High-frequency ventilation method and system for preventing lung injury

Abstract

The invention relates to the technical field of high-frequency ventilation methods, in particular to a high-frequency ventilation method and system for preventing lung injury, comprising: acquiring historical control parameters of a high-frequency ventilator during anesthesia within a preset time period; real-time ventilation information is obtained, and correlation evaluation is conducted on the standardized historical control parameters and the real-time ventilation information; accumulating the real-time ventilatory capacity data in a time dimension, comparing the real-time ventilatory capacity data with a preset total ventilatory capacity requirement during anesthesia, and determining a standard grade based on a comparison difference; a first control strategy, a second control strategy and a third control strategy are generated according to the standard grade, and control parameters of the high-frequency breathing machine are optimized; and performing real-time driving control on the high-frequency breathing machine based on the optimized control parameter group. The problem of risks of volume injury, pressure injury and the like in the prior art is solved, the amplitude, the frequency, the average airway pressure and the inhaled oxygen concentration are dynamically optimized in the whole anesthesia process, and the total lung stretching amount is always kept within a safe range.

Description

Technical Field

[0001] This invention relates to the field of high-frequency ventilation methods, and more particularly to a high-frequency ventilation method and system for preventing lung injury. Background Technology

[0002] High-frequency ventilation (HFV), with its high frequency and small tidal volume, can achieve oxygenation and carbon dioxide clearance during anesthesia at lower airway pressures, and has been used in difficult airway and one-lung ventilation scenarios. HFV ventilators are essential equipment for maintaining oxygenation and carbon dioxide clearance during anesthesia. By achieving gas exchange at lower airway pressures, it theoretically reduces volumetric and barotrauma, and has been widely used in difficult airway, one-lung ventilation, and thoracic surgery anesthesia scenarios.

[0003] However, existing open-loop preset, experience compensation, or single-point instantaneous feedback modes can only make passive corrections to instantaneous physiological indicators. They cannot extract the optimal parameter-ventilation mapping law using historical big data, nor do they have a dynamic comparison mechanism between "cumulative ventilation" and the preoperative total ventilation target. As a result, during prolonged anesthesia, the total lung stretch is difficult to be systematically limited, and there are still risks such as volumetric injury and barotrauma. Summary of the Invention

[0004] The purpose of this invention is to provide a high-frequency ventilation method and system for preventing lung injury, which solves the problems of volumetric injury and barotrauma in the prior art. It dynamically optimizes the amplitude, frequency, mean airway pressure and inhaled oxygen concentration throughout the anesthesia process to keep the total lung stretch within a safe range.

[0005] To achieve the above objectives, the present invention provides a high-frequency ventilation method for preventing lung injury, comprising the following steps: S1. Obtain the historical control parameters of the high-frequency ventilator during anesthesia within a preset time period, preprocess the historical control parameters, and obtain standardized historical control parameters. S2. Obtain real-time ventilation information, evaluate the correlation between standardized historical control parameters and real-time ventilation information, and obtain the correlation evaluation results. S3. Accumulate the real-time ventilation data over time to obtain the cumulative ventilation volume, and compare it with the preset total ventilation volume requirement during anesthesia to obtain the comparison difference. Determine the compliance level based on the comparison difference. S4. Generate a first control strategy, a second control strategy, and a third control strategy based on the achievement level. Optimize the control parameters of the high-frequency ventilator based on the first control strategy, the second control strategy, the third control strategy, and the correlation evaluation results. S5. Real-time drive control of the high-frequency ventilator based on the optimized control parameter set.

[0006] In some embodiments of this application, in step S1, historical control parameters of the high-frequency ventilator during anesthesia within a preset time period are obtained, and the historical control parameters are preprocessed to obtain standardized historical control parameters, including: Historical control parameters include respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint; During anesthesia, obtain the respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint of the high-frequency ventilator; The respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint are sequentially cleaned to remove outliers and null values, and the data is transformed to obtain the feature vectors of each control parameter. The feature vectors of each control parameter are then integrated to obtain standardized historical control parameters.

