A method and apparatus for assessing the compliance of static mechanical operations

Abstract

This invention discloses a method and device for assessing the compliance of static mechanical operations. This invention combines the characteristics of the respiratory waveform during breath-holding at the end of inspiration, accurately identifying whether the breath-holding operation performed by the doctor is compliant by analyzing the waveform morphology. This facilitates waveform analysis after the doctor performs the breath-holding operation and reduces the workload of medical staff. If applied to a ventilator, it can provide real-time reminders to the doctor whether the breath-holding operation just performed met the requirements.

Description

Technical Field

[0001] This invention relates to a method and apparatus for evaluating the compliance of static mechanical operations, belonging to the field of medical signal processing. Background Technology

[0002] In the medical field, mechanical ventilation is a common treatment method, widely used in intensive care and surgical procedures. Mechanical ventilation aims to support a patient's respiratory function by providing a normal breathing pattern and maintaining adequate oxygenation and carbon dioxide removal. However, incorrect ventilation settings and parameter selection can lead to ventilator-related complications and even threaten the patient's life.

[0003] By assessing a patient's static biomechanical parameters, physicians can understand the patient's respiratory function characteristics and adjust ventilation modes and parameter settings to provide safer and more effective mechanical ventilation support. In clinical practice, the gold standard for static biomechanical manipulation is end-inspiratory pausing, recording parameters such as end-inspiratory pressure and volume. Through the measurement and analysis of these parameters, physicians can obtain information about the patient's lung function, such as lung compliance (i.e., lung elasticity) and airway resistance. These indicators are crucial for assessing ventilation compliance, alveolar collapse, and determining appropriate ventilation strategies.

[0004] Due to the complex clinical environment and significant noise interference, the static mechanical parameters measured by the ventilator may be inaccurate. Furthermore, different healthcare professionals have varying levels of proficiency in static mechanical procedures, and inexperienced operators may make errors when performing end-inspiratory pauses. All these factors can lead to non-compliant static mechanical procedures, affecting the accuracy of calculated parameters such as lung compliance, and consequently influencing the physician's judgment and decisions regarding the patient. Therefore, ensuring the compliance of static mechanical procedures is crucial for accurately assessing a patient's lung function and ventilation status. Summary of the Invention

[0005] This invention provides a reliable method for determining the compliance of static biomechanical procedures. This will provide physicians with more reliable and accurate static biomechanical parameters, improving the reliability of assessments of patient lung function and ventilation status, and aiding in treatment decisions.

[0006] The technical solution adopted by this invention to solve its technical problem is:

[0007] A method for assessing compliance with static mechanical operations, comprising:

[0008] Obtain the set S of single breaths that perform static mechanical operations in a waveform record, where set S contains at least one single breath;

[0009] For each breath in set S, perform the following operation:

[0010] Feature points were extracted from the airway pressure and flow waveform sampling data of a single breath, including the inspiratory start point A, inspiratory end point B, breath-hold start point C, breath-hold end point E, and midpoint of the breath-hold segment D. The inspiratory start point is the first sampling point of the airway pressure and flow waveform, and the inspiratory end point B is the sampling point corresponding to the maximum airway pressure in the airway pressure waveform. The breath-hold start point C, breath-hold end point E, and midpoint of the breath-hold segment D were determined using the following method:

[0011] The initial start point of the defined segment is the first sampling point in the flow velocity waveform where the flow velocity is less than -0.2 of the previous sampling point and greater than 1 / 2 of the maximum flow velocity. If the minimum flow velocity in the flow velocity waveform is less than -10 L / Min or the last sampling point in the flow velocity waveform is 6 L / Min or more greater than the minimum flow velocity, then the initial end point is the first sampling point before the minimum flow velocity that is less than 1 / 3 of the minimum flow velocity. Otherwise, the initial end point is the sampling point corresponding to the minimum flow velocity.

[0012] Determine if the restricted segment meets the conditions; otherwise, narrow the range of the restricted segment, redetermine the start and end points of the restricted segment, and repeat the determination until the conditions are met. The specific conditions are: the slope of the fitted line of the restricted segment is less than the set value B_K, the velocity difference between the start and end points of the restricted segment is less than the set value B_DIFF, and the difference between the maximum and minimum values ​​of the velocity in the velocity waveform of the restricted segment is less than the set value B_RANGE. The start and end points of the restricted segment that meet the conditions are the breath-hold start point C and the breath-hold end point E, respectively, and the midpoint between the start and end points of the restricted segment is the breath-hold midpoint D.

[0013] Based on the extracted feature points, the inhalation segment AB, the breath-holding segment CE, and the second half of the breath-holding segment DE were divided.

