Invasive-noninvasive ventilation switching system and method based on modified gcs score

Abstract

The present application relates to the technical field of respirator control, in particular to invasive-noninvasive ventilation switching system and method based on modified GCS score. The method uses expiratory decay characteristics to construct passive reference flow rate conditions, decouples the real myogenic flow rate component from the mixed waveform; by comparing the characteristics of high and low load two groups of breathing cycles, the specific performance decay characteristics and response delay characteristics are extracted, and the mechanical saturation state when the ventilation benefit decreases while increasing the inspiratory effort is captured; based on the mechanical load saturation index and the modified GCS score, the interlocking decision mechanism is realized. The present application extracts the real breathing characteristics of the patient, quantifies the nonlinear saturation degree of the respiratory system under high load, increases the safety gate of the respiratory mechanics dimension on the basis of the consciousness evaluation, forms the double-dimensional safety interlocking decision, improves the high-risk switching operation of the central wake but mechanical decompensation, and improves the safety of ventilation switching.

Description

Technical Field

[0001] This invention relates to the field of ventilator control technology, specifically to an invasive-noninvasive ventilation switching system and method based on a modified GCS score. Background Technology

[0002] In the treatment of acute exacerbations of chronic obstructive pulmonary disease (AE-COPD), the invasive-noninvasive sequential ventilation strategy requires timely extubation and switching to noninvasive ventilation when the pulmonary infection control window is established. Clinically, the timing of this switch is primarily determined by assessing the patient's central consciousness using the Modified Glasgow Coma Scale (Modified GCS) and by evaluating respiratory capacity using standard respiratory parameters in pressure support ventilation (PSV) mode.

[0003] However, in pressure support ventilation mode, the positive pressure assistance provided by the ventilator can mask the patient's true respiratory mechanical load. Some patients, although their consciousness scores are within acceptable limits, may have their respiratory system in a critical state of mechanical saturation due to dynamic overinflation or airway trapping, exhibiting a latent decompensation of "high drive, low efficiency." Ignoring waveform recognition of this critical state and relying solely on the apparent state of consciousness for switching may result in the respiratory muscle load instantly exceeding the compensatory limit after the patient loses precise pressure support, leading to extubation failure and reintubation. Summary of the Invention

[0004] To address the technical problem in existing technologies that, without adding invasive sensors, cannot effectively remove the masking effect of positive pressure assist on the patient's true respiratory mechanics during pressure support ventilation, and to identify latent mechanical decompensation in conscious patients whose respiratory muscles are already in a critical state, the present invention aims to provide an invasive-noninvasive ventilation switching system and method based on a modified Glasgow Coma Scale (GCS) score. The specific technical solution adopted is as follows: This invention provides a method for invasive-noninvasive ventilation switching based on a modified GCS score, the method comprising: Collect flow rate data from patients in each respiratory cycle; construct a passive flow rate curve for the current inspiratory phase using the attenuation characteristics of the previous expiratory phase; separate the muscular flow rate component by the deviation between the measured inspiratory flow rate data and the passive flow rate curve in the current respiratory cycle; calculate the muscular inspiratory volume and peak flow rate delay time for a single respiratory cycle based on the muscular flow rate component. Based on the size of the muscular inspiratory volume, the respiratory cycles within the current analysis time window are divided into a low-load breathing group and a high-load breathing group. For the high-load breathing group and the low-load breathing group, the efficiency decay characteristics between the muscular inspiratory volume and the total inspiratory volume are analyzed, and the response delay characteristics between the peak flow rate delay time are also analyzed. The mechanical load saturation index is generated by nonlinearly coupling the efficiency decay characteristics and the response delay characteristics. Ventilation switching decisions were made based on the mechanical load saturation index and the patient's current modified Glasgow Coma Scale (GCS) score.

[0005] Furthermore, the method for obtaining the passive flow velocity curve includes: Extract flow rate data segments with attenuation characteristics during the expiratory phase of the previous respiratory cycle in the current respiratory cycle. The expiratory attenuation constant is obtained through regression analysis; the peak airway pressure during the inspiratory phase of the current respiratory cycle is obtained, and the difference between the peak airway pressure and the airway pressure at the last moment of the expiratory phase of the previous respiratory cycle is taken as the pressure difference; the ratio of the pressure difference to the estimated airway resistance constant is taken as the expected maximum flow rate. Based on the expected maximum flow rate and the expiratory decay constant, a passive flow rate curve is generated that decays sequentially with the current inspiratory phase.

