Oxygen supply system and method of integrating positive end-expiratory pressure monitoring

Abstract

This invention relates to the field of respiratory oxygen generation equipment control technology, specifically to an oxygen concentrator supply system and method integrating positive end-expiratory pressure (PEEP) monitoring. The method acquires airway pressure signals in real time and finely divides the inspiratory-expiratory cycle. Overall stability is obtained through consistency analysis of historical inspiratory and expiratory pressure signals. Furthermore, the matching degree between the PEEP pressure signal and a preset standard range is considered, and valid historical cycles are identified and selected based on signal consistency and frequency consistency. The real-time monitored inspiratory, expiratory, and PEEP pressure signals are compared with standard respiratory cycle data to quantify the deviation between the current respiratory state and the ideal model, thus providing a digital basis for adjusting the oxygen concentrator. This invention, by constructing an individualized standard model and implementing real-time state comparison and adjustment, achieves effective dynamic adjustment of the oxygen concentrator from passive constant oxygen supply to active adaptive respiratory support.

Description

Technical Field

[0001] This invention relates to the field of respiratory oxygen generation equipment control technology, specifically to an oxygen generator supply system and method integrating positive end-expiratory pressure monitoring. Background Technology

[0002] In oxygen therapy settings, patients with insufficient oxygenation but not yet meeting the criteria for mechanical ventilation, such as those with chronic obstructive pulmonary disease, require the introduction of positive end-expiratory pressure (PEEP) to maintain a certain positive pressure level at the end of expiration, prevent alveolar collapse, increase oxygen retention time in the lungs and gas exchange efficiency, thereby improving blood oxygen saturation. However, traditional oxygen generators only provide a fixed or regulated oxygen flow rate and cannot control airway pressure. They primarily rely on preset oxygen flow rates, addressing insufficient oxygenation only from the perspective of oxygen supply, without comprehensively analyzing changes in airway pressure and differences between inspiratory and expiratory phases during the patient's breathing process. This lack of effective control over PEEP leads to indiscriminate oxygen supply regulation. Summary of the Invention

[0003] To address the technical problem that existing oxygen concentrators only provide a fixed oxygen supply and cannot dynamically adjust based on the patient's real-time breathing process, the present invention aims to provide an oxygen concentrator supply system and method integrating positive end-expiratory pressure monitoring. The specific technical solution adopted is as follows: This invention proposes an oxygen supply method for an oxygen concentrator integrating positive end-expiratory pressure monitoring, the method comprising: Acquire real-time pressure signals of the patient's airway during the operation of the oxygen concentrator; divide multiple real-time inspiratory-expiratory cycles according to the changing trend of the pressure signals, and extract the real-time inspiratory pressure signal, real-time expiratory pressure signal, and real-time PEEP pressure signal for each real-time inspiratory-expiratory cycle; In the patient's historical database, for any target historical inspiratory-expiratory cycle, the pressure signals within adjacent historical inspiratory-expiratory cycles are used to construct a short-term respiratory pressure signal set. Within this short-term respiratory pressure signal set, overall stability is obtained based on the consistency of inspiratory and expiratory pressure signals between any two historical inspiratory-expiratory cycles. The degree of matching between the PEEP pressure signal range in each historical inspiratory-expiratory cycle within the short-term respiratory pressure signal set and a preset standard range is obtained. Based on overall stability, the degree of matching, and the frequency consistency of pressure signals between historical inspiratory-expiratory cycles, the state stability of the short-term respiratory pressure signal set is obtained. State stability is used to determine whether the target historical inspiratory-expiratory cycle is a valid historical cycle. Valid historical cycles are extracted from the historical database to form a standard respiratory cycle. The real-time inspiratory pressure signal, real-time expiratory pressure signal, and real-time PEEP pressure signal of each real-time inspiratory-expiratory cycle are compared with the corresponding standard pressure data in the standard respiratory cycle to obtain the respiratory status of each real-time inspiratory-expiratory cycle; the valve opening of the oxygen concentrator is adjusted according to the respiratory status.

[0004] Furthermore, the process of dividing the airflow into multiple real-time inhalation-exhalation cycles based on the changing trend of the pressure signal includes: The local maxima of the real-time pressure signal are obtained, and the range between adjacent local maxima is one real-time inhalation-exhalation cycle. In one real-time inhalation-exhalation cycle, the pressure change trend between adjacent moments is obtained one by one. If a moment does not show a pressure decrease trend and is a local minimum, it is determined as the inhalation endpoint. The inhalation endpoint divides the real-time inhalation-exhalation cycle into an inhalation phase and an exhalation phase. In the exhalation phase, the pressure change rate sequence is extracted. The moment when the absolute value of the pressure change rate sequence first meets the condition of being less than a preset absolute value of the change rate and is maintained for more than a preset duration is taken as the start moment of the PEEP phase. The range between the start moment of the PEEP phase and the end moment of the exhalation phase is taken as the PEEP phase.

[0005] Furthermore, the step of constructing a short-term respiratory pressure signal set from pressure signals within adjacent historical inspiratory-expiratory cycles includes: Starting from the target historical inspiratory-expiratory cycle, a preset number of historical inspiratory-expiratory cycles are searched sequentially, and the corresponding pressure signals are statistically analyzed to form a short-term respiratory pressure signal set.

[0006] Furthermore, the method for obtaining the overall stability includes: In the short-term respiratory pressure signal set, the proportion of expiratory time and the proportion of inspiratory time in each historical inspiratory-expiratory cycle are obtained. The average proportion of inspiratory time in all historical inspiratory-expiratory cycles is used as the first weight, and the average proportion of expiratory time is used as the second weight. For any two historical inspiratory-expiratory cycles, the first similarity between inspiratory pressure signals and the second similarity between expiratory pressure signals are obtained using the dynamic time warping algorithm. The average first weight of the two historical inspiratory-expiratory cycles is used as the weight of the first similarity, and the average second weight of the two historical inspiratory-expiratory cycles is used as the weight of the second similarity. The two are then weighted and summed to obtain the signal consistency between the two historical inspiratory-expiratory cycles. The average signal consistency of the statistical short-term respiratory pressure signal set is used as the overall stationarity.