[0007] In some embodiments of this application, in step S2, real-time ventilation information is acquired, and the correlation between standardized historical control parameters and real-time ventilation information is evaluated to obtain the correlation evaluation results, including: S21. Continuously collect the real-time ventilation of the high-frequency ventilator according to a fixed sampling period, and perform linear interpolation and low-pass filtering on missing or abnormal values ​​to generate a smooth real-time ventilation sequence. S22. Extract the same sampling cycle sequence corresponding to the current anesthesia stage from the standardized historical control parameter matrix, and construct a historical control parameter time alignment matrix; S23. Perform lag correlation analysis on the historical control parameter time alignment matrix and the real-time ventilation smoothing sequence, and compare the lag correlation analysis results with the preset correlation threshold to obtain the correlation evaluation results.

[0008] In some embodiments of this application, in S2, the expression for performing lag correlation analysis includes: ; in, This is a smoothed sequence of real-time ventilation. , , , , , These are the regression intercept, respiratory rate coefficient, tidal volume setpoint coefficient, oscillation pressure amplitude coefficient, respiratory ratio coefficient, and oxygen concentration setpoint coefficient, respectively. This is a characteristic of delayed respiratory rate. The hysteresis characteristic of the tidal volume setpoint The oscillation pressure amplitude exhibits a lag characteristic. This is a characteristic of respiratory lag. The oxygen concentration setpoint hysteresis characteristic. The optimal lag time is given by t, which is the time corresponding to the sampling point at a fixed time. Smooth the real-time ventilation sequence Compared with model predictions The difference is calculated using the following expression: ; in, To normalize the instantaneous error, Real-time ventilation smoothing sequence The sample standard deviation; Calculate the results of the lag correlation analysis The expression is: ; in, This is the allowable deviation threshold.

[0009] In some embodiments of this application, in step S3, the real-time ventilation data is accumulated over time to obtain the cumulative ventilation volume, including: ; in, Let be the instantaneous ventilation rate at the i-th sampling point. To accumulate ventilation, is the sampling period, and k is the total number of sampling points.

[0010] In some embodiments of this application, in step S3, the comparison is made with a preset total ventilation requirement during anesthesia to obtain a comparison difference. Determining the compliance level based on the comparison difference includes: The difference between the cumulative ventilation volume and the preset total ventilation volume required during anesthesia is calculated and compared, expressed as: ; in, To compare the differences, This is the preset total ventilation requirement during anesthesia; A threshold curve graph consisting of two threshold curves is generated based on the clinical allowable error ratio. The comparison difference located below the first threshold curve of the threshold curve graph is determined as the first compliance level; the comparison difference located above the first threshold curve and below the second threshold curve of the threshold curve graph is determined as the second compliance level; and the comparison difference located above the second threshold curve of the threshold curve graph is determined as the third compliance level.

[0011] In some embodiments of this application, in step S4, generating the first control strategy, the second control strategy, and the third control strategy based on the compliance level includes: First control strategy, second control strategy, and third control strategy are generated based on the first compliance level, second compliance level, and third compliance level, respectively.

[0012] In some embodiments of this application, in S4, the first control strategy includes: not adjusting the parameters, keeping the respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint at the previous cycle value, thereby accelerating the acquisition rate of real-time ventilation information; The second control strategy includes: adjusting respiratory rate, tidal volume setpoint, and oscillation pressure amplitude according to the configured importance index, and accelerating the acquisition rate of real-time ventilation information; The third control strategy includes adjusting the respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint according to the configured importance index, and accelerating the acquisition rate of real-time ventilation information.