[0006] Furthermore, the method for obtaining the muscular flow velocity component includes: Calculate the difference between the measured flow velocity data at the current inhalation stage and the value at the same moment in the passive flow velocity curve, and use it as the flow velocity difference. When the flow velocity difference is negative, the preset zero work value is used as the effective inhalation work value at that moment. When the flow velocity difference is non-negative, the flow velocity difference is used as the effective inhalation work value at the corresponding moment. The sequence of effective inspiratory work values ​​over time is taken as the muscular flow component.

[0007] Furthermore, the calculation of the muscular inspiratory volume and peak flow delay time of a single respiratory cycle based on the muscular flow component includes: Integrate the muscular flow component during the current inspiratory phase to obtain the muscular inspiratory volume; use the time difference between the moment of the maximum amplitude of the muscular flow component and the start time of the current inspiratory phase as the peak flow delay time.

[0008] Furthermore, the step of dividing the respiratory cycles within the current analysis time window into a low-load respiratory group and a high-load respiratory group includes: Use a preset number of consecutive historical respiratory cycles before the current respiratory cycle as the analysis time window; obtain the dispersion of muscular inspiratory volume for all respiratory cycles within the current analysis time window; When the dispersion is higher than the preset dispersion threshold, all respiratory cycles in the analysis time window are sorted in ascending order of muscular inspiratory volume values ​​to obtain a volume sequence. The first preset proportion of respiratory cycles in the volume sequence are grouped into a low-load breathing group; the last preset proportion of respiratory cycles in the volume sequence are grouped into a high-load breathing group.

[0009] Furthermore, the method for obtaining the performance degradation characteristics includes: For any breathing group between the low-load breathing group and the high-load breathing group, the muscular inspiratory volume and total inspiratory volume corresponding to each respiratory cycle in the breathing group are used as data pairs. Linear regression analysis is performed based on all data pairs in the breathing group, and the slope of the regression line is obtained as the volume expansion rate of the breathing group. Calculate the difference in volume expansion rate between the low-load breathing group and the high-load breathing group. If the difference is non-negative, use the difference as the performance decay characteristic; otherwise, use the preset decay value as the performance decay characteristic.

[0010] Furthermore, the method for obtaining the response delay characteristics includes: For any breathing group between the low-load breathing group and the high-load breathing group, the average peak flow rate delay time of all respiratory cycles in that breathing group is taken as the average delay time of that breathing group. The extent to which the mean delay of the high-load breathing group was higher than that of the low-load breathing group was analyzed to obtain response delay characteristics.

[0011] Furthermore, the method for obtaining the mechanical load saturation index includes: By using a logarithmic function to map the upper limit of the response delay characteristics, and then fusing the processed response delay characteristics with the performance decay characteristics, the mechanical load saturation index is obtained.

[0012] Furthermore, the decision to switch ventilation based on the mechanical load saturation index and the current patient's modified Glasgow Coma Scale (GCS) score includes: When the improved GCS score is greater than or equal to the preset score safety threshold and the mechanical load saturation index is lower than the preset mechanical safety threshold, a switchable instruction is generated and output. When the improved GCS score is greater than or equal to the preset score safety threshold and the mechanical load saturation index is not lower than the preset mechanical safety threshold, a switch prohibition command is generated and an insufficient reserve warning is output at the same time. When the improved GCS score is lower than the preset score safety threshold, a switch prohibition instruction is generated and a consciousness impairment signal is output at the same time.

[0013] The present invention also provides an invasive-noninvasive ventilation switching system based on a modified GCS score, comprising 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 the invasive-noninvasive ventilation switching method based on a modified GCS score as described above.

[0014] The present invention has the following beneficial effects: This invention utilizes expiratory attenuation characteristics to construct a passive baseline flow rate, decoupling the true muscular flow rate component from the mixed waveform, and quantifying the patient's actual inspiratory work under non-invasive conditions. Secondly, by comparing the characteristics of high- and low-load respiratory cycles, it specifically extracts efficacy attenuation features and response delay features. The former quantifies the loss of ventilation gain due to lung tissue sclerosis, while the latter identifies the deterioration of flow response caused by airway trapping, thereby capturing the mechanical saturation state where ventilation benefits decrease with increased inspiratory effort. Finally, an interlocked decision-making mechanism is implemented based on the mechanical load saturation index and the modified GCS score, supplementing the existing modified GCS score with necessary mechanical constraints to improve the safety of invasive-noninvasive sequential ventilation switching strategies. This invention extracts the patient's true respiratory characteristics to quantify the nonlinear saturation degree of the respiratory system under high load, adds a safety gating dimension of respiratory mechanics to the consciousness assessment, forming a two-dimensional safety interlocked decision, improving the safety of high-risk switching operations for centrally conscious but mechanically decompensated patients, and enhancing the safety of ventilation switching. Attached Figure Description

[0015] 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.

