A control method, device and medium based on a respiratory rehabilitation device

Abstract

The application discloses a control method and device based on a respiratory rehabilitation device and a medium, relates to the technical field of medical devices, and comprises the following steps: collecting a respiratory-related physiological signal and a stimulation coupling signal, constructing a basic respiratory phase sequence, a site phase response baseline and a reference airflow curve library, and determining a fixed charge budget training prescription; performing current respiratory cycle stimulation control according to the training prescription, and calculating an effective auxiliary area in a target stimulation phase window, a phase coordination debt increment, a response time delay offset and a contact offset; performing safety gating based on the contact offset and a recovery lag, and outputting a unique control action in combination with the phase coordination debt increment and a site recession; writing next respiratory cycle control parameters according to the unique control action, and performing low-disturbance pre-detection during recovery isolation. The application improves individualized control adaptability, effective assistance degree, closed-loop regulation accuracy and training stability.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to a control method, device and medium based on respiratory rehabilitation equipment. Background Technology

[0002] With the development of respiratory rehabilitation training, functional electrical stimulation, and intelligent medical devices, control methods for respiratory rehabilitation equipment are gradually evolving from fixed parameter output to individualized closed-loop regulation. Conventional methods typically identify the trainee's current ventilation status, respiratory rhythm, and recovery status by collecting respiratory-related physiological signals such as airflow, pressure, and blood oxygenation, and then set stimulation intensity, stimulation sequence, or training load accordingly. Some schemes also incorporate reference respiratory curves, rhythm guidance, and safety thresholds to implement phased adjustments to the training process, thereby improving the relevance and safety of respiratory training. These methods have provided a technical foundation for the application of respiratory rehabilitation equipment in scenarios such as chronic respiratory dysfunction, postoperative rehabilitation, and neuromuscular function training.

[0003] However, existing methods still have two main limitations: First, existing control methods mainly rely on fixed rhythm matching or single feedback correction, which are insufficient in jointly representing differences in stimulus site response, phase shift, and changes in contact state, making it difficult to maintain effective auxiliary consistency within the target stimulus phase window during continuous training; Second, existing methods mostly use single downgrading or shutdown to handle abnormal states during training, lacking a mechanism to separate the dominant causes of phase shift, site decay, and contact abnormalities, resulting in insufficient targeting of closed-loop regulation and insufficient training continuity. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a control method based on respiratory rehabilitation equipment to solve the problems of insufficient effective auxiliary stability within the target stimulus phase window and insufficient targeted closed-loop regulation under abnormal conditions in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a control method based on a respiratory rehabilitation device, comprising: collecting respiratory-related physiological signals and stimulation coupling signals from the trainee; constructing a basic respiratory phase sequence, a site phase response baseline, and a reference airflow curve library by combining low-intensity probe stimulation; determining a fixed charge budget training prescription for the current training phase based on the basic respiratory phase sequence, the site phase response baseline, and the recovery safety baseline; executing stimulation control of the current respiratory cycle according to the fixed charge budget training prescription, and calculating the effective auxiliary area within the target stimulation phase window based on the real-time airflow curve and the reference airflow curve; calculating the phase coordination debt increment, response delay offset, and contact offset by the effective auxiliary area, the target auxiliary area, the reference response delay, and the reference contact impedance; performing a safety gating judgment based on the contact offset and the recovery hysteresis, and, when the conditions for continued training are met, performing dominant cause separation by combining the phase coordination debt increment and the site decay, and outputting a unique control action; writing the control parameters for the next respiratory cycle according to the unique control action, and performing low-disturbance pre-probe during recovery isolation.

[0007] As a preferred embodiment of the control method based on respiratory rehabilitation equipment described in this invention, the specific steps for constructing the basic respiratory phase sequence, the site phase response baseline, and the reference airflow curve library are as follows: Time-scale alignment and periodic segmentation are performed on respiratory-related physiological signals and stimulation coupling signals to obtain the basic respiratory phase sequence; time normalization is performed on the steady-state respiratory cycle to generate the reference airflow curve library; low-intensity probe stimulation is applied to each stimulation site under different respiratory phase windows to obtain the unit probe charge auxiliary benefit, reference response delay, and reference contact impedance, thereby constructing the site phase response baseline.

[0008] As a preferred embodiment of the control method based on respiratory rehabilitation equipment described in this invention, the step of determining the fixed charge budget training prescription for the current training phase includes: screening effective stimulation site phase window combinations based on the site phase response baseline, determining the target stimulation phase window and the initial stimulation site group; determining the initial stimulation advance amount based on the target stimulation phase window and the reference response delay; determining the total periodic stimulation charge based on the recovery safety baseline; and generating a target auxiliary area sequence according to the proportional relationship between the total periodic stimulation charge and the probe charge.

[0009] As a preferred embodiment of the control method based on respiratory rehabilitation equipment described in this invention, the calculation of the effective auxiliary area within the target stimulus phase window includes: at the beginning of the current respiratory cycle, determining the starting point of the current respiratory cycle based on the zero-crossing moment when the airflow signal transitions from expiration to inspiration; determining the stimulus triggering moment of this cycle based on the starting moment of the target stimulus phase window and the stimulus advance, and outputting a stimulus pulse sequence to the starting stimulus site group when the stimulus triggering moment of this cycle is reached; acquiring the real-time airflow curve of the current respiratory cycle and reading the corresponding reference airflow curve from the reference airflow curve library; using the starting moment and ending moment of the target stimulus phase window as statistical boundaries, integrating the positive difference between the real-time airflow curve and the reference airflow curve of the current respiratory cycle within the target stimulus phase window to obtain the effective auxiliary area within the target stimulus phase window.

[0010] As a preferred embodiment of the control method based on respiratory rehabilitation equipment described in this invention, the fixed charge budget training prescription includes: phase allowable boundary, contact allowable boundary, site decay boundary, recovery allowable boundary, migration trigger boundary, and candidate site cache validity period.