[0013] In some embodiments of this application, a high-frequency ventilation system for preventing lung injury is also disclosed, comprising: The acquisition module is used to acquire historical control parameters of the high-frequency ventilator during anesthesia within a preset time period, and to preprocess the historical control parameters to obtain standardized historical control parameters. The evaluation module is used to acquire real-time ventilation information, evaluate the correlation between standardized historical control parameters and real-time ventilation information, and obtain the correlation evaluation results. The comparison module is used to accumulate real-time ventilation data over time to obtain cumulative ventilation, and compare it with the preset total ventilation requirement during anesthesia to obtain the comparison difference. The compliance level is determined based on the comparison difference. The optimization module is used to generate a first control strategy, a second control strategy, and a third control strategy based on the achievement level, and to optimize the control parameters of the high-frequency ventilator based on the first control strategy, the second control strategy, the third control strategy, and the correlation evaluation results. The control module is used to perform real-time drive control of the high-frequency ventilator based on the optimized set of control parameters.

[0014] The advantages and beneficial effects of this invention compared to the prior art are: 1. This invention simultaneously utilizes historical optimal parameters for real-time ventilation correlation evaluation and long-term constraint of cumulative ventilation target difference to lock the total lung stretch within a safe threshold range throughout the anesthesia process, reducing the risk of volumetric injury and barotrauma.

[0015] 2. This invention generates and switches between the first, second, and third control strategies in real time based on the achievement level, without the need for manual step-by-step adjustments. It can maintain the oxygenation and carbon excretion targets while automatically adjusting the control parameters such as the amplitude and frequency of the high-frequency ventilator to the optimal combination for lung protection, effectively reducing the incidence of atelectasis and over-expansion during anesthesia.

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] Figure 1 This is a flowchart of a high-frequency ventilation method for preventing lung injury according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a high-frequency ventilation system for preventing lung injury according to an embodiment of the present invention. Detailed Implementation

[0018] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product is in use. They are used only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," and "connect" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0019] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0020] like Figure 1 As shown, the present invention provides a high-frequency ventilation method for preventing lung injury, comprising the following steps: S1. Obtain the historical control parameters of the high-frequency ventilator during anesthesia within a preset time period, preprocess the historical control parameters, and obtain standardized historical control parameters.

[0021] S2. Obtain real-time ventilation information, evaluate the correlation between standardized historical control parameters and real-time ventilation information, and obtain the correlation evaluation results.

[0022] S3. Accumulate the real-time ventilation data over time to obtain the cumulative ventilation volume, and compare it with the preset total ventilation volume requirement during anesthesia to obtain the comparison difference. Determine the compliance level based on the comparison difference.

[0023] S4. Generate a first control strategy, a second control strategy, and a third control strategy based on the achievement level. Optimize the control parameters of the high-frequency ventilator based on the first control strategy, the second control strategy, the third control strategy, and the correlation evaluation results.

[0024] S5. Real-time drive control of the high-frequency ventilator based on the optimized control parameter set.

[0025] This invention generates and switches between first, second, and third control strategies in real time based on the achievement level, without the need for manual step-by-step adjustments. It can maintain the oxygenation and carbon excretion targets while automatically adjusting the control parameters such as the amplitude and frequency of the high-frequency ventilator to the optimal combination for lung protection, effectively reducing the incidence of atelectasis and over-expansion during anesthesia.

[0026] In some embodiments of this application, in step S1, historical control parameters of the high-frequency ventilator during anesthesia within a preset time period are obtained, and the historical control parameters are preprocessed to obtain standardized historical control parameters, including: Historical control parameters include respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint; During anesthesia, obtain the respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint of the high-frequency ventilator; The respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint are sequentially cleaned to remove outliers and null values, and the data is transformed to obtain the feature vectors of each control parameter. The feature vectors of each control parameter are then integrated to obtain standardized historical control parameters.

[0027] It's important to understand that respiratory rate refers to the number of oscillations per minute, directly determining the rate of gas exchange. A higher frequency means a smaller volume of air exhaled each time, which can lower airway pressure, but excessively high frequencies can increase intrapulmonary shear. Tidal volume setpoint refers to the volume of gas intended to be delivered to the lungs in a single oscillation. Oscillation pressure amplitude refers to the difference between peak inspiratory pressure and end-expiratory pressure, which is the actual driving force that propels gas into the lungs. Oxygen concentration setpoint refers to the volume percentage of oxygen in the delivered gas, determining the partial pressure of oxygen in arterial blood. Through data conversion, these parameters are transformed into a calculable, unified dimension, resulting in the characteristic vectors of each control parameter.