[0016] Figure 1 The flowchart illustrates a method for switching between invasive and non-invasive ventilation based on a modified GCS score, as provided in one embodiment of the present invention. Detailed Implementation

[0017] 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 invasive-noninvasive ventilation switching system and method based on a modified GCS score proposed 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.

[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 invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details a specific scheme for an invasive-noninvasive ventilation switching system and method based on a modified GCS score provided by the present invention.

[0020] Please see Figure 1 The diagram illustrates a flowchart of an invasive-noninvasive ventilation switching method based on a modified GCS score, according to an embodiment of the present invention. The method includes the following steps: S1: Collect the patient's flow rate data for each respiratory cycle; construct the passive flow rate curve for the current inspiratory phase using the attenuation characteristics of the patient's previous expiratory phase; separate the muscular flow rate component by the deviation between the measured inspiratory flow rate data and the passive flow rate curve in the current respiratory cycle; calculate the muscular inspiratory volume and peak flow rate delay time for a single respiratory cycle based on the muscular flow rate component.

[0021] The ventilator's ventilation process has a periodic rhythm, and the patient's respiratory status needs to be accurately monitored in terms of the respiratory cycle as the smallest unit. Therefore, real-time dynamic monitoring is performed using each respiratory cycle during the patient's ventilator ventilation process as the processing unit, and flow rate data is collected in real time. The airway flow rate value at each moment can be obtained through the flow rate sensor built into the ventilator. The respiratory cycle consists of a continuous inspiratory phase and an expiratory phase. The initiation time of inspiration, the end time of inspiration, the initiation time of expiration, and the end time of expiration are identified based on the positive and negative changes and duration of the flow rate value, thereby dividing the inspiratory and expiratory phases of each respiratory cycle.

[0022] Because lung tissue has the physiological characteristic of passive recoil, no active muscle work is involved during the patient's expiration phase, and the airway flow velocity exhibits an exponential decay pattern over time. Therefore, the flow velocity waveform during the expiration phase naturally possesses decay characteristics. Flow velocity data segments with decay characteristics can reflect the passive dynamic state of lung tissue and have key reference value for constructing an ideal passive flow velocity curve. In this embodiment of the invention, during the expiration phase of the previous respiratory cycle, a flow velocity data segment with decay characteristics is extracted. Specifically, the peak flow point of the expiration phase of the previous respiratory cycle is located, and a range from the peak flow point to a certain percentage of the peak flow is selected, such as a flow velocity data segment that decays from 75% to 25% of the peak flow point. This flow velocity data segment is in the laminar flow-dominated stage, conforms to the linear decay characteristics of the single-compartment model, and can reflect the core mechanical characteristics of passive lung tissue recoil, characterizing the combined state of airway resistance and lung compliance.

[0023] Furthermore, the expiratory attenuation constant is obtained through regression analysis. Specifically, logarithmic linear regression analysis is performed on the extracted flow rate data segments, and the expiratory attenuation constant is obtained by fitting the data using a single-compartment model formula. In particular, to ensure the reliability of the parameters, if the number of samples in the sampled segments is less than the preset minimum sample size, such as 5, the fitting conditions are deemed insufficient, and the expiratory attenuation constant value from the previous period is forcibly used. This ensures the stability and reliability of the parameters, ultimately achieving the acquisition of the expiratory attenuation constant.

[0024] The peak airway pressure during the inspiratory phase of the current respiratory cycle can be acquired using a pressure sensor built into the ventilator; this is the maximum airway pressure during the inspiratory phase. The difference between the peak airway pressure and the airway pressure at the end of the expiratory phase of the previous respiratory cycle is defined as the pressure differential, reflecting the effective driving pressure provided by the ventilator during the inspiratory phase. The ratio of this pressure differential to the estimated airway resistance constant is defined as the expected maximum flow rate, reflecting the maximum airflow velocity achievable solely through ventilator pressure without active muscle work. The airway resistance can be set to a fixed constant based on the type of artificial airway and is not restricted here, ensuring the accuracy of the baseline value.