[0007] Furthermore, the method for obtaining the matching degree includes: Historical inspiratory-expiratory cycles with PEEP pressure signals within a preset standard range are defined as effective historical inspiratory-expiratory cycles. The proportion of effective inspiratory-expiratory cycles in the short-term respiratory pressure signal set is statistically analyzed. The average PEEP pressure signal value of the effective historical inspiratory-expiratory cycles is defined as the overall PEEP pressure value. The stability of the overall PEEP pressure value of all effective historical inspiratory-expiratory cycles is obtained, and the matching degree of the short-term respiratory pressure signal set is obtained based on the stability and the proportion of the number of cycles.

[0008] Furthermore, the method for obtaining the frequency consistency of the pressure signal includes: For any historical inspiratory-expiratory cycle, the reciprocal of the time range is taken as the respiratory rate; In a short-term respiratory pressure signal set, the frequency consistency of the pressure signal is obtained based on the frequency difference between the respiratory rate of each historical inspiratory-expiratory cycle and the average respiratory rate.

[0009] Furthermore, the method for obtaining the standard respiratory cycle includes: If the stability of the target historical inhalation-exhalation cycle is greater than a preset stability threshold, it is marked as a valid historical cycle. All valid historical cycles in the historical database are statistically analyzed. Valid historical cycles that are consecutive in time are merged to obtain several consecutive time intervals. The valid historical cycle in the longest consecutive time interval is selected as the standard respiratory cycle.

[0010] Furthermore, obtaining the respiratory state for each real-time inspiratory-expiratory cycle includes: Extract the standard inspiratory pressure signal, standard expiratory pressure signal, and standard PEEP pressure signal of a standard respiratory cycle; combine the standard PEEP pressure signal range with the preset standard range to obtain an adaptive standard range; For any real-time inhalation-exhalation cycle, the real-time PEEP pressure signal range is compared with the adaptive standard range to determine the PEEP state of the real-time inhalation-exhalation cycle. The PEEP state is either abnormal or normal. Abnormal PEEP includes insufficient PEEP and excessive PEEP. At each PEEP state, the first deviation between the real-time inspiratory pressure signal and the standard inspiratory pressure signal during the real-time inspiratory-expiratory cycle, and the second deviation between the real-time expiratory pressure signal and the standard expiratory pressure signal are calculated. The inspiratory state is determined based on the first deviation, which includes normal inspiratory, insufficient inspiratory, and excessive inspiratory. The expiratory state is determined based on the second deviation, which includes normal expiratory, insufficient expiratory, and excessive expiratory.

[0011] Furthermore, adjusting the valve opening of the oxygen concentrator according to the breathing state includes: If the proportion of real-time inspiratory-expiratory cycles with normal PEEP is greater than a preset threshold, the oxygen concentrator valve parameters are maintained; otherwise, for real-time inspiratory-expiratory cycles with abnormal PEEP, the inspiratory and expiratory states are counted. Under both insufficient and excessive PEEP conditions, the number of real-time inspiratory-expiratory cycles for each inspiratory and expiratory state is counted, and the state with the most cycles is selected as the patient's current real-time respiratory state. The valves of the oxygen concentrator's inspiratory and expiratory channels are controlled according to the real-time respiratory state.

[0012] The present invention also proposes an oxygen concentrator oxygen supply system with integrated positive end-expiratory pressure monitoring, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of any one of the oxygen concentrator oxygen supply methods with integrated positive end-expiratory pressure monitoring.

[0013] The present invention has the following beneficial effects: This invention achieves precise perception of respiratory dynamics characteristics by acquiring airway pressure signals in real time and finely dividing the inspiratory-expiratory cycle. Overall stability is obtained through consistency analysis of historical inspiratory and expiratory pressure signals, and the stability of short-term respiratory pressure signal sets is quantitatively evaluated from multiple dimensions by combining the matching degree of PEEP pressure signals with preset standard ranges. This method of judging and selecting effective historical cycles based on signal and frequency consistency can construct standard respiratory cycles with biologically representative characteristics for different individuals, ensuring the individualization and accuracy of the comparison benchmark. By comparing real-time monitored inspiratory, expiratory, and PEEP pressure signals with standard respiratory cycle data, the system can accurately quantify the deviation between the current respiratory state and the ideal model, thus providing a digital basis for adjusting the oxygen concentrator. Finally, the valve opening of the oxygen concentrator is dynamically adjusted based on respiratory status feedback, achieving active intervention in airway pressure, ensuring that the positive end-expiratory pressure is stably maintained within the target range, and improving pulmonary gas exchange efficiency and oxygen utilization. This invention, by constructing an individualized standard model and implementing real-time state comparison and adjustment, achieves effective dynamic adjustment of the oxygen concentrator from passive constant oxygen supply to active adaptive respiratory support. Attached Figure Description

[0014] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1This is a flowchart of an oxygen concentrator oxygen supply method with integrated positive end-expiratory pressure monitoring, provided as an embodiment of the present invention. Detailed Implementation

[0016] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an oxygen concentrator oxygen supply system and method integrating positive end-expiratory pressure monitoring according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0017] 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 invention pertains.

[0018] The following description, in conjunction with the accompanying drawings, details the specific scheme of an oxygen concentrator oxygen supply system and method integrating positive end-expiratory pressure monitoring provided by the present invention.