[0028] In some embodiments of this application, in step S2, real-time ventilation information is acquired, and the correlation between standardized historical control parameters and real-time ventilation information is evaluated to obtain the correlation evaluation results, including: S21. Continuously collect the real-time ventilation of the high-frequency ventilator according to a fixed sampling period, and perform linear interpolation and low-pass filtering on missing or abnormal values ​​to generate a smooth real-time ventilation sequence. S22. Extract the same sampling cycle sequence corresponding to the current anesthesia stage from the standardized historical control parameter matrix, and construct a historical control parameter time alignment matrix; S23. Perform lag correlation analysis on the historical control parameter time alignment matrix and the real-time ventilation smoothing sequence, and compare the lag correlation analysis results with the preset correlation threshold to obtain the correlation evaluation results.

[0029] In some embodiments of this application, in S2, the expression for performing lag correlation analysis includes: ; in, This is a smoothed sequence of real-time ventilation. , , , , , These are the regression intercept, respiratory rate coefficient, tidal volume setpoint coefficient, oscillation pressure amplitude coefficient, respiratory ratio coefficient, and oxygen concentration setpoint coefficient, respectively. This is a characteristic of delayed respiratory rate. The hysteresis characteristic of the tidal volume setpoint. The oscillation pressure amplitude exhibits a lag characteristic. This is a characteristic of respiratory lag. The oxygen concentration setpoint hysteresis characteristic. The optimal lag time is given by t, which is the time corresponding to the sampling point at a fixed time. Smooth the real-time ventilation sequence Compared with model predictions The difference is calculated using the following expression: ; in, To normalize the instantaneous error, Real-time ventilation smoothing sequence The sample standard deviation; Calculate the results of the lag correlation analysis The expression is: ; in, This is the allowable deviation threshold.

[0030] In some embodiments of this application, in S3, the real-time ventilation data is accumulated over time to obtain the cumulative ventilation volume, including: ; in, Let be the instantaneous ventilation rate at the i-th sampling point. To accumulate ventilation, is the sampling period, and k is the total number of sampling points.

[0031] This invention utilizes real-time ventilation correlation evaluation based on historical optimal parameters and long-term constraint of cumulative ventilation target difference to lock the total lung stretch within a safe threshold range throughout anesthesia, reducing the risk of volumetric injury and barotrauma.

[0032] In some embodiments of this application, in step S3, the comparison with the preset total ventilation requirement during anesthesia is performed to obtain the comparison difference. Determining the compliance level based on the comparison difference includes: The difference between the cumulative ventilation volume and the preset total ventilation volume required during anesthesia is calculated and compared, expressed as: ; in, To compare the differences, This is the preset total ventilation requirement during anesthesia; A threshold curve graph with two threshold curves is generated based on the clinical allowable error ratio. The comparison difference below the first threshold curve of the threshold curve graph is determined as the first compliance level; the comparison difference above the first threshold curve and below the second threshold curve of the threshold curve graph is determined as the second compliance level; and the comparison difference above the second threshold curve of the threshold curve graph is determined as the third compliance level.

[0033] In some embodiments of this application, in step S4, generating the first control strategy, the second control strategy, and the third control strategy based on the compliance level includes: First control strategy, second control strategy, and third control strategy are generated based on the first compliance level, second compliance level, and third compliance level, respectively.

[0034] In some embodiments of this application, in S4, the first control strategy includes: not adjusting the parameters, keeping the respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint at the previous cycle value, thereby accelerating the acquisition rate of real-time ventilation information; The second control strategy includes: adjusting respiratory rate, tidal volume setpoint, and oscillation pressure amplitude according to the configured importance index, and accelerating the acquisition rate of real-time ventilation information; The third control strategy includes adjusting the respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint according to the configured importance index, and accelerating the acquisition rate of real-time ventilation information.

[0035] The present invention will now be verified with reference to specific embodiments.