[0025] Finally, based on the expected maximum flow rate and the expiratory decay constant, a passive flow rate curve is generated that decays sequentially with the current inspiratory phase. Passive lung recoil causes the airflow velocity to decay exponentially over time. In one specific embodiment of this invention, the start of the inspiratory phase is taken as the zero point. Combining the expected maximum flow rate and the expiratory decay constant, the theoretical flow rate value at each moment is calculated using an exponential decay formula, ultimately forming a continuous passive flow rate curve that reflects the theoretical flow rate change assuming the patient's respiratory muscles are completely relaxed. It should be noted that the passive flow rate curve constructed in this application is not an absolute reproduction of physiological passive exhalation, but rather an engineering reference baseline constructed based on the expiratory time constant. This baseline is used to relatively quantify the changing trend of respiratory effort, rather than for measuring an absolute physical quantity.

[0026] As an example, the expression for obtaining the velocity value at each moment on the passive velocity curve is: In the formula, Represented as the first The first respiratory cycle The flow velocity value at a given time. Represented as the first The expected maximum flow rate per respiratory cycle. Represented as the first The start of the inspiratory phase of a respiratory cycle Represented as the first The expiratory attenuation constant per respiratory cycle It is represented as a negative exponential function with the natural constant as the base.

[0027] The passive flow rate curve serves as a baseline reference assuming the patient is not performing any work. During the actual inspiratory phase, the patient's inspiratory muscle contractions superimposed on the passive waveform, causing flow rate distortion. The amplitude and timing of this distortion directly reflect the efficiency of the respiratory pump. Analyzing the muscular flow rate component reflects the patient's actual inspiratory effort and characterizes the intensity of respiratory drive. In this embodiment, the difference between the measured flow rate data at the current inspiratory phase and the value at the same instant in the passive flow rate curve is calculated as the flow rate difference. This difference reflects the deviation between the measured flow rate and the theoretical passive flow rate; the deviation represents the flow rate change caused by the active work of the respiratory muscles.

[0028] When the flow rate difference is negative, it indicates that the measured flow rate is lower than the theoretical passive flow rate, and there is no effective active inspiratory work. The preset zero work value is used as the effective inspiratory work value at that time. When the flow rate difference is non-negative, it indicates that the measured flow rate is higher than the theoretical passive flow rate, and the difference is generated by the active work of the respiratory muscles. The flow rate difference is used as the effective inspiratory work value at the corresponding time. The preset zero work value can be set to 0. Finally, the sequence of effective inspiratory work values ​​over time is used as the muscular flow rate component, reflecting the patient's active inspiratory effort at each time point.

[0029] Furthermore, key mechanical features are quantitatively extracted based on the muscular flow velocity component. In this embodiment of the invention, the muscular flow velocity component during the current inspiratory phase is integrated to obtain the muscular inspiratory volume. The integration interval covers the entire inspiratory phase. By integrating, the constantly changing flow velocity value is converted into the total gas volume, representing the amount of gas additionally obtained by the patient through the work of their own respiratory muscles, thereby quantifying the patient's inspiratory effort. The time difference between the moment corresponding to the maximum amplitude of the muscular flow velocity component and the start of the current inspiratory phase is used as the peak flow velocity delay time, that is, the lag time between the moment when the muscular flow velocity reaches its peak value and the start of inspiration. This indicates the latency of the airflow response after the patient's respiratory muscles have worked, and is used to assess changes in airway resistance and lung tissue ventilation response efficiency.

[0030] This completes the decoupling and quantification of the mechanical characteristics of a single respiratory cycle, providing standardized data input for subsequent nonlinear load analysis.

[0031] S2: Based on the size of the muscular inspiratory volume, the respiratory cycles within the current analysis time window are divided into a low-load breathing group and a high-load breathing group; for the high-load breathing group and the low-load breathing group, the efficiency decay characteristics between the muscular inspiratory volume and the total inspiratory volume are analyzed, and the response delay characteristics between the peak flow rate delay time are also analyzed; the mechanical load saturation index is generated by nonlinearly coupling the efficiency decay characteristics and the response delay characteristics.

[0032] Traditional monitoring focuses only on the statistical mean of respiratory parameters, which makes it difficult to capture the nonlinear mechanical hardening characteristics of lung tissue under different loads. Therefore, a refined analysis is carried out by stratified comparison, accumulating and grouping historical respiratory cycle data. Then, based on the parameter differences of different load groups, the mechanical attenuation law is analyzed to generate quantitative risk indicators, so as to analyze the compensatory margin of the patient's respiratory system in the mechanical dimension and identify the latent decompensation state with high drive but low efficiency.