[0019] Please see Figure 1 The diagram illustrates a flowchart of an oxygen concentrator oxygen supply method integrating positive end-expiratory pressure monitoring, according to an embodiment of the present invention. The method includes: Step S1: Acquire the real-time pressure signal of the patient's airway during the operation of the oxygen concentrator; divide the real-time inhalation-exhalation cycle according to the trend of pressure signal changes, and extract the real-time inhalation pressure signal, real-time expiratory pressure signal and real-time PEEP pressure signal for each real-time inhalation-exhalation cycle.

[0020] An oxygen concentrator integrating positive end-expiratory pressure (PEEP) control, a shared inspiratory and expiratory tubing, and a closed-face mask (or nasal mask) is included. The concentrator or the end of the tubing has a controlled expiratory valve. During respiration, the patient needs to inhale sufficient oxygen through the oxygen supply channel during inspiration and exhale through the expiratory channel during expiration to maintain a stable respiratory cycle. Positive end-expiratory pressure (PEEP) refers to the positive pressure maintained in the airway at the end of expiration, typically controlled within the range of approximately 3–8 cmH2O. Its main function is to increase functional residual capacity in the lungs, prolong the residence time of oxygen in the lungs, thereby improving gas exchange efficiency and blood oxygen saturation. When PEEP is too low, it indicates that gas is expelled too quickly during exhalation, resulting in insufficient gas retention in the lungs and an inability to form effective residual pressure. In this case, it is necessary to reduce the opening of the expiratory valve to increase expiratory resistance and appropriately increase the oxygen supply flow rate to promote airway pressure establishment and maintain positive end-expiratory pressure. When PEEP is too high, it indicates that expiratory resistance is too high or gas expulsion is restricted. In this case, it is necessary to increase the opening of the expiratory valve to reduce resistance and promote gas expulsion, while reducing the oxygen supply flow rate to prevent continuous gas accumulation, thereby bringing the positive end-expiratory pressure back to a reasonable range.

[0021] Therefore, the pressure value exhibited during the PEEP phase is an important indicator for determining the patient's condition. The PEEP phase is the final stage of an inspiratory-expiratory cycle. Therefore, this embodiment of the invention first acquires the real-time pressure signal of the patient's airway during the operation of the oxygen concentrator. This real-time pressure signal is a real-time signal with a certain time series, within which multiple respiratory cycles exist, because the change in airway pressure during a respiratory cycle is affected by both respiratory muscle activity and airway resistance. During the inspiratory phase, air is drawn into the lungs, and the airway pressure decreases due to being lower than the external atmospheric pressure. During the expiratory phase, air is expelled from the lungs, and the airway pressure increases due to being higher than the external atmospheric pressure. At the end of expiration, because the residual gas in the lungs maintains a certain pressure, the airway pressure tends to stabilize, forming positive end-expiratory pressure (PEEP). During the inspiratory phase, the airway pressure generally shows a continuous downward trend; therefore, the interval from the local maximum pressure to the local minimum pressure usually corresponds to the inspiratory phase. The starting point of inspiration can be determined as the local maximum, and the ending point of inspiration can be determined as the local minimum. Because the pressure value continuously decreases at adjacent time points during the descent, even if the pressure briefly rebounds during the patient's inhalation due to airflow interference, this phase is still considered as continuous inspiration as long as the pressure has not reached a new extreme point. Therefore, multiple real-time inspiratory-expiratory cycles can be divided based on the trend of pressure signal changes. Furthermore, each phase can be identified within each real-time inspiratory-expiratory cycle, and the real-time inspiratory pressure signal, real-time expiratory pressure signal, and real-time PEEP pressure signal can be extracted for each cycle.

[0022] It should be noted that, in this embodiment of the invention, to facilitate signal processing and period division, the pressure signal acquisition can be set with a monitoring duration; in this embodiment, the signal can be acquired every minute. After obtaining the pressure signal, it can be filtered to remove environmental noise and airflow disturbances, resulting in a smoother pressure waveform. Specific signal preprocessing methods are well-known to those skilled in the art and will not be elaborated upon here.

[0023] Preferably, in this embodiment of the invention, since the extreme points in the pressure signal waveform represent the moments of change in pressure trends, the division of each stage in the real-time inhalation-exhalation cycle can be directly based on the distribution of extreme points, specifically including: First, obtain the local maxima of the real-time pressure signal. As can be seen from the above explanation of the inspiratory-expiratory cycle, the starting position of the cycle should be the time corresponding to the maximum pressure in the airway. Therefore, the range between adjacent local maxima is defined as one real-time inspiratory-expiratory cycle.

[0024] Furthermore, within each predefined real-time inspiratory-expiratory cycle, starting from the initial point, the pressure change trend between each moment and adjacent moments is obtained. If a moment shows a normal pressure decrease trend, it indicates that the inspiratory phase is still underway. Conversely, if a moment does not show a pressure decrease trend and is a local minimum of the signal, it indicates that this moment is the inspiratory endpoint. This inspiratory endpoint divides each real-time inspiratory-expiratory cycle into an inspiratory phase and an expiratory phase, with the PEEP phase occurring within the expiratory phase. Because the PEEP phase is a stable airway pressure phase, exhibiting a smaller pressure change rate compared to previous phases and maintaining this rate for a certain duration, this embodiment further extracts a pressure change rate sequence during the expiratory phase. The moment in the pressure change rate sequence that first satisfies an absolute value less than a preset absolute value and remains so for a preset duration is taken as the starting moment of the PEEP phase. The range between the starting moment of the PEEP phase and the ending moment of the expiratory phase is defined as the PEEP phase.