[0014] Features extracted from airway pressure and flow waveform sampling data of a single breath include: the number of sampling points of the single breath waveform (length), the number of sampling points of the breath-hold interval (CE) (b_length), the proportion of the breath-hold interval (CE) to the entire respiratory cycle (b_rate), the mean of the flow waveform, the difference between the maximum and mean of the flow waveform in the inspiratory interval (AB) (f_in_sub_max_mean), the mean of the flow waveform in the breath-hold interval (CE) (f_b_mean), the slope of the fitted line of the pressure waveform in the breath-hold interval (CE) (p_b_k), the standard deviation of the pressure waveform in the breath-hold interval (CE) (p_b_std), the slope of the fitted line of the pressure waveform in the second half of the breath-hold interval (DE) (p_b_k_half), and the difference between the maximum and minimum values ​​of the pressure waveform in the second half of the breath-hold interval (DE) (p_b_p_p_half).

[0015] Determine whether a single static mechanical operation is compliant based on the extracted features:

[0016] If b_rate is less than or equal to the threshold B_RATE, it means that there was no breath-holding action at the end of inspiration during this breath, and it is not a static mechanical operation, so the prediction result is negative.

[0017] Otherwise, the prediction is considered non-compliant if any of the following five conditions are met:

[0018] ① If b_length is less than or equal to the threshold B_LENGTH_1, it means that the breath-holding time is too short and the plateau pressure has not yet stabilized.

[0019] ② If f_b_mean is less than or equal to the threshold F_B_MEAN, it indicates that the end-of-inspiratory pause operation is not compliant and the requirement of zero flow rate at the end of inspiratory inspiration has not been met.

[0020] ③ If f_in_sub_max_mean is greater than the threshold F_IN_SUB_MAX_MEAN, it indicates that the flow rate during the intake of air in this static mechanical operation is not a square wave air delivery.

[0021] ④ If f_mean is greater than the threshold F_MEAN, it indicates that the average flow rate of this breath is too high.

[0022] ⑤ If p_b_k_half is greater than the threshold P_B_K_HALF, it indicates that the platform pressure in the latter half of the air-holding section of this static mechanical operation is unstable.

[0023] Otherwise, compare p_b_std with the threshold P_B_STD:

[0024] ① If p_b_std is less than or equal to the threshold P_B_STD, it indicates that the platform pressure fluctuation of this static mechanical operation is small. Further determine the breath-holding time. If b_length is less than or equal to the threshold B_LENGTH_2 and length is less than or equal to the threshold LENGTH, it indicates that the breath-holding time of this static mechanical operation is reasonable and the prediction is qualified. Otherwise, it indicates that the breath-holding time of this static mechanical operation is unreasonable and the prediction is non-compliant.

[0025] ② If p_b_std is greater than the threshold P_B_STD, it indicates that the plateau pressure of this static mechanical operation has significant fluctuations. Further assessment of plateau pressure stability is needed. If p_b_k is less than or equal to the threshold P_B_K and p_b_p_p_half is less than or equal to the threshold P_B_P_P_HALF, it indicates that this static mechanical operation has a stable plateau pressure, and the prediction result is acceptable. Otherwise, it indicates that the plateau pressure of this static mechanical operation is unstable, and the prediction is non-compliant.

[0026] Furthermore, the acquisition of the set S of single breaths performing static mechanical operations in a waveform record specifically involves:

[0027] Examine the ventilation pattern of a single breath in the waveform recording to determine whether the breath was a static mechanical operation. The first breath with the ventilation pattern VCV and the previous 5 breaths of that breath constitute a set S.

[0028] Furthermore, the sampling frequency of the airway pressure and flow rate waveform sampling data for a single breath is above 50Hz.

[0029] Furthermore, the process of narrowing the scope of the defined segment and redetermining the start and end points of the defined segment specifically involves:

[0030] Obtain the median of the flow velocity waveform in the current limited segment; compare the difference between the start point, end point and median of the current limited segment, move the end with the larger difference inward by several sampling points to narrow the range of the limited segment, and redetermine the start point and end point of the limited segment.

[0031] Furthermore, it also includes:

[0032] If there is a compliant single breath in set S, then the waveform record is considered to have a compliant static mechanical operation.

[0033] Furthermore, in the feature extraction of airway pressure and flow waveform sampling data based on a single breath, if there is no breath-holding segment in a single breath, features are obtained through a filling strategy. Specifically, the number of sampling points b_length in the breath-holding segment CE is filled to 0; the mean f_b_mean of the flow waveform in the breath-holding segment CE, the standard deviation p_b_std of the pressure waveform in the breath-holding segment CE, the slope p_b_k of the fitted line of the pressure waveform in the breath-holding segment CE, the slope p_b_k_half of the fitted line of the pressure waveform in the second half of the breath-holding segment DE, and the difference p_b_p_p_half between the maximum and minimum values ​​of the pressure waveform in the second half of the breath-holding segment DE are filled to 999.