[0011] As a preferred embodiment of the control method based on respiratory rehabilitation equipment described in this invention, the calculation of phase coordination debt increment, response delay offset, and contact offset includes: determining the moment when the airflow growth rate first reaches a local peak after stimulation triggering based on the real-time airflow curve; determining the time difference between the moment when the airflow growth rate first reaches a local peak and the moment of stimulation triggering as the actual response delay of the current respiratory cycle; normalizing the difference between the actual response delay and the reference response delay of the current respiratory cycle relative to the duration of the current respiratory cycle to obtain the response delay offset; and stabilizing the contact impedance curve of the current respiratory cycle before stimulation triggering. The average value within a fixed sampling segment is calculated and determined as the real-time contact impedance for the current respiratory cycle. The difference between the real-time contact impedance for the current respiratory cycle and the reference contact impedance is normalized relative to the reference contact impedance to obtain the contact offset. The auxiliary benefit per unit stimulus charge for the current respiratory cycle is determined based on the ratio between the effective auxiliary area within the target stimulus phase window and the total stimulus charge for the current respiratory cycle. The site decay is obtained by combining the auxiliary benefit per unit probe charge of the reference window. Based on the difference between the target auxiliary area and the effective auxiliary area, combined with the response delay offset and the contact offset, the phase coordination debt increment for the current respiratory cycle is calculated.

[0012] As a preferred embodiment of the control method based on respiratory rehabilitation equipment described in this invention, the output unique control action includes: when the contact offset exceeds the contact allowable boundary, or the recovery hysteresis exceeds the recovery allowable boundary, outputting recovery isolation as the unique control action; when the effective auxiliary area reaches the target auxiliary area, or the phase coordination debt increment is zero, outputting maintain current prescription as the unique control action; when the effective auxiliary area does not reach the target auxiliary area and the phase coordination debt increment is greater than zero, performing dominant cause separation based on response delay offset and site decay.

[0013] As a preferred embodiment of the control method based on respiratory rehabilitation equipment according to the present invention, the process of separating the dominant cause includes: when the response delay offset is greater than the phase allowable boundary and the site decay is not greater than the site decay boundary, the dominant cause is determined to be phase offset, and phase correction is output as the only control action; when the response delay offset is greater than the phase allowable boundary and the site decay is greater than the site decay boundary, or when the response delay offset is not greater than the phase allowable boundary and the site decay is in the boundary proximity zone, the dominant cause is determined to be mixed offset, and double-series pulse rearrangement is output as the only control action; when the response delay offset is not greater than the phase allowable boundary and the site decay is greater than the site decay boundary, the dominant cause is determined to be site decay, and the candidate stimulus site group is judged to have substitution value based on the candidate site response parameter set cache; when the candidate stimulus site group has substitution value, site migration is output as the only control action, otherwise restoration isolation is output as the only control action.

[0014] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the control method based on a respiratory rehabilitation device as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any step of the control method based on a respiratory rehabilitation device as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: by constructing a basic respiratory phase sequence, a site phase response baseline, and a reference airflow curve library, the adaptability of individualized control is improved; by determining a fixed charge budget training prescription and executing stimulus control within the target stimulus phase window, the effectiveness of assistance is improved while ensuring training safety; by calculating the phase co-debt increment and outputting a unique control action, the accuracy of closed-loop regulation and training stability are improved. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. 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.

[0018] Figure 1 This is a flowchart of a control method based on respiratory rehabilitation equipment.

[0019] Figure 2 A flowchart for constructing a reference airflow curve library.

[0020] Figure 3 A flowchart for generating a training prescription for a fixed charge budget.

[0021] Figure 4 This is a flowchart for the unique control action output and control write-back.

[0022] Figure 5 A comparative data graph showing the effective auxiliary area within the phase window of the target stimulus.

[0023] Figure 6 A comparative data graph showing contact offset, recovery hysteresis, and corresponding safety boundaries. Detailed Implementation

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0026] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0027] Reference Figures 1-6 As one embodiment of the present invention, this embodiment provides a control method based on a respiratory rehabilitation device, comprising the following steps: S1. Collect respiratory-related physiological signals and stimulus coupling signals from trainees, and construct a basic respiratory phase sequence, site phase response baseline, and reference airflow curve library by combining low-intensity probe stimuli.

[0028] The airflow sensor of the respiratory rehabilitation device is placed at the inlet of the breathing tubing, the airway pressure sensor is placed at the connection between the mask and the airway, the chest displacement sensor and the abdominal displacement sensor are respectively attached to the lower edge of the sternum and the periumbilical region, the blood oxygen sensor is clipped to the fingertips of the trainee, and the heart rate acquisition electrode is attached to the precordial lead position.

[0029] Simultaneously, stimulation electrodes are attached to the corresponding stimulation sites on the body surface of the respiratory-related muscle groups.

[0030] It should be noted that the respiratory-related muscle groups include at least the corresponding sites of the inspiratory accessory muscle groups and the expiratory accessory muscle groups.

[0031] Using the sampling clock of the airflow sensor as a unified master clock, the airflow signal, airway pressure signal, chest displacement signal, abdominal displacement signal, blood oxygen saturation signal, heart rate signal, and electrode contact impedance signal are uniformly time-aligned to obtain raw respiratory monitoring data.

[0032] Furthermore, the raw respiratory monitoring data is preprocessed.

[0033] Specifically, baseline drift removal and smoothing filtering are performed on the airflow signal, high-frequency noise suppression is performed on the airway pressure signal, amplitude normalization and baseline correction are performed on the chest wall displacement signal and abdominal displacement signal, and abnormal jump rejection is performed on the blood oxygen saturation signal and heart rate signal. The zero-crossing time of the airflow signal from expiration to inspiration is determined as the inspiration start time, and the zero-crossing time of the airflow signal from inspiration to expiration is determined as the expiration start time. Combined with the same direction change relationship of the chest wall displacement signal and abdominal displacement signal, the consistency of the inspiration start time and the expiration start time is checked. Abnormal cycles that have a relative change of airflow peak value relative to the corresponding peak value of the previous complete respiratory cycle that is greater than the body movement interference judgment boundary, or where the chest wall displacement signal and abdominal displacement signal show opposite changes in the same respiratory cycle, are deleted to obtain a continuous complete respiratory cycle sequence.

[0034] It should be noted that the boundary for determining body movement interference is obtained by continuously collecting multiple complete respiratory cycles during the trainee's resting guided breathing phase, reading the relative change between the airflow peaks of adjacent complete respiratory cycles, and taking the maximum measured value. The preferred value range is [15%, 35%].

