle number of respiratory cycles in front in time domain as a predicted comparison signal and obtaining a respiratory mode of each respiratory cycle in the predicted comparison signal; extracting a preset number of comparison pairs from the prediction comparison signal and the respiratory pressure signal in the seasonal period, the comparison pairs being the respiratory cycles with the same time sequence number; obtaining a prediction credibility according to the difference between the identification numbers of the respiratory modes of the two respiratory cycles in each of the comparison pairs, in combination with the difference between the respiratory stability coefficients of adjacent seasonal periods, the difference between the identification numbers being negatively correlated with the prediction credibility; when the prediction credibility is less than a preset credibility threshold, determining to re-predict and re-determining after re-prediction; when the prediction credibility is greater than or equal to the preset credibility threshold, determining not to re-predict. 6.The respiratory dynamic monitoring method of big data smart health according to claim 1, characterized in that, the method for obtaining the predicted respiratory mode comprises: in the prediction comparison signal of the final predicted respiratory pressure signal, taking the proportion of the number of respiratory cycles of each respiratory mode in the total number of all respiratory cycles as the proportion of each respiratory mode; selecting the respiratory mode with the largest proportion as the predicted respiratory mode. 7.The respiratory dynamic monitoring method of big data smart health according to claim 1, wherein, the method for obtaining the work amount comprises: integrating the pressure values in each of the time domain intervals, and taking the integral value as the work amount corresponding to the respiratory cycle. 8.The respiratory dynamic monitoring method of big data smart health according to claim 1, wherein, the preset prediction model is an ARIMA model.

9. A big data intelligent health respiratory dynamic monitoring system, the system comprising: a data acquisition module: acquiring a respiratory pressure signal and a pressure value of a current patient; and obtaining a time domain interval of each respiratory cycle of the respiratory pressure signal according to the change characteristics of the respiratory pressure signal; a respiratory analysis module: obtaining a work amount corresponding to each of the respiratory cycles based on all the pressure values in each of the time domain intervals; dividing a respiratory mode for each of the respiratory cycles based on the work amount; extracting a latest seasonal period of the respiratory pressure signal based on an STL algorithm; and obtaining a respiratory stability coefficient based on the change characteristics of the work amount of the time-sequentially adjacent respiratory cycles in the seasonal period, in combination with the number distribution characteristics of the respiratory cycles of the respiratory mode; a respiratory prediction module: when the respiratory stability coefficient is greater than or equal to a preset stability threshold, obtaining a predicted respiratory pressure signal by using a preset prediction model; and determining whether to re-predict based on the difference characteristics of the respiratory pressure signal and the predicted respiratory pressure signal in the seasonal period, in combination with the difference between the respiratory stability coefficients of adjacent seasonal periods; when it is determined not to re-predict, obtaining a predicted respiratory mode according to the final predicted respiratory pressure signal. 10.A terminal for monitoring respiratory dynamics in a big data smart health, the terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein, the processor executes the computer program to implement the steps of the big data intelligent health respiratory dynamic monitoring method according to any one of claims 1-8.

Images