[0034] A static mechanical operation compliance assessment device, comprising:

[0035] The data acquisition module is used to acquire a set S of single breaths that perform static mechanical operations in a waveform record, where set S contains at least one single breath.

[0036] The single-breath compliance assessment module is used to assess the compliance of each breath in set S, including:

[0037] The feature segmentation unit is used to extract feature points based on the airway pressure and flow rate waveform sampling data of a single breath and to segment the airway into an inspiratory segment AB, a breath-holding segment CE, and a second half of the breath-holding segment DE. The feature points include the inspiratory start point A, the inspiratory end point B, the breath-hold start point C, the breath-hold end point E, and the midpoint of the breath-holding segment D. The inspiratory start point is the first sampling point of the airway pressure and flow rate waveform, and the inspiratory end point B is the sampling point corresponding to the maximum airway pressure value in the airway pressure waveform. The breath-hold start point C, the breath-hold end point E, and the midpoint of the breath-holding segment D are determined using the following method:

[0038] The initial start point of the defined segment is the first sampling point in the flow velocity waveform where the flow velocity is less than -0.2 of the previous sampling point and greater than 1 / 2 of the maximum flow velocity. If the minimum flow velocity in the flow velocity waveform is less than -10 L / Min or the last sampling point in the flow velocity waveform is 6 L / Min or more greater than the minimum flow velocity, then the initial end point is the first sampling point before the minimum flow velocity that is less than 1 / 3 of the minimum flow velocity. Otherwise, the initial end point is the sampling point corresponding to the minimum flow velocity.

[0039] Determine if the restricted segment meets the conditions; otherwise, narrow the range of the restricted segment, redetermine the start and end points of the restricted segment, and repeat the determination until the conditions are met. The specific conditions are: the slope of the fitted line of the restricted segment is less than the set value B_K, the velocity difference between the start and end points of the restricted segment is less than the set value B_DIFF, and the difference between the maximum and minimum values ​​of the velocity in the velocity waveform of the restricted segment is less than the set value B_RANGE. The start and end points of the restricted segment that meet the conditions are the breath-hold start point C and the breath-hold end point E, respectively, and the midpoint between the start and end points of the restricted segment is the breath-hold midpoint D.

[0040] The feature extraction unit is used to extract features based on the airway pressure and flow waveform sampling data of a single breath, including: the number of sampling points of the single breath waveform (length), the number of sampling points of the breath-holding CE (b_length), the proportion of the breath-holding CE to the entire respiratory cycle (b_rate), the mean of the flow waveform, the difference between the maximum and mean of the flow waveform in the inspiratory AB (f_in_sub_max_mean), the mean of the flow waveform in the breath-holding CE (f_b_mean), the slope of the fitted line of the pressure waveform in the breath-holding CE (p_b_k), the standard deviation of the pressure waveform in the breath-holding CE (p_b_std), the slope of the fitted line of the pressure waveform in the second half of the breath-holding DE (p_b_k_half), and the difference between the maximum and minimum values ​​of the pressure waveform in the second half of the breath-holding DE (p_b_p_p_half).

[0041] The compliance assessment unit is used to determine whether a single static mechanical operation is compliant based on the extracted features.

[0042] (1) If b_rate is less than or equal to the threshold B_RATE, then no static mechanical operation was performed on the breath;

[0043] (2) Otherwise, if one of the following conditions is met, the prediction is unqualified: b_length is less than or equal to the threshold B_LENGTH_1 or f_b_mean is less than or equal to the threshold F_B_MEAN or f_in_sub_max_mean is greater than the threshold F_IN_SUB_MAX_MEAN or f_mean is greater than the threshold F_MEAN or p_b_k_half is greater than the threshold P_B_K_HALF;

[0044] (3) Otherwise, further determine the standard deviation p_b_std of the pressure waveform at CE during the breath-holding segment: If p_b_std is less than or equal to the threshold P_B_STD, then perform the following judgment:

[0045] If b_length is less than or equal to the threshold B_LENGTH_2 and length is less than or equal to the threshold LENGTH, then the respiratory static mechanics operation is compliant.