[0035] Based on a continuous complete respiratory cycle sequence, the tidal volume, inspiratory duration, expiratory duration, peak inspiratory flow rate, peak expiratory flow rate, chest-abdominal phase difference, and airway pressure fluctuation amplitude corresponding to each respiratory cycle are extracted to obtain a set of core respiratory mechanics parameters.

[0036] Steady-state screening is performed on continuous complete respiratory cycle sequences based on whether the variation amplitudes of cycle duration, tidal volume, and chest-abdomen phase difference of adjacent respiratory cycles all fall within a stable variation bandwidth, thus obtaining a set of steady-state respiratory cycles.

[0037] It should be noted that the stable variation bandwidth is obtained by continuously collecting multiple complete respiratory cycles during the resting guided breathing phase of the trainee, deleting abnormal respiratory cycles, and reading the maximum measured values ​​of the relative change rate of adjacent cycle duration, the relative change rate of tidal volume, and the absolute change of chest-abdomen phase difference in the remaining respiratory cycles. Preferably, the stable variation bandwidth of the relative change rate of cycle duration is [3%, 10%], the stable variation bandwidth of the relative change rate of tidal volume is [5%, 15%], and the stable variation bandwidth of the absolute change of chest-abdomen phase difference is [4°, 15°].

[0038] The steady-state respiratory cycle set is time-normalized according to the respiratory cycle duration, and the median value of the corresponding airflow value is read at the same normalized phase position to generate an initial reference airflow curve library.

[0039] Within the set of steady-state respiratory cycles, the recovery time required for blood oxygen saturation to return to the steady-state fluctuation band formed by the average blood oxygen saturation value corresponding to the set of steady-state respiratory cycles and the recovery tolerance of fluctuation above and below the average blood oxygen saturation value is calculated, as well as the recovery time required for heart rate to return to the steady-state fluctuation band formed by the average heart rate value corresponding to the set of steady-state respiratory cycles and the recovery tolerance of fluctuation above and below the average heart rate value is calculated, and a safe recovery baseline is obtained.

[0040] It should be noted that the recovery tolerance is obtained by reading the offset of blood oxygen saturation and heart rate relative to their respective average values ​​within the steady-state respiratory cycle set, and taking the maximum measured value of the offset. The preferred range of the recovery tolerance corresponding to blood oxygen saturation is [1%, 3%], and the preferred range of the recovery tolerance corresponding to heart rate is [3%, 8%].

[0041] Furthermore, low-intensity probe stimuli are applied sequentially to each stimulation site.

[0042] It should be noted that the low-intensity probe stimulus is triggered sequentially in the order of the inspiratory initiation phase window, the inspiratory mid-phase phase window, and the expiratory initiation phase window, and the probe pulse width, probe frequency, and number of probe pulses are kept consistent within each phase window.

[0043] For each stimulation site, the airflow response curve and contact impedance change curve were acquired after detection within each phase window, and the auxiliary incremental area of ​​the detection window was calculated. The expression is as follows: ; in, Indicates the first The stimulation site at the ... The incremental area of ​​the detector window under each phase window Indicates the first The start time of each phase window Indicates the first The end time of each phase window Indicates the first The stimulation site at the ... Airflow response curves after applying low-intensity probe stimulation under each phase window This indicates the reference airflow curve called from the initial reference airflow curve library. Indicates when When the value is negative, it is set to zero, which is used to retain only the positive auxiliary part brought about by the detection stimulus.

[0044] Based on the incremental area of ​​the probe window and the amount of probe charge corresponding to the probe stimulus, the auxiliary benefit per unit probe charge is calculated, expressed as follows: ; in, Indicates the first The stimulation site at the ... The unit probe charge auxiliary gain under each phase window Indicates the first The stimulation site at the ... The amount of probe charge corresponding to each phase window.

[0045] Simultaneously, the time difference between the stimulus initiation moment and the moment when the airflow growth rate first reaches its local peak is recorded to obtain the reference response delay.

[0046] Read the stable values ​​of contact impedance before and after the detection stimulus is applied, and obtain the reference contact impedance.

[0047] The site phase response baseline is constructed by combining the unit probe charge assist gain, reference response delay, and reference contact impedance for each stimulation site under each phase window.

[0048] Furthermore, the stimulation sites were initially sorted from largest to smallest based on the auxiliary benefit of the unit detection charge.

[0049] When the difference in the unit probe charge assist gain between two adjacent stimulation sites does not exceed the gain equivalence band, they are reordered in ascending order of reference response delay.

[0050] When the reference response delay is still the same, the candidate sites are reordered in ascending order of reference contact impedance offset to obtain the initial value of the candidate site order.

[0051] It should be noted that the reference contact impedance offset was obtained by repeatedly performing low-intensity probing stimulation at the same stimulation site, the same respiratory phase window, and the same attachment state, and by reading the relative offset of the reference contact impedance relative to the corresponding average value for each instance.

[0052] It should be noted that the benefit equivalence band is obtained by repeatedly performing low-intensity probe stimulation under the same stimulation site, the same target stimulation phase window, and the same probe output conditions, reading the difference between the auxiliary benefits of each unit probe charge and taking the maximum difference. Preferably, the value range of the benefit equivalence band is [3%, 12%].

[0053] S2. Based on the basic respiratory phase sequence, site phase response baseline, and recovery safety baseline, determine the fixed charge budget training prescription for the current training phase.

[0054] Each complete respiratory cycle in the basic respiratory phase sequence is divided into several (e.g., 3) continuous phase windows according to a uniform normalized time axis.

[0055] The unit probe charge assist gain, reference response delay, and reference contact impedance in each phase window and site phase response baseline are mapped one-to-one.

[0056] Delete stimulation site phase window combinations where the reference response delay is greater than the corresponding phase window length or the reference contact impedance deviates from the contact stability band, and obtain effective stimulation site phase window combinations.

[0057] It should be noted that the contact stabilization band is obtained by repeatedly attaching the band to the same stimulation site and collecting multiple contact impedance stabilization values, reading the relative offset between adjacent sampled stabilization values ​​and taking the maximum measured value. Preferably, the range of the contact stabilization band is [5%, 18%].

[0058] Furthermore, the effective stimulation site phase window combinations are sorted in descending order of the auxiliary benefit per unit probe charge.