[0025] In one specific implementation of this invention, the method for obtaining the pressure change trend is as follows: The pressure value at one moment is subtracted from the pressure value at the previous moment from the pressure signal to obtain the pressure change rate at each moment. If the pressure change rate is negative, it indicates a decreasing pressure trend; if it is positive, it indicates an increasing pressure trend; and if it is 0, it indicates a stable pressure trend. The absolute value of the change rate can be set to 0.05 cmH2O / s, and the preset duration can be set to 0.2s. Specific threshold settings can be obtained through statistical analysis of medical big data and are not limited here.

[0026] Step S2: In the patient's historical database, for any target historical inspiratory-expiratory cycle, construct a short-term respiratory pressure signal set by combining the pressure signals within adjacent historical inspiratory-expiratory cycles; within the short-term respiratory pressure signal set, obtain overall stability based on the consistency of inspiratory and expiratory pressure signals between any two historical inspiratory-expiratory cycles; obtain the degree of matching between the PEEP pressure signal range and the preset standard range in each historical inspiratory-expiratory cycle within the short-term respiratory pressure signal set; obtain the state stability of the short-term respiratory pressure signal set based on overall stability, matching degree, and the frequency consistency of pressure signals between historical inspiratory-expiratory cycles; determine whether the target historical inspiratory-expiratory cycle is a valid historical cycle based on the state stability; extract valid historical cycles from the historical database to construct a standard respiratory cycle.

[0027] This invention aims to tailor the oxygen concentrator to the unique respiratory state of each patient. Traditional devices with fixed parameters or fixed respiratory state reference standards often result in poor adjustment outcomes due to individual patient factors. Therefore, this invention further utilizes the patient's historical database, which stores airway pressure data from multiple recent inspiratory-expiratory cycles. Under stable respiratory conditions, the respiratory dynamics of the same patient are consistent; therefore, the changes in airway pressure curves are largely consistent across multiple respiratory cycles, exhibiting high similarity after time alignment. Simultaneously, the respiratory rate and inspiratory-expiratory ratio also remain relatively stable, collectively indicating a stable and regular respiratory state. Therefore, step S2 of this invention analyzes the pressure data from the historical database to identify a series of recent stable respiratory state data as standard data for determining the real-time status.

[0028] To determine whether a historical cycle represents a stable respiratory state, it is necessary to perform a combined analysis of the cycle and its adjacent cycles. Therefore, this embodiment of the invention first takes any historical inspiratory-expiratory cycle from the historical database as a target historical inspiratory-expiratory cycle and analyzes it separately. The pressure signals within the target historical inspiratory-expiratory cycle and its adjacent historical inspiratory-expiratory cycles are then used to construct a short-term respiratory pressure signal set. That is, one target historical inspiratory-expiratory cycle corresponds to one short-term respiratory pressure signal set. If this short-term respiratory pressure signal set exhibits a stable pressure distribution, it indicates that the target historical inspiratory-expiratory cycle is in a stable respiratory state.

[0029] In short-term respiratory pressure signal concentrations, overall stability is first determined by the consistency of inspiratory and expiratory pressure signals between any two historical inspiratory-expiratory cycles. That is, within this signal concentration, the more similar the pressure signals during the inspiratory and expiratory phases of each cycle, the more stable the patient's respiratory state is within that time period, and the greater the corresponding overall stability.

[0030] Furthermore, considering a stable and standard breathing state, which should remain even more stable during the PEEP phase and with the signal range within the normal physiological range, this embodiment of the invention further compares the PEEP pressure signal range in each historical inspiratory-expiratory cycle with a preset standard range to obtain the matching degree of the entire short-term respiratory pressure signal set. That is, the greater the matching degree, the more consistent the pressure state exhibited during the PEEP phase is with the normal physiological state, and the more likely the signal set is to be in a stable breathing state, making the target historical inspiratory-expiratory cycle more reliable.

[0031] Further analysis of the frequency consistency of pressure signals between historical inspiratory-expiratory cycles reveals that frequency consistency, similar to overall stationarity, characterizes the frequency uniformity of different pressure signals within a short-term respiratory pressure signal set. Therefore, combining the overall stationarity and matching degree mentioned above yields the state stationarity of the short-term respiratory pressure signal set. Greater state stationarity indicates a higher likelihood of the target historical inspiratory-expiratory cycle being in a stable respiratory state, thus making the pressure signal of that cycle more reliable. Therefore, based on state stationarity, the validity of a target historical inspiratory-expiratory cycle is determined. All valid historical cycles are extracted from the historical database and then selected or combined to form a standard respiratory cycle for assessing the patient's real-time respiratory status.

[0032] In one specific implementation of this invention, a preset standard range is set between 3 and 8 cmH2O, which represents the default healthy physiological range and will not be elaborated further. The historical database can be accessed by doctors selecting the period of most recent physiological stability for each patient.

[0033] Preferably, in this embodiment of the invention, the pressure signals within adjacent historical inspiratory-expiratory cycles are used to form a short-term respiratory pressure signal set, including: Starting with the target historical inspiratory-expiratory cycle, a preset number of historical inspiratory-expiratory cycles are searched sequentially backward, and the corresponding pressure signals are statistically analyzed to form a short-term respiratory pressure signal set. It should be noted that if there are no preset number of other cycles after a certain historical inspiratory-expiratory cycle in the historical database, its validity is not evaluated. In one specific implementation of this invention, the preset number is set to four, meaning the final short-term respiratory pressure signal set includes data from five cycles. This sequential backward construction of the signal set allows the target historical inspiratory-expiratory cycle to be used as the "starting point" for analysis, making the final state stability more consistent with the characteristics of the target historical inspiratory-expiratory cycle.