[0046] Otherwise, the static mechanical operation of breathing is not compliant;

[0047] If p_b_std is greater than the threshold P_B_STD, further determine the slope p_b_k of the fitted line of the pressure waveform in the breath-holding segment CE and the difference p_b_p_p_half between the maximum and minimum values ​​of the pressure waveform in the second half of the breath-holding segment DE. If p_b_k is less than or equal to the threshold P_B_K and p_b_p_p_half is less than or equal to the threshold P_B_P_P_HALF, then the respiratory static mechanics operation is compliant; otherwise, the respiratory static mechanics operation is non-compliant.

[0048] Furthermore, it also includes: a comprehensive compliance judgment module, used to determine whether there is a compliant static mechanical operation in the waveform record based on the set S: if there is a compliant single breath in the set S, then the waveform record is considered to have a compliant static mechanical operation.

[0049] The main advantages of this invention are: it accurately identifies whether a doctor's breath-holding maneuver is compliant by analyzing waveform morphology characteristics. This facilitates waveform analysis after a doctor performs a breath-holding maneuver and reduces the workload of medical staff. If applied to a ventilator, it can provide real-time reminders to doctors regarding the compliantness of their recent breath-holding maneuver. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of the process of the present invention.

[0051] Figure 2This is a schematic diagram illustrating the method of finding feature points on different types of airway pressure and flow rate waveforms, where a represents a compliant static mechanical operation waveform, b represents a non-compliant static mechanical operation waveform, and c represents a non-static mechanical operation waveform.

[0052] Figure 3 This is a flowchart illustrating the process of determining whether a static mechanical operation is compliant based on characteristics during a single breath in this invention. Detailed Implementation

[0053] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] like Figure 1 As shown, this invention provides a method for assessing the compliance of static mechanical procedures. By identifying characteristic points on the airway pressure and flow rate waveforms, the respiratory cycle is segmented. Then, features are calculated based on these segments, and the compliance of the static mechanical procedures is determined according to the feature values. The specific implementation steps are as follows:

[0055] S1. Obtain the set S of single breaths that perform static mechanical operations in a waveform record; set S contains at least one single breath, specifically:

[0056] The determination of whether a single breath is a static mechanical maneuver is made by examining the ventilation pattern of the waveform recording. If the ventilation pattern is VCV, then the breath is considered to be a static mechanical maneuver. Furthermore, because there is a certain delay in switching ventilator ventilation modes in clinical practice, judging solely by the displayed ventilation mode may miss static mechanical maneuvers. Therefore, the first 5 breaths in the record with the first ventilation mode of VCV are also included in set S; that is, set S consists of the breath with the first ventilation mode of VCV and the 5 breaths preceding that breath.

[0057] S2. For each breath in set S, perform the following operation:

[0058] S2.1 Acquire airway pressure and flow rate waveform sampling data for a single breath, with a sampling frequency of 50Hz or higher.

[0059] S2.2 Extract feature points from airway pressure and flow rate waveform sampling data of a single breath, such as... Figure 2 As shown, this includes the inhalation start point A, the inhalation end point B, the breath-holding start point C, the breath-holding end point E, and the midpoint D of the breath-holding segment. The specific extraction method is as follows:

[0060] S2.2.1. The first sampling point of the airway pressure and flow rate waveform is regarded as the inhalation start point A.

[0061] S2.2.2. The sampling point corresponding to the maximum airway pressure in the airway pressure waveform is regarded as the end point of inhalation B.

[0062] S2.2.3 The starting point C of breath-holding, the ending point E of breath-holding, and the midpoint D of the breath-holding segment are determined by the following method:

[0063] First, define the limiting segment by first defining the initial range of the breath-holding segment, and then finding the start and end points of the breath-holding segment within this initial range. Specifically:

[0064] The first point in the flow velocity waveform where the flow velocity is less than -0.2 of the previous sampling point's flow velocity but greater than 1 / 2 of the maximum flow velocity value is taken as the initial starting point C' of the limited segment.

[0065] If the minimum flow velocity is normal, i.e., less than -10 L / min, or if the last sampling point of the flow velocity waveform is 6 L / min or more greater than the minimum flow velocity, then the first sampling point less than 1 / 3 of the minimum flow velocity before the minimum flow velocity is taken as the initial and final point of the limited segment. Otherwise, the sampling point corresponding to the minimum flow velocity is regarded as the initial and final point E' of the limited segment.

[0066] If the limit segment does not exist, it is considered that there is no breath-hold segment CE.

[0067] If a defined segment exists, determine the start point C of the breath-holding period, the end point E of the breath-holding period, and the midpoint D of the breath-holding segment by following these steps:

[0068] Calculate the median of the bounded segment on the flow velocity waveform, denoted as f_median.