[0059] When there are two or more effective stimulation site phase window combinations with the same unit probe charge assist benefit, the effective stimulation site phase window combination with the smallest reference response delay value is selected.

[0060] When the reference response delay values ​​are still the same, select the effective stimulus site phase window combination with the smallest initial value of the candidate site sequence to obtain the target stimulus phase window and the initial stimulus site group.

[0061] Furthermore, based on the start time of the target stimulus phase window and the reference response delay corresponding to the initial stimulus site group, an initial stimulus lead is generated.

[0062] It should be noted that the initial stimulus advance is obtained by reading the reference response delay corresponding to the starting stimulus site group under the target stimulus phase window, and directly mapping the reference response delay to the advance duration of the stimulus trigger relative to the start time of the target stimulus phase window. Preferably, the value range of the initial stimulus advance is [40ms, 220ms].

[0063] Using the initial stimulation site group as the target, candidate cycle stimulation outputs are applied in ascending order during a continuous steady-state respiratory cycle, and the blood oxygen recovery time and heart rate recovery time corresponding to each candidate cycle stimulation output are read.

[0064] When the blood oxygen recovery time or heart rate recovery time corresponding to a candidate periodic stimulus output exceeds the safe recovery baseline, the previous level of the candidate periodic stimulus output is determined as the total periodic stimulus charge, and the total periodic stimulus charge is written into the fixed charge budget training prescription.

[0065] To convert the total periodic stimulus charge into a usable target auxiliary area sequence, the probe charge and probe window auxiliary increment area corresponding to the initial stimulus site group under the target stimulus phase window are read. Under the condition that the initial stimulus site group, target stimulus phase window, and current training phase remain unchanged, the target auxiliary area corresponding to each respiratory cycle in the current training phase is obtained by calculating the ratio between the total periodic stimulus charge and the probe charge, as expressed by: ; in, Indicates the first The target auxiliary area corresponding to each training breathing cycle Indicates the first The total cyclic stimulation charge corresponding to each training breathing cycle This represents the amount of probe charge corresponding to the initial stimulation site group within the target stimulation phase window. This represents the area of ​​the probe window auxiliary increment corresponding to the initial stimulation site group under the target stimulation phase window.

[0066] It should be noted that the target auxiliary area represents the positive airflow increment brought about by the stimulus within the target stimulus phase window. Under the condition that the initial stimulus site group, the target stimulus phase window, and the current training phase remain unchanged, the stimulus path, the phase interval, and the trainee's current fatigue level do not change, and the correspondence between the probe window auxiliary increment area and the stimulus charge remains under the same conditions. Therefore, by taking the probe charge corresponding to the initial stimulus site group under the target stimulus phase window as the baseline charge and the probe window auxiliary increment area as the baseline auxiliary area, the target auxiliary area corresponding to the current training breathing cycle can be obtained by mapping the ratio of the total stimulus charge of the current training breathing cycle to the baseline charge.

[0067] The target auxiliary area sequence is obtained by arranging the target auxiliary areas corresponding to each training breathing cycle in chronological order.

[0068] Furthermore, by reading the reference response delays obtained from repeated detections at the same stimulus site under the same target stimulus phase window, the offset of each reference response delay relative to the corresponding average value is calculated, and the maximum measured value is taken to obtain the phase allowable boundary.

[0069] By reading the reference contact impedance obtained by repeated detection at the same stimulation site, calculating the offset of each reference contact impedance relative to the corresponding average value, and taking the maximum measured value, the contact allowable boundary is obtained.

[0070] By reading the unit probe charge boost obtained from repeated detections at the same stimulus site, calculating the relative decrease in unit probe charge boost between two adjacent detections, and taking the maximum measured value, the site decay boundary is obtained.

[0071] By reading the blood oxygen recovery time and heart rate recovery time from the recovery safety baseline, the offset of the recovery time of each steady-state respiratory cycle relative to the corresponding recovery safety baseline is calculated, and the maximum measured value is taken to obtain the recovery allowable boundary.

[0072] The migration trigger boundary is obtained by reading the unit probe charge auxiliary benefit corresponding to adjacent candidate sites in the initial value of the candidate site sequence, calculating the relative difference between the unit probe charge auxiliary benefits of adjacent candidate sites, and taking the minimum measured value.

[0073] The validity period of the candidate site cache is obtained by repeatedly performing low-intensity probing stimulation under the same attachment state and recording the number of consecutive respiratory cycles in which the initial value of the candidate site order remains unchanged.

[0074] The target stimulus phase window, the initial stimulus site group, the initial stimulus lead, the total periodic stimulus charge, the target auxiliary area sequence, the phase allowable boundary, the contact allowable boundary, the site decay boundary, the recovery allowable boundary, the migration trigger boundary, the site migration order, and the candidate site cache validity period are combined to generate a fixed charge budget training prescription.

[0075] It should be noted that the site migration order is the sequential replacement order of the candidate stimulus site groups determined based on the initial value of the candidate site order.

[0076] S3. Execute the current respiratory cycle stimulation control according to the fixed charge budget training prescription, and calculate the effective auxiliary area within the target stimulation phase window based on the real-time airflow curve and the reference airflow curve. Calculate the phase coordination debt increment, response delay offset, and contact offset using the effective auxiliary area, target auxiliary area, reference response delay, and reference contact impedance.

[0077] At the start of the current respiratory cycle, the starting point of the current respiratory cycle is determined by the zero-crossing moment when the airflow signal changes from expiration to inspiration, and the stimulation triggering time of this cycle is determined according to the start time of the target stimulus phase window and the initial stimulus advance.

[0078] When the stimulation trigger time of this cycle is reached, the stimulation pulse sequence corresponding to the total stimulation charge of the cycle is output to the starting stimulation site group, and the real-time airflow curve, chest displacement curve, abdominal displacement curve, blood oxygen saturation curve, heart rate curve and contact impedance curve of the current respiratory cycle are collected simultaneously to obtain real-time monitoring data of the current respiratory cycle.

[0079] Furthermore, the reference airflow curve corresponding to the current respiratory cycle duration is read from the initial reference airflow curve library.

[0080] If there is no reference airflow curve in the initial reference airflow curve library that is exactly the same as the duration of the current respiratory cycle, select the reference airflow curves corresponding to the durations of two adjacent cycles, and perform linear interpolation according to the duration of the current respiratory cycle to generate the reference airflow curve for the current respiratory cycle.