[0034] Preferably, in some implementations of the present invention, considering that there are two phases, inhalation and exhalation, within a respiratory cycle, the pressure signals of these two phases should be analyzed separately before being combined to determine the overall stability. Therefore, the specific method for obtaining the overall stability may include: In short-term respiratory pressure signal aggregation, the proportion of expiratory time and the proportion of inspiratory time within each historical inspiratory-expiratory cycle are obtained. The average proportion of inspiratory time across all historical inspiratory-expiratory cycles is used as the first weight, and the average proportion of expiratory time as the second weight. This weighting method reveals which phase of a historical inspiratory-expiratory cycle is dominant, thus representing the confidence level of the consistency analysis results for that phase based on its dominance. Subsequent weighted processing yields effective consistency analysis results.

[0035] For any two historical inspiratory-expiratory cycles, a first similarity score between inspiratory pressure signals and a second similarity score between expiratory pressure signals are obtained using a dynamic time warping algorithm. Both similarity scores are results of separate analyses of the inspiratory and expiratory phases. To fuse these two separate analysis results, the average first weight of the two historical inspiratory-expiratory cycles is used as the weight for the first similarity score, and the average second weight of the two historical inspiratory-expiratory cycles is used as the weight for the second similarity score. These are then weighted and summed to obtain the signal consistency between the two historical inspiratory-expiratory cycles.

[0036] Obtain the signal consistency between any two historical inspiratory-expiratory cycles in the short-term respiratory pressure signal set, and calculate the average signal consistency in the short-term respiratory pressure signal set as the overall stationarity.

[0037] It should be noted that the Dynamic Time Warping (VTW) algorithm eliminates signal velocity differences through nonlinear alignment, accurately measuring shape similarity. The algorithm typically outputs a VTW distance, which requires negative correlation mapping and normalization to obtain the similarity score. In one specific implementation of this invention, a negative exponential function exp(-x) with the natural constant as the base is used for mapping. The distance output by the algorithm is substituted into x, and the result of the function mapping is the final similarity score.

[0038] It should be noted that the exhalation time percentage is the proportion of the total cycle time in the exhalation phase, and the inhalation time percentage is the proportion of the total cycle time in the inhalation phase. The specific average value and weighted sum are all basic mathematical operations, which will not be elaborated here.

[0039] Preferably, in order to more effectively analyze the degree of matching between the PEEP stage and the standard range corresponding to each historical inspiratory-expiratory cycle in the short-term respiratory pressure signal set, some implementations of the present invention quantify the matching degree from two dimensions: quantity proportion and pressure stability. Specific methods for obtaining the degree of matching include: Historical inspiratory-expiratory cycles whose PEEP pressure signals fall within a preset standard range are considered valid historical inspiratory-expiratory cycles. The proportion of valid inspiratory-expiratory cycles in the short-term respiratory pressure signal set is statistically analyzed. This proportion is the initial evaluation of the short-term respiratory pressure signal set based on the number of cycles meeting the standard range. In a specific implementation of this invention, the average pressure value of the PEEP pressure signal can be calculated. If the average pressure value is within a preset standard range, the corresponding historical inspiratory-expiratory cycle is considered a valid historical inspiratory-expiratory cycle.

[0040] The average PEEP pressure signal value of all effective historical inspiratory-expiratory cycles is used as the overall PEEP pressure value. The stability of the overall PEEP pressure value across all effective historical inspiratory-expiratory cycles is obtained. Higher stability indicates a more stable distribution across all PEEP phases in the short-term respiratory pressure signal set, making the signal set more reliable and the data more effective. In one specific implementation of this invention, the variance of all overall PEEP pressure values ​​is obtained, and this variance is negatively correlated to obtain the stability. The negative correlation mapping method can employ the reciprocal method. To prevent the denominator from being zero, the variance is truncated to a minimum value, for example, 0.01. If the calculated variance is less than 0.01, it is set to 0.01. After truncating the minimum value, the stability can be obtained directly by reciprocal.

[0041] The matching degree of the short-term respiratory pressure signal set is obtained based on the stability and the quantity proportion. In this embodiment of the invention, the product of the stability and the quantity proportion can be directly used as the matching degree.

[0042] Preferably, in this embodiment of the invention, the method for obtaining the frequency consistency of the pressure signal includes: For any historical inspiratory-expiratory cycle, the reciprocal of the time range is taken as the respiratory rate; In a short-term respiratory pressure signal set, the frequency consistency of the pressure signal is obtained based on the frequency difference between the respiratory rate of each historical inspiratory-expiratory cycle and the average respiratory rate. That is, if there are many historical inspiratory-expiratory cycles in the signal set whose respiratory rates deviate more from the average value, it indicates that the frequency distribution consistency of the pressure signal in the signal set is poor.

[0043] In one specific implementation of this invention, after obtaining the frequency difference between each respiratory rate and the average respiratory rate, the frequency differences are accumulated to obtain the overall frequency difference. The smaller the overall frequency difference, the greater the frequency consistency of the pressure signals in the signal set. Therefore, the sum of the overall frequency difference and a preset constant is used as the denominator, and the positive integer 1 is used as the numerator to obtain the pressure signal frequency consistency. The preset constant can be set to 0.01.

[0044] In the specific implementation of this invention, after the matching degree, overall stability and pressure signal frequency consistency are quantified, the three data can be directly fused by multiplication, and the product is mapped using the sigmoid function to obtain a state stability between 0 and 1.

[0045] Preferably, in this embodiment of the invention, in order to screen out the most effective standard respiratory cycle, the method for obtaining the standard respiratory cycle includes: If the stability of the target historical inhalation-exhalation cycle is greater than a preset stability threshold, it is marked as a valid historical cycle.