[0069] To determine if the restricted section meets the following conditions: Check if the slope of the fitted straight line between the start point C' and end point E' of the restricted section is less than the set value B_K = 1, if the velocity difference between the start point C' and end point E' is less than the set value B_DIFF = 10 L / Min, and if the difference between the maximum and minimum values ​​of the velocity waveform in the restricted section is less than the set value B_RANGE = 15 L / Min. If these conditions are met, then the start point C' of the restricted section is the start point C of the breath-holding, the end point E' is the end point E of the breath-holding, and the midpoint between the start point C' and end point E' is the midpoint D of the breath-holding section.

[0070] If the above conditions are not met, the range of the defined segment is narrowed, the start and end points of the defined segment are redefined and judged, until the conditions are met; specifically, the difference between the flow velocity at the start point C' and f_median, and the difference between the flow velocity at the end point E' and f_median are compared. The end with the larger difference is moved inward by several sampling points (preferably one) to narrow the range of the defined segment.

[0071] If the condition is still not met even when the range of the limited segment is narrowed down to 0, it is considered that no breath-holding segment CE has been found.

[0072] S2.3. Based on the extracted feature points, divide the inhalation segment AB, the breath-holding segment CE, and the second half of the breath-holding segment DE.

[0073] S2.4. Feature extraction based on airway pressure and flow rate waveform sampling data from a single breath, comprising ten features:

[0074] length: The number of sampling points in a single breath waveform;

[0075] b_length: The number of sampling points in the CE segment of the breath-holding section;

[0076] b_rate: The proportion of the breath-holding CE to the entire respiratory cycle, i.e., b_length / length;

[0077] f_mean: The mean value of the flow velocity waveform;

[0078] f_in_sub_max_mean: The difference between the maximum value and the mean value of the flow velocity waveform in the intake section AB, that is, the maximum value of the flow velocity waveform in the intake section AB minus the mean value of the flow velocity waveform in the intake section AB;

[0079] f_b_mean, the mean value of the flow velocity waveform on the CE section of the breath-holding segment;

[0080] p_b_k: The slope of the fitted straight line of the pressure waveform in the breath-holding segment CE;

[0081] p_b_std: Standard deviation of the pressure waveform at CE during the breath-holding segment;

[0082] p_b_k_half: The slope of the fitted straight line of the pressure waveform in the second half of the breath-holding phase of DE;

[0083] p_b_p_p_half: The difference between the maximum and minimum values ​​of the pressure waveform in the second half of the breath-holding DE, that is, the maximum value of the pressure waveform in the second half of the breath-holding DE minus the minimum value.

[0084] Furthermore, in this step, for a single breath without a breath-holding segment, it can be directly determined that there was no static mechanical operation. To facilitate standardized processing, the missing values ​​of the breath-holding segment characteristics in single breaths without a breath-holding segment can be filled using the following strategy:

[0085] For the number of sampling points b_length in the breath-holding segment CE, missing values ​​will be filled with 0.

[0086] For the mean f_b_mean of the flow velocity waveform in the breath-holding section CE, the standard deviation p_b_std of the pressure waveform in the breath-holding section CE, the slope p_b_k of the fitted line of the pressure waveform in the breath-holding section CE, the slope p_b_k_half of the fitted line of the pressure waveform in the second half of the breath-holding section DE, and the difference p_b_p_p_half between the maximum and minimum values ​​of the pressure waveform in the second half of the breath-holding section DE, missing values ​​are filled with 999.

[0087] Furthermore, in this step, single breaths with lengths outside the range [L_MIN, L_MAX] are excluded from the set S. Generally, single breaths with lengths outside the range [50, 1000] are excluded.

[0088] S2.5. Determine whether a single static mechanical operation is compliant based on the extracted features:

[0089] like Figure 2 and 3 As shown, based on the characteristics of the respiratory waveform during the breath-holding action at the end of inspiration, and according to the relationship between 10 feature values ​​and corresponding thresholds, it is determined whether a single static mechanical operation is compliant. "No" indicates that no static mechanical operation was performed during that breath. The following thresholds are set by this invention based on experience and experimental results, but are not limited to them:

[0090] (1) If b_rate <= 0.002, it means that there was no breath-holding action at the end of inhalation in this static mechanical operation, and the prediction result is negative.

[0091] (2) Otherwise, if one of the following conditions is met, the prediction is unqualified:

[0092] b_length<=17 indicates that the breath-holding time is too short and the plateau pressure cannot be stabilized.

[0093] f_b_mean <= -0.5726, indicating that the pause operation at the end of inhalation is not compliant.

[0094] The value of f_in_sub_max_mean > 3.5512 indicates that the flow rate waveform during inhalation is not a square wave.