[0081] Using the start and end times of the target stimulus phase window as the integration boundaries, the positive difference between the real-time airflow curve and the reference airflow curve of the current respiratory cycle is integrated to obtain the effective auxiliary area within the target stimulus phase window. The expression is as follows: ; in, Indicates the first Effective auxiliary area within the target stimulus phase window per respiratory cycle Indicates the first The start time of the target stimulus phase window for each respiratory cycle. Indicates the first The end time of the target stimulus phase window in each respiratory cycle Indicates the first Real-time airflow curve for each respiratory cycle Indicates the first Reference airflow curve for one respiratory cycle.

[0082] In this embodiment, to verify the effect of the method of the present invention on improving the effective assistance level within the target stimulus phase window and maintaining the stability of the auxiliary output under perturbation conditions, a comparison of the effective assistance area within the target stimulus phase window under different control methods was conducted; such as Figure 5Control group 2 uses a fixed stimulus advance, fixed stimulus site group, and fixed total stimulus charge per cycle; control group 1 uses a control method that only corrects the stimulus advance. As shown in the figure, the effective auxiliary area within the target stimulus phase window of the experimental group is generally higher than that of control group 1 and control group 2, and is generally closer to the target auxiliary area curve of the experimental group. This indicates that the fixed charge budget training prescription generated based on the basic respiratory phase sequence, the site phase response baseline, and the reference airflow curve library can make the stimulus effect more concentrated within the target stimulus phase window. The magnified view further shows that in the perturbation interval, the effective auxiliary area within the target stimulus phase window of control group 1 and control group 2 decreases more significantly, while the experimental group still maintains a higher level. This indicates that the present invention can still maintain a good auxiliary output through phase correction, double-pulse rearrangement, or site migration when respiratory rhythm changes or state perturbations occur, thereby improving the effective auxiliary level and enhancing the continuity and stability of the training process.

[0083] The actual response delay of the current respiratory cycle is determined by the time difference between the moment when the airflow growth rate first reaches a local peak after the stimulus triggering moment and the moment when the stimulus triggering moment, using the real-time airflow curve.

[0084] The actual response delay of the current respiratory cycle is compared with the reference response delay, and the difference between the actual response delay and the reference response delay is normalized relative to the duration of the current respiratory cycle to obtain the response delay offset.

[0085] Read the reference contact impedance corresponding to the initial stimulation site group under the target stimulation phase window.

[0086] The average value of the contact impedance curve of the current respiratory cycle within a preset stable sampling segment before stimulation is obtained, and the average value is determined as the real-time contact impedance of the current respiratory cycle.

[0087] It should be noted that the stable sampling segment is obtained by continuously reading multiple contact impedance sampling values ​​before the stimulus is triggered, and selecting a continuous sampling interval in which the direction of contact impedance change remains consistent. Preferably, the duration of the stable sampling segment is in the range of [20ms, 80ms].

[0088] The real-time contact impedance of the current respiratory cycle is compared with the reference contact impedance, and the difference between the real-time contact impedance and the reference contact impedance is normalized relative to the reference contact impedance to obtain the contact offset.

[0089] Furthermore, the ratio of the effective auxiliary area within the target stimulus phase window to the total cyclic stimulus charge of the current respiratory cycle is used as the auxiliary benefit per unit stimulus charge in the current respiratory cycle.

[0090] The reference unit probe charge assist gain of the initial stimulation site group under the target stimulation phase window is read, and the difference between the reference unit probe charge assist gain and the current respiratory cycle unit stimulation charge assist gain is normalized to obtain the site decay amount.

[0091] Read the reference blood oxygen recovery time and reference heart rate recovery time from the safe baseline.

[0092] At the end of the current respiratory cycle, continue to track the blood oxygen saturation curve and heart rate curve along the time axis to determine the moment when blood oxygen saturation returns to the steady-state fluctuation zone and the moment when heart rate returns to the steady-state fluctuation zone, and determine the corresponding time difference as the current blood oxygen recovery time and the current heart rate recovery time, respectively.

[0093] The deviation of the current blood oxygen recovery time from the reference blood oxygen recovery time and the deviation of the current heart rate recovery time from the reference heart rate recovery time are compared, and the one with the largest deviation is determined as the recovery lag.

[0094] It should be noted that the current blood oxygen recovery time and the current heart rate recovery time are obtained by continuing to track the blood oxygen saturation curve and heart rate curve after the end of the current respiratory cycle, and reading the time taken for the blood oxygen saturation and heart rate to return to the corresponding steady state zone. Preferably, the value range of the current blood oxygen recovery time is [1s, 12s], and the value range of the current heart rate recovery time is [1s, 15s].

[0095] Read the target supplementary area corresponding to the current respiratory cycle from the target supplementary area sequence, and calculate the phase-coordinated debt increment of the current respiratory cycle, expressed as: ; in, Indicates the first Phase-coordinated debt increment per breathing cycle Indicates the first The target auxiliary area corresponding to each respiratory cycle Indicates the first The response delay offset per respiratory cycle Indicates the first Contact offset per respiratory cycle Indicates when Take zero when the value is negative.

[0096] Furthermore, the first The effective auxiliary area, response delay offset, contact offset, site decay, recovery hysteresis, and phase coordination debt increment within the target stimulus phase window of each respiratory cycle are sequentially combined according to a unified cycle number to obtain the current respiratory cycle state quantity set.

[0097] It should be noted that the actual response delay is obtained by reading the first derivative of the real-time airflow curve after the stimulus triggering time and the time difference between the moment when the first derivative first reaches a local peak and the moment when the stimulus triggering time. Preferably, the value range of the actual response delay is [40ms, 260ms].

[0098] S4. Based on the contact offset and recovery hysteresis, a safety gating judgment is made. When the conditions for continued training are met, the dominant cause is separated by combining the phase collaborative debt increment and the site decay, and a unique control action is output.

[0099] Read the candidate site response parameter set cache generated in the historical control cycle and determine whether the candidate site response parameter set cache is in a valid state.

[0100] When the number of interval cycles between the generation time of the candidate site response parameter set cache and the current respiratory cycle number is not greater than the validity period of the candidate site cache, the candidate site response parameter set cache is determined to be a valid cache; when the number of interval cycles is greater than the validity period of the candidate site cache, the candidate site response parameter set cache is determined to be an invalid cache.