[0033] Therefore, the load is first differentiated by the numerical distribution of muscular inspiratory volume to amplify the differences in mechanical characteristics under different respiratory efforts. In this embodiment of the invention, a preset number of consecutive historical respiratory cycles preceding the current respiratory cycle are used as the analysis time window. That is, the current respiratory cycle and the preset number of consecutive historical respiratory cycles preceding it are used together as the current analysis time window. The analysis time window is also a first-in-first-out data buffer used to store recent respiratory mechanics data. The preset number can be set to 50, which can be adjusted by the implementer. The window is updated iteratively with the respiratory cycle to ensure that the analysis is always based on the latest valid data.

[0034] To further obtain the dispersion of muscular inspiratory volume for all respiratory cycles within the current analysis time window, in this embodiment of the invention, the coefficient of variation of all muscular inspiratory volume values ​​within the window is calculated to reflect the degree of data fluctuation and distribution characteristics. Higher dispersion indicates more significant differences in patients' respiratory effort, which is more conducive to stratified comparative analysis. When the dispersion is low, i.e., when the dispersion is not higher than a preset dispersion threshold, the patient's breathing pattern is determined to be in an extremely monotonous steady state, lacking the gradient differences required to calculate mechanical characteristics. In this case, the mechanical load saturation index of the previous moment remains unchanged. The preset dispersion threshold can be set to 0.1, and the implementer can flexibly adjust it according to different clinical scenarios.

[0035] When the dispersion is higher than the preset dispersion threshold, it indicates that the respiratory data has a sufficient dynamic range and can be effectively analyzed in a hierarchical manner. All respiratory cycles in the analysis time window are sorted in ascending order of muscular inspiratory volume values ​​to obtain a volume sequence. This sequence achieves an orderly arrangement from the basic respiratory state to the extreme exertion state.

[0036] The first preset proportion of respiratory cycles in the volumetric sequence constitutes the low-load breathing group. Specifically, the preset proportion can be set to 30%, which can be adjusted by the implementer according to the specific implementation situation. The data of the low-load breathing group represents the patient's basic biomechanical characteristics at rest or with slight exertion, serving as a benchmark reference. The last preset proportion of respiratory cycles in the volumetric sequence constitutes the high-load breathing group, that is, selecting the sample of the greatest exertion. The data of this group reflects the patient's biomechanical performance under extreme load, used to detect the compensatory limit of the system.

[0037] Because lung tissue is prone to mechanical hardening, which leads to decreased compliance and reduced ventilation efficiency under high load conditions, it is necessary to compare the volume expansion efficiency of high and low load groups to identify the decay pattern of ventilation efficiency. Therefore, we first analyzed the correlation characteristics between muscular inspiratory volume and total inspiratory volume to quantify the ventilation conversion efficiency under different loads.

[0038] In this embodiment of the invention, for any respiratory group between the low-load and high-load respiratory groups, the muscular inspiratory volume and total inspiratory volume corresponding to each respiratory cycle in that respiratory group are used as data pairs. Each data pair corresponds to the matching relationship between the patient's active work volume and the actual total ventilation volume in one respiratory cycle. For example, if the muscular inspiratory volume in a certain respiratory cycle is 0.2L and the total inspiratory volume is 0.8L, then the two constitute a corresponding data pair (0.2, 0.8). Based on all data pairs in the respiratory group, linear regression analysis is performed, and the least squares method is used to fit the regression line. The slope of the regression line is used as the volume expansion rate of the respiratory group. The magnitude of the slope reflects the total inspiratory volume that can be converted from a unit of muscular inspiratory volume. The higher the slope, the better the ventilation conversion efficiency, and vice versa, reflecting the ventilation efficiency of the respiratory system under this load condition.

[0039] The difference in volume expansion rate between the low-load and high-load breathing groups is calculated, i.e., the volume expansion rate of the low-load group minus the volume expansion rate of the high-load group. If the difference is non-negative, it indicates that the ventilation efficiency under high-load conditions is lower than that under low-load conditions, resulting in diminishing marginal returns. That is, increasing effort yields less volume gain, indicating a significant efficacy decline. In this case, the difference is used as the efficacy decline characteristic. Otherwise, a preset decline value is used as the efficacy decline characteristic, indicating that there is currently no significant efficacy decline, the lung tissue is still within the linear elastic range, and no sclerosis has occurred, so it does not need to be included in risk considerations. The preset decline value can be set to 0 to ensure that it only makes a positive contribution when efficacy decline is observed.