[0095] The f_mean > 2.5075 indicates that the mean flow rate of this breath is relatively high.

[0096] p_b_k_half>1.5105 indicates that the plateau pressure did not decrease steadily in the latter half of the breath-holding phase.

[0097] If neither of the conditions in (1) and (2) above are met, then the standard deviation p_b_std of the pressure waveform at CE during the breath-holding segment should be further determined:

[0098] If p_b_std <= 0.845, it indicates that the plateau pressure does not fluctuate significantly. Further assessment of breath-holding time is needed. If b_length <= 20 and length <= 174, the breath-holding time for this breath is reasonable, and the prediction result is acceptable. Otherwise, the prediction result is unacceptable.

[0099] If p_b_std > 0.845, it indicates significant fluctuations in the plateau pressure. Further assessment of the plateau pressure's stability is needed. If p_b_k <= 0.566 and p_b_p_p_half <= 4.075, it indicates a stable plateau pressure during this static mechanical operation, and the prediction result is considered acceptable. Otherwise, the prediction result is unacceptable.

[0100] S3. Determine if there is a qualified static mechanical operation in this waveform record:

[0101] If there is a qualified single breath in set S, then the waveform record is considered to have a qualified static mechanical operation.

[0102] This invention discloses a method for identifying breathing movements involving static mechanical manipulation from ventilator waveforms and determining the compliance of these manipulations. By analyzing waveform morphology, it accurately identifies whether the breath-holding maneuver performed by the doctor is compliant. This method facilitates waveform analysis after breath-holding maneuvers for doctors and reduces the workload of medical staff. When applied to ventilators, it can provide real-time reminders to doctors regarding the compliantness of recently performed breath-holding maneuvers, better guiding the implementation of further mechanical ventilation.

[0103] The waveforms obtained from 3371 breath-holding operations were sampled and evaluated using the method of this invention. The method tested 536 compliant waveforms, 1010 non-compliant waveforms, and 1825 non-static mechanical operations. After final evaluation and verification by medical experts, 575 waveforms were found to be compliant, 972 non-compliant waveforms, and 1824 non-static mechanical operations. The method of this invention has an accuracy rate of 97.75%, demonstrating high sensitivity and specificity.

[0104] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for assessing the compliance of static mechanical operations, characterized in that, include: Obtain the set S of single breaths that perform static mechanical operations in a waveform record, where set S contains at least one single breath; For each breath in set S, perform the following operation: Feature points were extracted from the airway pressure and flow waveform sampling data of a single breath, including the inspiratory start point A, inspiratory end point B, breath-hold start point C, breath-hold end point E, and midpoint of the breath-hold segment D. The inspiratory start point is the first sampling point of the airway pressure and flow waveform, and the inspiratory end point B is the sampling point corresponding to the maximum airway pressure in the airway pressure waveform. The breath-hold start point C, breath-hold end point E, and midpoint of the breath-hold segment D were determined using the following method: The initial start point of the defined segment is the first sampling point in the flow velocity waveform where the flow velocity is less than -0.2 of the previous sampling point and greater than 1 / 2 of the maximum flow velocity. If the minimum flow velocity in the flow velocity waveform is less than -10 L / Min or the last sampling point in the flow velocity waveform is 6 L / Min or more greater than the minimum flow velocity, then the initial end point is the first sampling point before the minimum flow velocity that is less than 1 / 3 of the minimum flow velocity. Otherwise, the initial end point is the sampling point corresponding to the minimum flow velocity. Determine if the restricted segment meets the conditions; otherwise, narrow the range of the restricted segment, redetermine the start and end points of the restricted segment, and repeat the determination until the conditions are met. The specific conditions are: the slope of the fitted line of the restricted segment is less than the set value B_K, the velocity difference between the start and end points of the restricted segment is less than the set value B_DIFF, and the difference between the maximum and minimum values ​​of the velocity in the velocity waveform of the restricted segment is less than the set value B_RANGE. The start and end points of the restricted segment that meet the conditions are the breath-hold start point C and the breath-hold end point E, respectively, and the midpoint between the start and end points of the restricted segment is the breath-hold midpoint D. Based on the extracted feature points, the inhalation segment AB, the breath-holding segment CE, and the second half of the breath-holding segment DE were divided. Features extracted from airway pressure and flow waveform sampling data of a single breath include: the number of sampling points (length) of the single breath waveform, the number of sampling points (b_length) of the breath-hold interval (CE), the proportion (b_rate) of the breath-hold interval (CE) to the entire respiratory cycle, the mean of the flow waveform, the difference between the maximum and mean of the flow waveform in the inspiratory interval (AB) (f_in_sub_max_mean), the mean of the flow waveform in the breath-hold interval (CE) (f_b_mean), the slope (p_b_k) of the fitted line of the pressure waveform in the breath-hold interval (CE), the standard deviation (p_b_std) of the pressure waveform in the breath-hold interval (CE), the slope (p_b_k_half) of the fitted line of the pressure waveform in the second half of the breath-hold interval (DE), the mean of the flow waveform (f_mean), and the difference between the maximum and minimum values ​​of the pressure waveform in the second half of the breath-hold interval (DE) (p_b_p_p_half). Determine whether a single static mechanical operation is compliant based on the extracted features: (1) If b_rate is less than or equal to the threshold B_RATE, then no static mechanical operation was performed on the breath; (2) Otherwise, if one of the following conditions is met, the prediction is unqualified: b_length is less than or equal to the threshold B_LENGTH_1 or f_b_mean is less than or equal to the threshold F_B_MEAN or f_in_sub_max_mean is greater than the threshold F_IN_SUB_MAX_MEAN or f_mean is greater than the threshold F_MEAN or p_b_k_half is greater than the threshold P_B_K_HALF; (3) If neither of the conditions in (1) and (2) above is met, then the standard deviation p_b_std of the pressure waveform in the breath-holding segment CE is further determined: If p_b_std is less than or equal to the threshold P_B_STD, then the following judgment is made: If b_length is less than or equal to the threshold B_LENGTH_2 and length is less than or equal to the threshold LENGTH, then the respiratory static mechanics operation is compliant. Otherwise, the static mechanical operation of breathing is not compliant; If p_b_std is greater than the threshold P_B_STD, further determine the slope p_b_k of the fitted line of the pressure waveform in the breath-holding segment CE and the difference p_b_p_p_half between the maximum and minimum values ​​of the pressure waveform in the second half of the breath-holding segment DE. If p_b_k is less than or equal to the threshold P_B_K and p_b_p_p_half is less than or equal to the threshold P_B_P_P_HALF, then the respiratory static mechanics operation is compliant; otherwise, the respiratory static mechanics operation is non-compliant.