[0046] All valid historical cycles in the historical database are statistically analyzed, and consecutive valid historical cycles are merged to obtain several continuous time intervals. There may be many such continuous time intervals. In order to select a more effective quasi-respiratory cycle, the valid historical cycles within the longest continuous time interval are selected as the standard respiratory cycle. That is, if the duration of this continuous time interval is longer and the patient's breathing shows a long period of stable state, then the data within this continuous time interval are considered to be more reliable as standard data.

[0047] In a specific implementation of this invention, since the state stability has been normalized, the stability threshold can be set to 0.7.

[0048] Step S3: Compare the real-time inspiratory pressure signal, real-time expiratory pressure signal, and real-time PEEP pressure signal of each real-time inspiratory-expiratory cycle with the corresponding standard pressure data in the standard respiratory cycle to obtain the respiratory status of each real-time inspiratory-expiratory cycle; adjust the valve opening of the oxygen concentrator according to the respiratory status.

[0049] After obtaining a standard respiratory cycle, the inspiratory pressure, expiratory pressure, and PEEP pressure signals can be statistically analyzed to obtain standard signals in three dimensions. These standard signals represent the data under the current patient's normal and stable breathing state. Therefore, by comparing the real-time inspiratory pressure, expiratory pressure, and PEEP pressure signals of each real-time inspiratory-expiratory cycle with the corresponding standard pressure data in the standard respiratory cycle, the respiratory state of each real-time inspiratory-expiratory cycle compared to the stable and normal state can be obtained. Based on the real-time respiratory state, the valve opening of the oxygen concentrator can be adjusted to make the oxygen concentrator's parameter settings more suitable for the patient.

[0050] Preferably, in this embodiment of the invention, obtaining the respiratory state for each real-time inspiratory-expiratory cycle includes: Extract the standard inspiratory pressure signal, standard expiratory pressure signal, and standard PEEP pressure signal of a standard respiratory cycle; combine the range of the standard PEEP pressure signal with the preset standard range to obtain an adaptive standard range.

[0051] In a specific implementation of this invention, taking the standard inspiratory pressure signal as an example, all inspiratory pressure signals in a standard respiratory cycle are extracted. These inspiratory pressure signals may not be sequentially consistent; the longest temporal range is selected as the standard. Other signals are then aligned temporally using an interpolation algorithm. The average of the aligned inspiratory pressure signals yields the standard inspiratory pressure signal. Similarly, the standard expiratory pressure signal and the standard PEEP pressure signal can be obtained. Interpolation algorithms are well-known techniques to those skilled in the art; appropriate interpolation algorithms can be selected for processing during specific implementation.

[0052] In a specific implementation of this invention, after obtaining the standard PEEP pressure signal, the pressure signal range can be extracted, and its intersection with the standard range can be calculated. The resulting intersection range is used as the adaptive standard range. If the intersection is an empty set, the boundary point closest to the standard PEEP pressure signal range is selected as the adaptive standard range, meaning that the range contains only one data point.

[0053] For any real-time inspiratory-expiratory cycle, the real-time PEEP pressure signal range is compared with the adaptive standard range to determine the PEEP state of the real-time inspiratory-expiratory cycle. The PEEP state is either abnormal or normal, and abnormal PEEP includes insufficient PEEP and excessive PEEP. In a specific implementation of this invention, if the average signal value of the real-time PEEP pressure signal is within the adaptive standard range, it is determined that PEEP is normal; if it is below the adaptive standard range, it is determined that PEEP is insufficient; and if it is above the adaptive standard range, it is determined that PEEP is excessive.

[0054] At each PEEP state, the first deviation between the real-time inspiratory pressure signal and the standard inspiratory pressure signal during the real-time inspiratory-expiratory cycle, and the second deviation between the real-time expiratory pressure signal and the standard expiratory pressure signal are calculated. The inspiratory state is determined based on the first deviation, which includes inadequate inspiratory and excessive inspiratory. The expiratory state is determined based on the second deviation, which includes inadequate expiration and excessive expiration.

[0055] As a specific example, in the specific implementation of this embodiment of the invention, the first deviation and the second deviation can be quantified by the curve integral difference. Taking the first deviation as an example, the real-time inspiratory pressure signal and the standard inspiratory pressure signal can be time-aligned using a linear interpolation resampling method. After time alignment, the integral deviation between the inspiratory curves is calculated. ,in The first deviation, This is the real-time inspiratory pressure signal after timing alignment. This is the standard inspiratory pressure signal after timing alignment. The specific integration calculation and interpolation methods are well-known techniques to those skilled in the art and will not be elaborated here.

[0056] After obtaining the first and second deviations, considering that the inspiratory phase is a negative pressure process—the more forceful the inhalation, the deeper the pressure drops, and the lower the actual pressure; conversely, the weaker the inhalation, the shallower the pressure drops, and the higher the actual pressure—if the first deviation is greater than a preset first threshold, it indicates that the actual pressure is higher than the standard pressure, meaning the negative pressure drop is insufficient, indicating inadequate inhalation. If the first deviation is less than a negative second threshold, it indicates that the actual pressure is lower than the standard pressure, meaning the negative pressure drop is too deep, indicating excessive inhalation. Other conditions indicate that the patient's inhalation is normal.

[0057] Exhalation is a positive pressure process; the more forceful the exhalation, the higher the positive pressure, and the weaker the exhalation, the lower the positive pressure. Therefore, if the second deviation is greater than the preset third threshold, it indicates that the patient's actual expiratory pressure is too high, resulting in excessive exhalation; if the second deviation is less than a negative fourth threshold, it indicates that the patient's exhalation is insufficient. Other conditions indicate that the patient's exhalation is normal.

[0058] In a specific implementation of this invention, the process of setting the first threshold to the fourth threshold may include: (1) Calculate the absolute integral area corresponding to the standard inspiratory pressure signal and the standard expiratory pressure signal respectively to obtain the inspiratory reference area and the expiratory reference area.