[0036] Ten minutes into anesthesia, the system read and standardized historical control parameters from the past 30 minutes. Real-time ventilation was then collected and compared with historical data for correlation assessment. The ventilation per second was then accumulated to obtain a cumulative value, compared with the pre-set total volume requirement, and the difference was calculated and determined to be the second target level. Based on this, the second control strategy was invoked, reducing the oscillation pressure amplitude from 22 cmH2O to 18 cmH2O and maintaining a respiratory rate of 600 breaths / min. After optimization, the high-frequency ventilator was activated. Five minutes later, EtCO2 decreased from 45 mmHg to 38 mmHg, lung compliance increased from 42 mL / cmH2O to 48 mL / cmH2O, and no atelectasis was observed.

[0037] During the historical control parameter acquisition phase, five null values ​​and two abnormal jump points appeared in the original records (the respiratory rate momentarily displayed as zero breaths per minute). The system used linear interpolation to fill in the missing values ​​and remove the jump points. Then, the respiratory rate, tidal volume, oscillation pressure amplitude, respiratory ratio, and oxygen concentration were normalized and integrated into a standardized historical control parameter vector. In the subsequent correlation evaluation, this vector was aligned with the real-time sequence and successfully matched the optimal correlation segment with a lag of eight seconds, ensuring that the subsequent strategy generation was based on reliable data.

[0038] The real-time ventilation sampling period was set to 50ms. The original signal had high-frequency spikes due to slight changes in the patient's position. After the system performed low-pass filtering, a smooth sequence was obtained and aligned with the standardized historical parameter matrix at the same sampling beat. The correlation coefficient was calculated to be 0.92, which is higher than the preset threshold of 0.85, and was judged to be "highly correlated". Therefore, it is sufficient to make minor adjustments based on the historical trend without the need for radical adjustments, thus reducing fluctuations in lung stress.

[0039] The lag correlation analysis module found that when the oscillation pressure amplitude changed 12 seconds in advance, it had the highest correlation with the current ventilation volume. Based on this, the system set 12 seconds as the optimal lag time and substituted this lag relationship into subsequent predictions. Clinical results showed that fine-tuning the amplitude 12 seconds in advance can keep the error of the next ventilation volume within ±5%, avoiding excessive overshoot.

[0040] At 60 minutes into the surgery, the system accumulated the measured ventilation per second in 1-second increments, resulting in a cumulative ventilation of 18.5L. This value was 1.5L lower than the preoperative target of 20L, falling within the second threshold range. Therefore, the second control strategy was triggered, which only increased the tidal volume setting by 5%. After 10 minutes, the cumulative volume rose to 19.6L, and the difference narrowed to 0.4L. The total lung stretch was still below the safe upper limit of 10mL / cmH2O / kg.

[0041] The threshold curve has two boundaries: the first threshold curve has an allowable error of ±10%, and the second threshold curve has an allowable error of ±20%. At 70 minutes into the operation, if the difference ΔV is -1.8L, which is between the first and second curves, the system determines it to be at the "second target level" and only increases the amplitude; if ΔV is -2.5L, which exceeds the second curve, it is upgraded to the "third target level", and the amplitude, respiratory ratio, and oxygen concentration are adjusted simultaneously to ensure an increasing lung protection level.

[0042] The system automatically matches the strategy library according to the achievement level: the first level corresponds to "zero adjustment", the second level corresponds to "dual parameter fine adjustment of amplitude and tidal volume", and the third level corresponds to "full adjustment of five parameters". After the anesthesiologist confirms, the strategy is issued with one click, without the need for manual input of each item, which shortens the reaction time by about 60% and avoids overexpansion caused by human delay.

[0043] After reaching the third level of compliance, the system sorts the data according to a preset importance index: oscillation pressure amplitude weight 0.4, respiratory rate 0.3, oxygen concentration 0.2, and respiratory ratio 0.1. Based on this, the amplitude is first increased by 15%, the frequency by 8%, the oxygen concentration by 5%, and the respiratory ratio is shortened by 10%. After the adjustment, SpO2 increased from 93% to 98%, while the driving pressure only increased by 2 cmH2O, and the CT scan showed no new atelectasis.