2. The method according to claim 1, characterized in that, The specific steps for obtaining the set S of single breaths undergoing static mechanical operation in a waveform record are as follows: Examine the ventilation pattern of a single breath in the waveform recording to determine whether the breath was a static mechanical operation. The first breath with the ventilation pattern VCV and the previous 5 breaths of that breath constitute a set S.

3. The method according to claim 1, characterized in that, The sampling frequency of the airway pressure and flow rate waveform data for a single breath is above 50 Hz.

4. The method according to claim 1, characterized in that, The process of narrowing the defined segment and redetermining the start and end points of the defined segment specifically involves: Obtain the median of the flow velocity waveform in the current limited segment; compare the difference between the start point, end point and median of the current limited segment, move the end with the larger difference inward by several sampling points to narrow the range of the limited segment, and redetermine the start point and end point of the limited segment.

5. The method according to claim 1, characterized in that, Also includes: If there is a compliant single breath in set S, then the waveform record is considered to have a compliant static mechanical operation.

6. The method according to claim 1, characterized in that, In the feature extraction of airway pressure and flow waveform sampling data based on a single breath, if there is no breath-hold segment in a single breath, features are obtained through a filling strategy. Specifically, the number of sampling points b_length in the breath-hold segment CE is filled to 0; the mean f_b_mean of the flow waveform in the breath-hold segment CE, the standard deviation p_b_std of the pressure waveform in the breath-hold segment CE, the slope p_b_k of the fitted line of the pressure waveform in the breath-hold segment CE, the slope p_b_k_half of the fitted line of the pressure waveform in the second half of the breath-hold segment DE, and the difference p_b_p_p_half between the maximum and minimum values ​​of the pressure waveform in the second half of the breath-hold segment DE are filled to 999.

7. The method according to claim 1, characterized in that, The thresholds are: B_RATE = 0.002, B_LENGTH_1 = 17, F_B_MEAN = -0.5726, F_IN_SUB_MAX_MEAN = 3.5512, F_MEAN = 2.5075, and P_B_K_HALF = 1.5105; P_B_STD = 0.845, B_LENGTH_2 = 20, LENGTH = 174, P_B_K = 0.566, and P_B_P_P_HALF = 4.

075.