[0059] (2) A preset physiological tolerance ratio coefficient K is introduced to shield normal respiratory fluctuations. The clinical routine recommendation is a value range of 10% to 20% (i.e., between 0.1 and 0.2). The value in this implementation is 0.1.

[0060] (3) Dynamically generate four thresholds: the first threshold = K × inspiratory reference area, the second threshold is the opposite of the first threshold; the third threshold = K × expiratory reference area, and the fourth threshold is the opposite of the third threshold. In the specific implementation of this invention, the first threshold is 7.5 and the third threshold is 6 after calculation. The specific calculation methods are all conventional mathematical methods for those skilled in the art, and will not be elaborated here.

[0061] Preferably, considering that a real-time pressure signal includes multiple real-time inspiratory-expiratory cycles, in order to effectively assess the current patient's respiratory status, the valve opening of the oxygen concentrator is adjusted according to the respiratory status, including: If the proportion of real-time inspiratory-expiratory cycles with normal PEEP exceeds a preset threshold, the oxygen concentrator valve parameters are maintained. Otherwise, for real-time inspiratory-expiratory cycles with abnormal PEEP, the inspiratory and expiratory states are statistically analyzed. Under both insufficient and excessive PEEP conditions, the number of real-time inspiratory-expiratory cycles for each inspiratory and expiratory state is counted, and the state with the highest number of cycles is selected as the patient's current real-time respiratory state. For example, the real-time respiratory state may include (insufficient PEEP, excessive inspiration) or (insufficient PEEP, insufficient inspiration), etc. This represents a two-dimensional comprehensive state result.

[0062] The valves of the oxygen concentrator's inhalation and exhalation channels are controlled according to the real-time breathing status.

[0063] In a specific implementation of this invention, if the real-time breathing state shows insufficient PEEP and insufficient inhalation, then both the inspiratory and expiratory channels need to be enhanced simultaneously; if both PEEP and expiratory functions are insufficient, then the inspiratory channel needs to be stabilized and the expiratory channel increased; if both PEEP and inhalation are excessive, then both the inspiratory and expiratory channels need to be reduced; if both PEEP and expiratory functions are excessive, then the inspiratory channel needs to be stabilized and the expiratory channel reduced. For other situations, if PEEP is insufficient but both inspiratory and expiratory states are normal, then the expiratory channel can be increased; if PEEP is excessive but both inspiratory and expiratory states are normal, then the expiratory channel can be reduced. In actual adjustment, adjustment compensation can be set. For example, every M inspiratory-expiratory cycles, the valve opening is adjusted by 1% according to the adjustment direction until the percentage of real-time inspiratory-expiratory cycles with normal PEEP exceeds a preset percentage threshold. The preset percentage threshold can be set to 80%.

[0064] In summary, this invention acquires airway pressure signals in real time and finely divides the inspiratory-expiratory cycle. Overall stability is obtained through consistency analysis of historical inspiratory and expiratory pressure signals. Furthermore, the degree of matching between the PEEP pressure signal and a preset standard range is considered, and valid historical cycles are identified and selected based on signal and frequency consistency. The real-time monitored inspiratory, expiratory, and PEEP pressure signals are compared with standard respiratory cycle data to quantify the deviation between the current respiratory state and the ideal model, thus providing a digital basis for adjusting the oxygen concentrator. This invention, by constructing an individualized standard model and implementing real-time state comparison and adjustment, achieves effective dynamic adjustment of the oxygen concentrator from passive constant oxygen supply to active adaptive respiratory support.

[0065] Based on the same inventive concept, the present invention also proposes an oxygen concentrator oxygen supply system with integrated positive end-expiratory pressure monitoring, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the steps of the oxygen concentrator oxygen supply method with integrated positive end-expiratory pressure monitoring described above.

[0066] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0067] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. An integrated positive end-expiratory pressure monitoring oxygen generator oxygen supply method, characterized in that, The method comprises: acquiring real-time pressure signals of a patient's respiratory tract during operation of an oxygen generator; dividing a plurality of real-time inhalation-exhalation cycles according to the change trend of the pressure signals, and extracting real-time inhalation pressure signals, real-time exhalation pressure signals and real-time PEEP pressure signals of each real-time inhalation-exhalation cycle; in a historical database of the patient, for any target historical inhalation-exhalation cycle, constructing short-time breathing pressure signals in adjacent historical inhalation-exhalation cycles; in the short-time breathing pressure signals, obtaining overall stability according to the consistency of the inhalation pressure signals and the exhalation pressure signals between any two historical inhalation-exhalation cycles; obtaining the matching degree of the PEEP pressure signal range in each historical inhalation-exhalation cycle in the short-time breathing pressure signals with a preset standard range; obtaining state stability of the short-time breathing pressure signals according to the overall stability, the matching degree, and the frequency consistency of the pressure signals between the historical inhalation-exhalation cycles; judging whether the target historical inhalation-exhalation cycle is an effective historical cycle according to the state stability; extracting the effective historical cycles in the historical database to construct standard breathing cycles; comparing the real-time inhalation pressure signals, the real-time exhalation pressure signals and the real-time PEEP pressure signals of each real-time inhalation-exhalation cycle with the corresponding standard pressure data in the standard breathing cycles to obtain the breathing state of each real-time inhalation-exhalation cycle; and adjusting the valve opening degree of the oxygen generator according to the breathing state.