[0044] At the hardware level, the acquisition module exports the logs of the past 2 hours from the high-frequency ventilator via the RS-485 bus; the evaluation module performs hysteresis correlation calculations within the ARM processor; the comparison module writes the cumulative amount to the DDR cache in real time; the optimization module outputs three sets of strategies in parallel based on the FPGA; and the control module writes the final parameters back to the ventilator via the DAC board, with a total latency of less than 200ms, achieving real-time lung protection closed loop.

[0045] In some embodiments of this application, such as Figure 2 As shown, a high-frequency ventilation system for preventing lung injury is also disclosed, comprising: The acquisition module is used to acquire historical control parameters of the high-frequency ventilator during anesthesia within a preset time period, and to preprocess the historical control parameters to obtain standardized historical control parameters. The evaluation module is used to acquire real-time ventilation information, evaluate the correlation between standardized historical control parameters and real-time ventilation information, and obtain the correlation evaluation results. The comparison module is used to accumulate real-time ventilation data over time to obtain cumulative ventilation, and compare it with the preset total ventilation requirement during anesthesia to obtain the comparison difference. The compliance level is determined based on the comparison difference. The optimization module is used to generate a first control strategy, a second control strategy, and a third control strategy based on the achievement level, and to optimize the control parameters of the high-frequency ventilator based on the first control strategy, the second control strategy, the third control strategy, and the correlation evaluation results. The control module is used to perform real-time drive control of the high-frequency ventilator based on the optimized set of control parameters.

[0046] In this application, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. In case of any inconsistency, the meaning set forth in this specification or derived from the content described herein shall prevail. Furthermore, the terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit the scope of this application.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A high-frequency ventilation method for preventing lung injury, characterized in that, Includes the following steps: S1. Obtain the historical control parameters of the high-frequency ventilator during anesthesia within a preset time period, preprocess the historical control parameters, and obtain standardized historical control parameters. S2. Obtain real-time ventilation information, evaluate the correlation between standardized historical control parameters and real-time ventilation information, and obtain the correlation evaluation results. S3. Accumulate the real-time ventilation data over time to obtain the cumulative ventilation volume, and compare it with the preset total ventilation volume requirement during anesthesia to obtain the comparison difference. Determine the compliance level based on the comparison difference. S4. Generate a first control strategy, a second control strategy, and a third control strategy based on the achievement level, and optimize the control parameters of the high-frequency ventilator based on the first control strategy, the second control strategy, the third control strategy, and the correlation evaluation results. S5. Real-time drive control of the high-frequency ventilator based on the optimized control parameter set.

2. The high-frequency ventilation method for preventing lung injury according to claim 1, characterized in that, In step S1, historical control parameters of the high-frequency ventilator during anesthesia within a preset time period are obtained, and the historical control parameters are preprocessed to obtain standardized historical control parameters, including: The historical control parameters include respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint. During anesthesia, obtain the respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint of the high-frequency ventilator; The respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint are sequentially cleaned to remove outliers and null values, and the data is transformed to obtain the feature vectors of each control parameter. The feature vectors of each control parameter are then integrated to obtain standardized historical control parameters.

3. The high-frequency ventilation method for preventing lung injury according to claim 2, characterized in that, In step S2, real-time ventilation information is acquired, and the standardized historical control parameters are correlated with the real-time ventilation information to obtain the correlation evaluation results, including: S21. Continuously collect the real-time ventilation of the high-frequency ventilator according to a fixed sampling period, and perform linear interpolation and low-pass filtering on missing or abnormal values ​​to generate a smooth real-time ventilation sequence. S22. Extract the same sampling cycle sequence corresponding to the current anesthesia stage from the standardized historical control parameter matrix, and construct a historical control parameter time alignment matrix; S23. Perform lag correlation analysis on the historical control parameter time alignment matrix and the real-time ventilation smoothing sequence, and compare the lag correlation analysis results with the preset correlation threshold to obtain the correlation evaluation results.