[0101] It should be noted that the candidate site response parameter set cache is historical cache data generated and saved after the low-disturbance pre-probe was performed in the previous recovery isolation phase. When the candidate site response parameter set cache has not been generated in the current training phase, the candidate site response parameter set cache is determined to be invalid cache.

[0102] Further, it involves entering a security gate control decision.

[0103] Specifically, the contact offset of the current respiratory cycle is compared with the contact allowable boundary, and the recovery hysteresis of the current respiratory cycle is compared with the recovery allowable boundary.

[0104] When the contact offset is greater than the contact allowable boundary, or the recovery hysteresis is greater than the recovery allowable boundary, it is determined that the current respiratory cycle does not meet the conditions for continued enhanced stimulation, and the recovery isolation is output as the only control action.

[0105] Simultaneously, the total stimulating charge of the initial stimulation site group in the next respiratory cycle is corrected to the product of the total stimulating charge of the current cycle and the recovery deload coefficient. The stimulation advance amount of the next respiratory cycle is kept at the stimulation advance amount of the current cycle, and the candidate site refresh request is set to the triggered state to obtain the parameter correction amount corresponding to the recovery isolation.

[0106] It should be noted that the recovery deload factor is obtained by comparing the blood oxygen recovery time and heart rate recovery time corresponding to the stimulation output of multiple consecutive candidate cycles during the prescription familiarization phase, and reading the relative charge reduction ratio corresponding to the simultaneous return of blood oxygen recovery time and heart rate recovery time to within the recovery safety baseline. Preferably, the value range of the recovery deload factor is [0.60, 0.90].

[0107] When the contact offset is not greater than the contact allowable boundary and the recovery hysteresis is not greater than the recovery allowable boundary, the collaborative debt activation determination stage begins.

[0108] Furthermore, the target assist area and the effective assist area within the target stimulus phase window corresponding to the current respiratory cycle are read to determine whether there is an assist gap in the current respiratory cycle.

[0109] When the effective auxiliary area within the target stimulus phase window is not less than the target auxiliary area, or when the phase coordination debt increment is zero, it is determined that coordination debt control has not been activated in the current respiratory cycle, and the output maintains the current prescription as the only control action.

[0110] Meanwhile, the stimulation site group, stimulation advance amount, and total stimulation charge of the next respiratory cycle are maintained as the parameters corresponding to the current cycle, and the candidate site refresh request is set to non-triggered state to obtain the parameter correction amount corresponding to the current prescription.

[0111] Furthermore, when the effective auxiliary area within the target stimulus phase window is smaller than the target auxiliary area and the phase coordination debt increment is greater than zero, the process enters the dominant cause separation stage.

[0112] The response delay offset is compared with the phase allowable boundary, and the site decay is compared with the site decay boundary.

[0113] When the response delay offset is greater than the phase allowable boundary and the site decay is not greater than the site decay boundary, the dominant cause of the current respiratory cycle is determined to be the phase offset, and the output phase correction is used as the only control action.

[0114] The actual response delay and reference response delay of the current respiratory cycle are read, and the time difference between the actual response delay and the reference response delay is used as the stimulus advance correction value. When the actual response delay is greater than the reference response delay, the stimulus advance is corrected forward according to the corresponding time difference in the next respiratory cycle. When the actual response delay is less than the reference response delay, the stimulus advance is corrected backward according to the corresponding time difference in the next respiratory cycle. At the same time, the stimulation site group and the total stimulation charge of the cycle are kept unchanged in the next respiratory cycle, and the candidate site refresh request is set to non-triggered state. The parameter correction amount corresponding to the phase correction is obtained.

[0115] When the response delay offset is greater than the phase allowable boundary and the site decay is greater than the site decay boundary, or when the response delay offset is not greater than the phase allowable boundary and the absolute value of the difference between the site decay and the site decay boundary is not greater than the boundary proximity zone corresponding to the site decay boundary, the dominant cause of the current respiratory cycle is determined to be mixed offset, and the dual-series pulse rearrangement is output as the sole control action.

[0116] It should be noted that the boundary proximity band is obtained by reading the site decay amount of multiple consecutive respiratory cycles (e.g., 3) during the prescription familiarization phase, and calculating the difference between the site decay amounts of adjacent respiratory cycles and taking the maximum measured value. Preferably, the value range of the boundary proximity band is [0.02, 0.08].

[0117] Furthermore, after rearranging the output double pulse trains, the total cyclic stimulation charge corresponding to the next respiratory cycle is divided into the charge of the leading pulse train and the charge of the compensation pulse train. The leading pulse train is placed in the pre-phase interval before the start of the target stimulation phase window, and the compensation pulse train is placed in the subsequent interval inside the target stimulation phase window.

[0118] Meanwhile, the stimulation site group for the next respiratory cycle remains unchanged, and the candidate site refresh request is set to non-triggered state to obtain the parameter correction amount corresponding to the double-pulse rearrangement.

[0119] It should be noted that the charge of the preamble is obtained by reading the phase coordination debt increment and response delay offset of the current respiratory cycle, and by allocating it from the total periodic stimulus charge according to the ratio of the phase coordination debt increment and the response delay offset to the sum of the two. Preferably, the ratio of the preamble charge to the total periodic stimulus charge is in the range of [30%, 70%].

[0120] It should be noted that the amount of charge in the compensation pulse train is obtained by subtracting the charge in the leading pulse train from the total periodic stimulation charge. Preferably, the proportion of the charge in the compensation pulse train to the total periodic stimulation charge is in the range of [30%, 70%].

[0121] It should be noted that the pre-interval is obtained by reading the reference response delay corresponding to the initial stimulus site group under the target stimulus phase window and the response delay offset of the current respiratory cycle, and adding the time corresponding to the reference response delay and the response delay offset. Preferably, the value range of the pre-interval is [20ms, 180ms].

[0122] Furthermore, when the response delay offset is not greater than the phase allowable boundary and the site decay is greater than the site decay boundary, the dominant cause of the current respiratory cycle is determined to be site decay, and the cycle enters the site substitution confirmation stage.

[0123] In the site substitution confirmation phase, it is first determined whether the candidate site response parameter set cache is a valid cache.