[0040] Furthermore, if patients under high load conditions have intrinsic positive end-expiratory pressure or airway trapping, it will lead to a slower airflow response and a longer peak flow delay time. Therefore, we further analyzed the differences in peak flow delay time between high and low load groups to quantify the degree of deterioration of airway resistance.

[0041] In this embodiment of the invention, for any one of the low-load breathing group and the high-load breathing group, the average peak flow rate delay time of all breathing cycles in the breathing group is taken as the delay average of the breathing group. By statistical averaging, the random fluctuation of a single breath is eliminated, reflecting the overall level of airflow response delay under the load group.

[0042] By analyzing the degree to which the mean delay of the high-load breathing group is higher than that of the low-load breathing group, a response delay characteristic is obtained, reflecting the degree of deterioration of airway resistance with increasing respiratory load. Specifically, the difference between the mean delay of the high-load breathing group and the mean delay of the low-load breathing group is used as the response delay characteristic. The larger the response delay characteristic, the more severe the airflow response delay under high load, and the larger the ratio, the more severe the resistance deterioration.

[0043] It should be noted that, in order to avoid division by zero anomalies, a preset lower limit value is set to 0.05. If the average delay value of the low-load group is lower than the preset lower limit value, the preset lower limit value is used as the average delay value of the low-load breathing group in the ratio to obtain the response delay characteristics.

[0044] Finally, by comprehensively considering the gain decay in the capacity dimension and the resistance deterioration in the time dimension, the compensatory margin of the respiratory system can be fully evaluated, thus coupling the calculation of the mechanical load saturation index. In a specific embodiment of this invention, a logarithmic function is used to map the response delay feature to a constrained upper limit. Only when the response delay feature is positive does it contribute positively to the index. At the same time, the constraint of the logarithmic function and the upper limit of 1 avoids the excessive influence of a single extreme value on the overall index. Specifically, the non-negative part is first extracted for subsequent analysis by analyzing the response delay feature and the magnitude of 0. Then, a logarithmic operation with base 2 is performed on the non-negative part. Finally, the result is limited to a range not exceeding 1 by a minimum value function.

[0045] Furthermore, by fusing the processed response delay characteristics and performance degradation characteristics, a mechanical load saturation index is obtained. Specifically, the processed response delay characteristics and performance degradation characteristics are multiplied to obtain the mechanical load saturation index. As an example, the expression for the mechanical load saturation index is: In the formula, This is expressed as the current mechanical load saturation index. This represents the current performance degradation characteristics. This is represented as the current response latency characteristic. This is represented as a function that takes the maximum value. Represented as a function that takes the minimum value, It is represented as a logarithmic function with base 2.

[0046] This completes the deep decoupling and nonlinear quantification of respiratory mechanical load status, outputting a single risk indicator that can be used for decision-making.

[0047] S3: Make ventilation switching decisions based on the mechanical load saturation index and the current modified GCS score of the patient.

[0048] By analyzing the mechanical load saturation index, the mechanical compensation margin and load saturation degree of the patient's respiratory system are quantified. This can be combined with traditional clinical consciousness assessment indicators to construct a two-dimensional interlocking mechanism of central drive and peripheral mechanics, which is used together for ventilation switching decision analysis. In this embodiment of the invention, the patient's current modified Glasgow Coma Scale (GCS) score, which is usually assessed and entered into the system regularly by medical staff, and the mechanical load saturation index are calculated in real time to ensure the freshness and validity of the decision data.

[0049] The modified Glasgow Coma Scale (GCS) score primarily reflects the patient's central nervous system status and airway protection capacity, serving as a prerequisite for determining whether the patient meets the basic conditions for extubation. The mechanical load saturation index specifically reflects the compensatory margin of the respiratory pump under the current load, acting as a key gating mechanism to prevent pseudo-tolerance.

[0050] In this embodiment of the invention, when the modified GCS score is less than a preset safety threshold, it is determined that the patient's central consciousness level is not up to standard, lacking the basic conditions for spontaneous cooperation with non-invasive ventilation, which easily leads to clinical risks such as aspiration and ineffective ventilation. A switching prohibition instruction is generated, and a consciousness impairment signal is simultaneously output. This instruction reminds medical staff that the patient's current consciousness level does not meet the requirements for ventilation switching. The preset safety threshold can be set to 10, which can be adjusted by the implementer.