8. A static mechanical operation compliance assessment device, characterized in that, include: The data acquisition module is used to acquire a set S of single breaths that perform static mechanical operations in a waveform record, where set S contains at least one single breath. The single-breath compliance assessment module is used to assess the compliance of each breath in set S, including: The feature segmentation unit is used to extract feature points based on the airway pressure and flow rate waveform sampling data of a single breath and to segment the airway into an inspiratory segment AB, a breath-holding segment CE, and a second half of the breath-holding segment DE. The feature points include the inspiratory start point A, the inspiratory end point B, the breath-hold start point C, the breath-hold end point E, and the midpoint of the breath-holding segment D. The inspiratory start point is the first sampling point of the airway pressure and flow rate waveform, and the inspiratory end point B is the sampling point corresponding to the maximum airway pressure value in the airway pressure waveform. The breath-hold start point C, the breath-hold end point E, and the midpoint of the breath-holding segment D are determined using the following method: The initial start point of the defined segment is the first sampling point in the flow velocity waveform where the flow velocity is less than -0.2 of the previous sampling point and greater than 1 / 2 of the maximum flow velocity. If the minimum flow velocity in the flow velocity waveform is less than -10 L / Min or the last sampling point in the flow velocity waveform is 6 L / Min or more greater than the minimum flow velocity, then the initial end point is the first sampling point before the minimum flow velocity that is less than 1 / 3 of the minimum flow velocity. Otherwise, the initial end point is the sampling point corresponding to the minimum flow velocity. Determine if the restricted segment meets the conditions; otherwise, narrow the range of the restricted segment, redetermine the start and end points of the restricted segment, and repeat the determination until the conditions are met. The specific conditions are: the slope of the fitted line of the restricted segment is less than the set value B_K, the velocity difference between the start and end points of the restricted segment is less than the set value B_DIFF, and the difference between the maximum and minimum values ​​of the velocity in the velocity waveform of the restricted segment is less than the set value B_RANGE. The start and end points of the restricted segment that meet the conditions are the breath-hold start point C and the breath-hold end point E, respectively, and the midpoint between the start and end points of the restricted segment is the breath-hold midpoint D. The feature extraction unit is used to extract features based on the airway pressure and flow waveform sampling data of a single breath, including: the number of sampling points of the single breath waveform (length), the number of sampling points of the breath-hold interval (CE) (b_length), the proportion of the breath-hold interval (CE) to the entire respiratory cycle (b_rate), the mean of the flow waveform, the difference between the maximum and mean of the flow waveform in the inspiratory interval (AB) (f_in_sub_max_mean), the mean of the flow waveform in the breath-hold interval (CE) (f_b_mean), the slope of the fitted line of the pressure waveform in the breath-hold interval (CE) (p_b_k), the standard deviation of the pressure waveform in the breath-hold interval (CE) (p_b_std), the slope of the fitted line of the pressure waveform in the second half of the breath-hold interval (DE) (p_b_k_half), the mean of the flow waveform (f_mean), and the difference between the maximum and minimum values ​​of the pressure waveform in the second half of the breath-hold interval (DE) (p_b_p_p_half). The compliance assessment unit is used to determine whether a single static mechanical operation is compliant based on the extracted features. (1) If b_rate is less than or equal to the threshold B_RATE, then no static mechanical operation was performed on the breath; (2) Otherwise, if one of the following conditions is met, the prediction is unqualified: b_length is less than or equal to the threshold B_LENGTH_1 or fb mean is less than or equal to the threshold FB MEAN or f in sub max mean is greater than the threshold F_IN_SUB_MAX_MEAN or f_mean is greater than the threshold F_MEAN or p_b_k_half is greater than the threshold P_B_K_HALF. (3) If neither of the conditions in (1) and (2) above is met, then the standard deviation p_b_std of the pressure waveform in the breath-holding segment CE is further determined: If p_b_std is less than or equal to the threshold P_B_STD, then the following judgment is made: If b_length is less than or equal to the threshold B_LENGTH_2 and length is less than or equal to the threshold LENGTH, then the respiratory static mechanics operation is compliant. Otherwise, the static mechanical operation of breathing is not compliant; If p_b_std is greater than the threshold P_B_STD, further determine the slope p_b_k of the fitted line of the pressure waveform in the breath-holding segment CE and the difference p_b_p_p_half between the maximum and minimum values ​​of the pressure waveform in the second half of the breath-holding segment DE. If p_b_k is less than or equal to the threshold P_B_K and p_b_p_p_half is less than or equal to the threshold P_B_P_P_HALF, then the respiratory static mechanics operation is compliant; otherwise, the respiratory static mechanics operation is non-compliant.

9. The apparatus according to claim 8, characterized in that, Also includes: The comprehensive compliance judgment module is used to determine whether there is a compliant static mechanical operation in the waveform record based on the set S: if there is a compliant single breath in the set S, then the waveform record is considered to have a compliant static mechanical operation.

Images