[0124] When the candidate site response parameter set cache is invalidated, site migration is not performed. Instead, recovery isolation is directly output as the only control action. The candidate site refresh request is marked as triggered. At the same time, the total stimulation charge of the next respiratory cycle is corrected to the product of the total stimulation charge of the current cycle and the recovery de-loading coefficient. The stimulation advance of the next respiratory cycle remains unchanged, and the parameter correction amount corresponding to recovery isolation is obtained.

[0125] When the candidate site response parameter set cache is valid, read the candidate stimulus site group that is immediately following the current stimulus site group in the site migration order, and read the candidate unit stimulus charge auxiliary gain of the candidate stimulus site group in the candidate site response parameter set cache.

[0126] Simultaneously, the auxiliary benefit of candidate unit stimulus charge is compared with the auxiliary benefit of current respiratory cycle unit stimulus charge, and the difference between the auxiliary benefit of candidate unit stimulus charge and the auxiliary benefit of current respiratory cycle unit stimulus charge is normalized relative to the auxiliary benefit of current respiratory cycle unit stimulus charge to obtain the alternative benefit of candidate site.

[0127] When the substitution benefit of a candidate site is greater than the migration trigger boundary, the candidate stimulus site group is determined to have substitution value, and the site migration is output as the sole control action.

[0128] Simultaneously, the stimulation site group for the next respiratory cycle is corrected to the candidate stimulation site group, the stimulation advance for the next respiratory cycle is corrected to the reference response delay mapping value corresponding to the candidate stimulation site group under the target stimulation phase window, the total stimulation charge of the next respiratory cycle is kept at the same as the total stimulation charge of the current cycle, and the candidate site refresh request is set to a non-triggered state to obtain the parameter correction amount corresponding to the site migration.

[0129] Furthermore, when the substitution benefit of candidate sites is not greater than the migration trigger boundary, the candidate stimulus site group is determined to have no substitution value, and the recovery isolation is output as the only control action.

[0130] Simultaneously, the total stimulation charge of the next respiratory cycle is corrected to the product of the total stimulation charge of the current cycle and the recovery de-load coefficient. The stimulation site group and stimulation advance amount of the next respiratory cycle remain unchanged, and the candidate site refresh request is set to the triggered state to obtain the parameter correction amount corresponding to the recovery isolation.

[0131] Furthermore, after the unique control action is output, the unique control action identifier, the correction value of the stimulation site group for the next respiratory cycle, the correction value of the stimulation advance for the next respiratory cycle, the correction value of the total stimulation charge for the next respiratory cycle, the double-pulse rearrangement structure parameters, and the candidate site refresh request flag are combined according to a unified cycle number to obtain the control decision data for the current respiratory cycle.

[0132] In this embodiment, to verify the constraining effect of the method of the present invention on training safety and its impact on maintaining operation within the safety boundary under disturbance conditions, a comparison was made of contact offset, recovery hysteresis, and corresponding safety boundaries at different training stages; such as Figure 6 The overall changes in contact offset, recovery hysteresis, and corresponding safety boundaries are presented. As can be seen from the figure, although the contact offset and recovery hysteresis increase with the perturbation phase throughout the training process, they remain below the contact allowable boundary and recovery allowable boundary. This indicates that the present invention does not simply pursue higher stimulus output, but rather provides auxiliary enhancement under the constraint of the safety gating mechanism. The magnified view shows that in the section with more obvious local fluctuations, although the two state curves are close to the boundary, they do not cross the boundary. This indicates that the present invention can promptly control and constrain when the contact state changes and the recovery burden increases, thus avoiding the expansion of training risks.

[0133] S5. Write the control parameters for the next respiratory cycle based on the unique control action, and perform low-disturbance pre-detection when isolation is restored.

[0134] When the unique control action is to maintain the current prescription, the stimulation site group, stimulation advance amount, total stimulation charge of the cycle, and target stimulation phase window corresponding to the current respiratory cycle are directly written into the control queue for the next respiratory cycle.

[0135] When the unique control action is phase correction, the stimulation site group is kept as the current stimulation site group, the stimulation advance is corrected to the sum of the current stimulation advance and the stimulation advance correction value, and the total stimulation charge and the target stimulation phase window are kept unchanged before being written into the control queue for the next respiratory cycle.

[0136] When the unique control action is identified as double pulse rearrangement, the stimulation site group is kept as the current stimulation site group, the double pulse rearrangement structure parameters are written into the control queue of the next respiratory cycle, and the total stimulation charge of the cycle is split into the charge of the leader pulse train and the charge of the compensation pulse train and written into the control queue of the next respiratory cycle respectively.

[0137] When the unique control action is identified as site migration, the stimulation site group is corrected to the candidate stimulation site group, the stimulation advance is corrected to the reference response delay mapping value corresponding to the candidate stimulation site group under the target stimulation phase window, and the total stimulation charge and the target stimulation phase window are kept unchanged before being written into the control queue for the next respiratory cycle.

[0138] When the unique control action is identified as restoring isolation, the total cyclic stimulation charge is corrected to the charge reduction corresponding to restoring isolation, and the stimulation site group, stimulation advance amount, and target stimulation phase window are written into the control queue for the next respiratory cycle.

[0139] Furthermore, after completing the writing of the control parameter set for the next respiratory cycle, it is determined whether the unique control action identifier is to restore isolation, and whether the candidate site refresh request flag is in a triggered state.

[0140] When the unique control action identifier is not "Restore Isolation" or the candidate site refresh request is not in the triggered state, low-disturbance pre-probing is not performed, and the candidate site response parameter set cache from the previous round remains unchanged; when the unique control action identifier is "Restore Isolation" and the candidate site refresh request is in the triggered state, the low-disturbance pre-probing phase begins.

[0141] Furthermore, in the low-disturbance pre-detection phase, candidate stimulus sites that are immediately following the current stimulus site group in the site migration sequence are read first, and these candidate stimulus site groups are identified as the pre-detection targets for this round.

[0142] A low-perturbation pre-probe pulse is applied to the candidate stimulation site group at the reduced charge level corresponding to the restoration of isolation.

[0143] It should be noted that the low-perturbation pre-detection pulse is placed in the recovery isolation detector window outside the target stimulus phase window, and the detector pulse width, detector frequency, and number of detector pulses are consistent with the definition of low-intensity detector stimulus.