[0051] Furthermore, when the modified GCS score is greater than or equal to the preset safety threshold and the mechanical load saturation index is lower than the preset mechanical safety threshold, the patient is deemed to have both good central awareness and cooperation, sufficient respiratory muscle mechanical reserve, and no risk of latent mechanical decompensation. This fully meets the dual-dimensional conditions for invasive-noninvasive ventilation switching, generating and outputting a switchable command that reminds medical staff to perform the ventilation switching operation. The preset mechanical safety threshold can be set to 0.5, which can be adjusted by the practitioner.

[0052] Furthermore, when the modified GCS score is greater than or equal to the preset safety threshold and the mechanical load saturation index is not lower than the preset mechanical safety threshold, it is determined that although the patient's level of consciousness is within acceptable limits, the respiratory muscles are already under mechanical load saturation and mechanical reserve is insufficient. If the patient is abruptly switched to non-invasive ventilation, it is easy for the load to exceed the compensatory limit of the respiratory muscles, leading to ventilation failure or re-intubation and other adverse consequences. A switch prohibition instruction is generated, and a reserve insufficiency warning is simultaneously output. This instruction warning can be recorded after being issued, reminding medical staff to continue to optimize invasive ventilation parameters, improve the patient's respiratory mechanical status, and then reassess the timing of the switch.

[0053] The various ventilation switching commands, consciousness impairment signals, and reserve depletion warnings output are all real-time visualized and multi-terminal synchronized. They can be clearly displayed on the ventilator control panel, bedside monitor, and central monitoring system. At the same time, all decision results and judgment indicators are automatically stored in the patient's respiratory monitoring database, which facilitates medical staff to retrospectively analyze the dynamic changes in the patient's respiratory status and level of consciousness.

[0054] In summary, this invention utilizes expiratory attenuation characteristics to construct a passive baseline flow rate, decoupling the true muscular flow component from the mixed waveform, and quantifying the patient's actual inspiratory work under non-invasive conditions. Secondly, by comparing the characteristics of respiratory cycles at high and low loads, it specifically extracts efficacy attenuation features and response delay features. The former quantifies the loss of ventilation gain due to lung tissue sclerosis, while the latter identifies the deterioration of flow response caused by airway trapping, thereby capturing the mechanical saturation state where ventilation benefits decrease with increased inspiratory effort. Finally, based on the mechanical load saturation index and the modified GCS score, an interlocked decision-making mechanism is implemented, supplementing the existing modified GCS score with necessary mechanical constraints to improve the safety of invasive-noninvasive sequential ventilation switching strategies. This invention extracts the patient's true respiratory characteristics to quantify the nonlinear saturation degree of the respiratory system under high load, adds a safety gating dimension of respiratory mechanics to the consciousness assessment, forming a two-dimensional safety interlocked decision, improving the safety of high-risk switching operations for centrally conscious but mechanically decompensated patients, and enhancing the safety of ventilation switching.

[0055] The present invention also provides an invasive-noninvasive ventilation switching system based on a modified GCS score, comprising 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 the invasive-noninvasive ventilation switching method based on a modified GCS score as described above.

[0056] 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.

[0057] 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. A method for switching between invasive and non-invasive ventilation based on a modified GCS score, characterized in that, The method includes: Collect flow rate data from patients in each respiratory cycle; construct a passive flow rate curve for the current inspiratory phase using the attenuation characteristics of the previous expiratory phase; separate the muscular flow rate component by the deviation between the measured inspiratory flow rate data and the passive flow rate curve in the current respiratory cycle; calculate the muscular inspiratory volume and peak flow rate delay time for a single respiratory cycle based on the muscular flow rate component. Based on the size of the muscular inspiratory volume, the respiratory cycles within the current analysis time window are divided into a low-load breathing group and a high-load breathing group. For the high-load breathing group and the low-load breathing group, the efficiency decay characteristics between the muscular inspiratory volume and the total inspiratory volume are analyzed, and the response delay characteristics between the peak flow rate delay time are also analyzed. The mechanical load saturation index is generated by nonlinearly coupling the efficiency decay characteristics and the response delay characteristics. Ventilation switching decisions were made based on the mechanical load saturation index and the patient's current modified Glasgow Coma Scale (GCS) score.

2. The invasive-noninvasive ventilation switching method based on the improved GCS score according to claim 1, characterized in that, The method for obtaining the passive flow velocity curve includes: Extract flow rate data segments with attenuation characteristics during the expiratory phase of the previous respiratory cycle in the current respiratory cycle. The expiratory attenuation constant is obtained through regression analysis; the peak airway pressure during the inspiratory phase of the current respiratory cycle is obtained, and the difference between the peak airway pressure and the airway pressure at the last moment of the expiratory phase of the previous respiratory cycle is taken as the pressure difference; the ratio of the pressure difference to the estimated airway resistance constant is taken as the expected maximum flow rate. Based on the expected maximum flow rate and the expiratory decay constant, a passive flow rate curve is generated that decays sequentially with the current inspiratory phase.