2. The method of claim 1, wherein the integrated positive end-expiratory pressure monitoring oxygen machine oxygen supply method is characterized by, The method comprises: acquiring local maximum values of the real-time pressure signals, and the range between adjacent local maximum values is a real-time inhalation-exhalation cycle; in a real-time inhalation-exhalation cycle, the pressure change trend between adjacent time points is obtained one by one, if a time point is not a pressure decreasing trend and is a local minimum value, the time point is determined as an inhalation end point, the inhalation end point divides the real-time inhalation-exhalation cycle into an inhalation stage and an exhalation stage, a pressure change rate sequence is extracted in the exhalation stage, a time point in the pressure change rate sequence is taken as a starting time point of a PEEP stage when the absolute value of the time point is less than a preset change rate absolute value and the time of the time point is more than a preset time length, and the range between the starting time point of the PEEP stage and an exhalation stage termination time point is taken as a PEEP stage.

3. The method of claim 1, wherein the integrated positive end-expiratory pressure monitoring oxygen machine oxygen supply method is characterized by, The method comprises: taking the target historical inhalation-exhalation cycle as a starting point, searching a preset number of historical inhalation-exhalation cycles backward in time sequence, and counting the corresponding pressure signals to construct a short-time breathing pressure signal set.

4. The method of claim 1, wherein the integrated positive end-expiratory pressure monitoring oxygen machine oxygen supply method is characterized by, The method for obtaining the overall stability comprises: in the short-time breathing pressure signal set, acquiring the exhalation time proportion and the inhalation time proportion in each historical inhalation-exhalation cycle, taking the average inhalation time proportion of all historical inhalation-exhalation cycles as a first weight, and taking the average exhalation time proportion as a second weight; For any two historical inspiratory-expiratory cycles, the first similarity between inspiratory pressure signals and the second similarity between expiratory pressure signals are obtained using the dynamic time warping algorithm. The average first weight of the two historical inspiratory-expiratory cycles is used as the weight of the first similarity, and the average second weight of the two historical inspiratory-expiratory cycles is used as the weight of the second similarity. The two are then weighted and summed to obtain the signal consistency between the two historical inspiratory-expiratory cycles. The average signal consistency of the statistical short-term respiratory pressure signal set is used as the overall stationarity.

5. The method of claim 1, wherein the integrated positive end-expiratory pressure monitoring oxygen generator oxygen supply method is characterized by, The methods for obtaining the matching degree include: Historical inspiratory-expiratory cycles with PEEP pressure signals within a preset standard range are defined as effective historical inspiratory-expiratory cycles. The proportion of effective inspiratory-expiratory cycles in the short-term respiratory pressure signal set is statistically analyzed. The average PEEP pressure signal value of the effective historical inspiratory-expiratory cycles is defined as the overall PEEP pressure value. The stability of the overall PEEP pressure value of all effective historical inspiratory-expiratory cycles is obtained, and the matching degree of the short-term respiratory pressure signal set is obtained based on the stability and the proportion of the number of cycles.

6. The method of claim 1, wherein the integrated positive end-expiratory pressure monitoring oxygen generator oxygen supply method is characterized by, The method for obtaining the frequency consistency of the pressure signal includes: For any historical inspiratory-expiratory cycle, the reciprocal of the time range is taken as the respiratory rate; In a short-term respiratory pressure signal set, the frequency consistency of the pressure signal is obtained based on the frequency difference between the respiratory rate of each historical inspiratory-expiratory cycle and the average respiratory rate.

7. The method of claim 1, wherein the integrated positive end-expiratory pressure monitoring oxygen machine oxygen supply method is characterized by, The method for obtaining the standard respiratory cycle includes: If the stability of the target historical inhalation-exhalation cycle is greater than a preset stability threshold, it is marked as a valid historical cycle. All valid historical cycles in the historical database are statistically analyzed. Valid historical cycles that are consecutive in time are merged to obtain several consecutive time intervals. The valid historical cycle in the longest consecutive time interval is selected as the standard respiratory cycle.

8. The method of claim 1, wherein the integrated positive end-expiratory pressure monitoring oxygen generator oxygen supply method is characterized by, The process of obtaining the respiratory state for each real-time inspiratory-expiratory cycle includes: Extract the standard inspiratory pressure signal, standard expiratory pressure signal, and standard PEEP pressure signal of a standard respiratory cycle; combine the standard PEEP pressure signal range with the preset standard range to obtain an adaptive standard range; For any real-time inhalation-exhalation cycle, the real-time PEEP pressure signal range is compared with the adaptive standard range to determine the PEEP state of the real-time inhalation-exhalation cycle. The PEEP state is either abnormal or normal. Abnormal PEEP includes insufficient PEEP and excessive PEEP. At each PEEP state, the first deviation between the real-time inspiratory pressure signal and the standard inspiratory pressure signal during the real-time inspiratory-expiratory cycle, and the second deviation between the real-time expiratory pressure signal and the standard expiratory pressure signal are calculated. The inspiratory state is determined based on the first deviation, which includes normal inspiratory, insufficient inspiratory, and excessive inspiratory. The expiratory state is determined based on the second deviation, which includes normal expiratory, insufficient expiratory, and excessive expiratory.

9. The method of claim 8, wherein the method further comprises, The adjustment of the oxygen concentrator's valve opening based on respiratory status includes: If the proportion of real-time inspiratory-expiratory cycles with normal PEEP is greater than a preset threshold, the oxygen concentrator valve parameters are maintained; otherwise, for real-time inspiratory-expiratory cycles with abnormal PEEP, the inspiratory and expiratory states are counted. Under both insufficient and excessive PEEP conditions, the number of real-time inspiratory-expiratory cycles for each inspiratory and expiratory state is counted, and the state with the most cycles is selected as the patient's current real-time respiratory state. The valves of the oxygen concentrator's inspiratory and expiratory channels are controlled according to the real-time respiratory state.

10. An integrated positive end-expiratory pressure monitoring oxygen generator oxygen supply system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the oxygen supply method of an oxygen concentrator with integrated positive end-expiratory pressure monitoring as described in any one of claims 1 to 9.

Images