4. The high-frequency ventilation method for preventing lung injury according to claim 3, characterized in that, In S2, the expression for performing lagged correlation analysis includes: ； in, This is a smoothed sequence of real-time ventilation. , , , , , These are the regression intercept, respiratory rate coefficient, tidal volume setpoint coefficient, oscillation pressure amplitude coefficient, respiratory ratio coefficient, and oxygen concentration setpoint coefficient, respectively. This is a characteristic of delayed respiratory rate. The hysteresis characteristic of the tidal volume setpoint The oscillation pressure amplitude exhibits a lag characteristic. This is a characteristic of respiratory lag. The oxygen concentration setpoint hysteresis characteristic. The optimal lag time is given by t, which is the time corresponding to the sampling point at a fixed time. Smooth the real-time ventilation sequence Compared with model predictions The difference is calculated using the following expression: ； in, To normalize the instantaneous error, Real-time ventilation smoothing sequence The sample standard deviation; Calculate the results of the lag correlation analysis The expression is: ； in, This is the allowable deviation threshold.

5. The high-frequency ventilation method for preventing lung injury according to claim 4, characterized in that, In step S3, the real-time ventilation data are accumulated over time to obtain the cumulative ventilation volume, which includes: ； in, Let be the instantaneous ventilation rate at the i-th sampling point. To accumulate ventilation, is the sampling period, and k is the total number of sampling points.

6. The high-frequency ventilation method for preventing lung injury according to claim 5, characterized in that, In step S3, the total ventilation volume during anesthesia is compared with the preset requirement to obtain the comparison difference. The achievement level is determined based on the comparison difference, including: The difference between the cumulative ventilation volume and the preset total ventilation volume required during anesthesia is calculated and compared, expressed as: ； in, To compare the differences, This is the preset total ventilation requirement during anesthesia; A threshold curve graph with two threshold curves is generated based on the clinical allowable error ratio. The comparison difference below the first threshold curve of the threshold curve graph is determined as the first compliance level; the comparison difference above the first threshold curve and below the second threshold curve of the threshold curve graph is determined as the second compliance level; and the comparison difference above the second threshold curve of the threshold curve graph is determined as the third compliance level.

7. The high-frequency ventilation method for preventing lung injury according to claim 6, characterized in that, In step S4, generating the first control strategy, the second control strategy, and the third control strategy based on the achievement level includes: First control strategy, second control strategy, and third control strategy are generated based on the first compliance level, second compliance level, and third compliance level, respectively.

8. The high-frequency ventilation method for preventing lung injury according to claim 7, characterized in that, In S4, the first control strategy includes: not adjusting the parameters, keeping the respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint at the previous cycle value, thereby accelerating the acquisition rate of real-time ventilation information; The second control strategy includes: adjusting respiratory rate, tidal volume setpoint, and oscillation pressure amplitude according to the configured importance index, and accelerating the acquisition rate of real-time ventilation information; The third control strategy includes adjusting the respiratory rate, tidal volume setpoint, oscillation pressure amplitude, respiratory ratio, and oxygen concentration setpoint according to the configured importance index, and accelerating the acquisition rate of real-time ventilation information.

9. A high-frequency ventilation system for preventing lung injury, characterized in that, include: The acquisition module is used to acquire historical control parameters of the high-frequency ventilator during anesthesia within a preset time period, and to preprocess the historical control parameters to obtain standardized historical control parameters. The evaluation module is used to acquire real-time ventilation information, evaluate the correlation between standardized historical control parameters and real-time ventilation information, and obtain the correlation evaluation results. The comparison module is used to accumulate real-time ventilation data over time to obtain cumulative ventilation, and compare it with the preset total ventilation requirement during anesthesia to obtain the comparison difference. The compliance level is determined based on the comparison difference. The optimization module is used to generate a first control strategy, a second control strategy, and a third control strategy according to the achievement level, and to optimize the control parameters of the high-frequency ventilator based on the first control strategy, the second control strategy, the third control strategy, and the correlation evaluation results. The control module is used to perform real-time drive control of the high-frequency ventilator based on the optimized set of control parameters.

Images