3. The invasive-noninvasive ventilation switching method based on the improved GCS score according to claim 1, characterized in that, The method for obtaining the muscular flow velocity component includes: Calculate the difference between the measured flow velocity data at the current inhalation stage and the value at the same moment in the passive flow velocity curve, and use it as the flow velocity difference. When the flow velocity difference is negative, the preset zero work value is used as the effective inhalation work value at that moment. When the flow velocity difference is non-negative, the flow velocity difference is used as the effective inhalation work value at the corresponding moment. The sequence of effective inspiratory work values ​​over time is taken as the muscular flow component.

4. The invasive-noninvasive ventilation switching method based on the improved GCS score according to claim 1, characterized in that, The calculation of the muscular inspiratory volume and peak flow delay time of a single respiratory cycle based on the muscular flow component includes: Integrate the muscular flow component during the current inspiratory phase to obtain the muscular inspiratory volume; use the time difference between the moment of the maximum amplitude of the muscular flow component and the start time of the current inspiratory phase as the peak flow delay time.

5. The invasive-noninvasive ventilation switching method based on the improved GCS score according to claim 1, characterized in that, The step of dividing respiratory cycles within the current analysis time window into a low-load respiratory group and a high-load respiratory group includes: Use a preset number of consecutive historical respiratory cycles before the current respiratory cycle as the analysis time window; obtain the dispersion of muscular inspiratory volume for all respiratory cycles within the current analysis time window; When the dispersion is higher than the preset dispersion threshold, all respiratory cycles in the analysis time window are sorted in ascending order of muscular inspiratory volume values ​​to obtain a volume sequence. The first preset proportion of respiratory cycles in the volume sequence are grouped into a low-load breathing group; the last preset proportion of respiratory cycles in the volume sequence are grouped into a high-load breathing group.

6. The invasive-noninvasive ventilation switching method based on the improved GCS score according to claim 1, characterized in that, The method for obtaining the performance degradation characteristics includes: For any breathing group between the low-load breathing group and the high-load breathing group, the muscular inspiratory volume and total inspiratory volume corresponding to each respiratory cycle in the breathing group are used as data pairs. Linear regression analysis is performed based on all data pairs in the breathing group, and the slope of the regression line is obtained as the volume expansion rate of the breathing group. Calculate the difference in volume expansion rate between the low-load breathing group and the high-load breathing group. If the difference is non-negative, use the difference as the performance decay characteristic; otherwise, use the preset decay value as the performance decay characteristic.

7. The invasive-noninvasive ventilation switching method based on the improved GCS score according to claim 1, characterized in that, The method for obtaining the response delay characteristics includes: For any breathing group between the low-load breathing group and the high-load breathing group, the average peak flow rate delay time of all respiratory cycles in that breathing group is taken as the average delay time of that breathing group. The extent to which the mean delay of the high-load breathing group was higher than that of the low-load breathing group was analyzed to obtain response delay characteristics.

8. The method for switching between invasive and non-invasive ventilation based on the improved GCS score according to claim 1, characterized in that, The method for obtaining the mechanical load saturation index includes: By using a logarithmic function to map the upper limit of the response delay characteristics, and then fusing the processed response delay characteristics with the performance decay characteristics, the mechanical load saturation index is obtained.

9. The invasive-noninvasive ventilation switching method based on the improved GCS score according to claim 1, characterized in that, The decision to switch ventilation based on the mechanical load saturation index and the current modified Glasgow Coma Scale (GCS) score includes: When the improved GCS score is greater than or equal to the preset score safety threshold and the mechanical load saturation index is lower than the preset mechanical safety threshold, a switchable instruction is generated and output. When the improved GCS score is greater than or equal to the preset score safety threshold and the mechanical load saturation index is not lower than the preset mechanical safety threshold, a switch prohibition command is generated and an insufficient reserve warning is output at the same time. When the improved GCS score is lower than the preset score safety threshold, a switch prohibition instruction is generated and a consciousness impairment signal is output at the same time.

10. An invasive-noninvasive ventilation switching system based on a modified GCS score, 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 invasive-noninvasive ventilation switching method based on the improved GCS score as described in any one of claims 1 to 9.

Images