[0144] It should be noted that the recovery isolation detection window is obtained by reading the stimulation-off segment of multiple consecutive respiratory cycles during the recovery isolation phase and selecting a continuous time interval that is outside the target stimulation phase window and whose contact impedance change direction is consistent. Preferably, the duration of the recovery isolation detection window is in the range of [30ms, 200ms].

[0145] After applying a low-disturbance pre-detection pulse, real-time airflow curves and contact impedance curves corresponding to the candidate stimulation site groups are acquired simultaneously, and reference airflow curves corresponding to the current respiratory cycle duration are read from the reference airflow curve library.

[0146] Using the start and end times of the recovery isolation probe window as statistical boundaries, the auxiliary incremental area of ​​the probe window for candidate stimulus sites within the recovery isolation probe window is calculated.

[0147] The ratio of the incremental area of ​​the probe window to the amount of probe charge corresponding to the low-disturbance pre-probe pulse is used as the candidate unit stimulus charge auxiliary gain.

[0148] Simultaneously, the moment when the airflow growth rate first reaches a local peak after the candidate stimulus site group is triggered is read, the candidate reference response delay is obtained, and the average value of the stable sampling segment of the contact impedance curve within the recovery isolation detection window is read to obtain the candidate reference contact impedance.

[0149] Furthermore, the candidate unit stimulus charge assist gain, candidate reference response delay, candidate reference contact impedance, and current period number are combined to generate a candidate site response parameter set cache.

[0150] The current cycle number is determined as the generation cycle number of the candidate site response parameter set cache; then, the candidate site response parameter set cache is written into the cache record area, and the candidate site response parameter set cache of the corresponding candidate stimulus site group in the previous round in the cache record area is overwritten and updated to obtain the latest candidate site response parameter set cache.

[0151] Furthermore, it is necessary to continue to determine whether stable data has been formed during the recovery isolation phase. Specifically, real-time airflow curves, blood oxygen saturation curves, heart rate curves, and contact impedance curves are continuously read during multiple respiratory cycles (e.g., 3) within the recovery isolation phase.

[0152] The relative changes in peak airflow, blood oxygen recovery time, heart rate recovery time, and contact impedance between adjacent respiratory cycles were compared with the corresponding stability criteria boundaries.

[0153] It should be noted that the stability determination boundary is obtained by reading the relative changes in airflow peak, blood oxygen recovery time, heart rate recovery time, and contact impedance corresponding to the training breathing cycle that has been confirmed to be in a stable recovery state during the prescription familiarization phase, and taking the maximum measured value of each type of relative change. Preferably, the value range of the stability determination boundary is [0.03, 0.15].

[0154] When all relative changes are not greater than the corresponding stability determination boundary, the isolation phase of the recovery is considered to have formed stable data.

[0155] When any relative change exceeds the corresponding stability threshold, it is determined that stable data has not been formed during the recovery isolation phase, and the reference airflow curve library and site phase response baseline remain unchanged.

[0156] When stable data is determined to be formed during the recovery isolation phase, multiple consecutive complete respiratory cycles (e.g., 3) are read from within the recovery isolation phase. Time normalization is performed on each consecutive complete respiratory cycle, and the corresponding airflow value is read at the same normalized phase position. The airflow values ​​corresponding to the same phase position of each normalized respiratory cycle are arranged in chronological order, and the median value is taken to generate an updated reference airflow curve. The updated reference airflow curve is written into the reference airflow curve library at the curve position corresponding to the current cycle duration, and the updated reference airflow curve library is obtained.

[0157] After updating the reference airflow curve library, continue to update the site phase response baseline. Specifically, read the candidate unit stimulus charge assist gain, candidate reference response delay, and candidate reference contact impedance corresponding to the candidate stimulus site group during the recovery isolation phase. Then read the original reference unit probe charge assist gain, original reference response delay, and original reference contact impedance corresponding to the candidate stimulus site group under the target stimulus phase window in the site phase response baseline.

[0158] The candidate unit stimulus charge-assisted gain is compared with the original reference unit probe charge-assisted gain, and the candidate unit stimulus charge-assisted gain whose difference between the candidate unit stimulus charge-assisted gain and the original reference unit probe charge-assisted gain does not exceed the gain equivalence band is directly written into the site phase response baseline.

[0159] When the difference between the auxiliary gain of the candidate unit stimulus charge and the auxiliary gain of the original reference unit probe charge is greater than the gain equivalence band, the auxiliary gain of the candidate unit stimulus charge and the auxiliary gain of the original reference unit probe charge are averaged in the order of adjacent periods and written into the site phase response baseline. The candidate reference response delay and the candidate reference contact impedance are written into the site phase response baseline in the same way to obtain the updated site phase response baseline.

[0160] Furthermore, the control parameter set for the next respiratory cycle, the cached latest candidate site response parameter set, the updated reference airflow curve library, and the updated site phase response baseline are combined according to a unified cycle number to generate control write-back data for use in the next respiratory cycle stimulation control call.

[0161] This embodiment also provides a computer device applicable to the control method based on respiratory rehabilitation equipment, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the control method based on respiratory rehabilitation equipment as proposed in the above embodiment.

[0162] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0163] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the control method based on a respiratory rehabilitation device as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0164] In summary, this invention improves the adaptability of individualized control by constructing a basic respiratory phase sequence, a site phase response baseline, and a reference airflow curve library; it enhances the effectiveness of assistance while ensuring training safety by determining a fixed charge budget training prescription and executing stimulus control within the target stimulus phase window; and it improves the accuracy of closed-loop regulation and training stability by calculating the phase co-debt increment and outputting a unique control action.

[0165] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. 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 be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A control method based on a respiratory rehabilitation device, characterized in that, include: Collect respiratory-related physiological signals and stimulus coupling signals from trainees, and construct a basic respiratory phase sequence, site phase response baseline, and reference airflow curve library by combining low-intensity probe stimuli. Based on the basic respiratory phase sequence, site phase response baseline, and recovery safety baseline, the fixed charge budget training prescription for the current training phase is determined. Perform stimulation control for the current respiratory cycle according to the fixed charge budget training prescription, and calculate the effective auxiliary area within the target stimulation phase window based on the real-time airflow curve and the reference airflow curve; The phase coordination debt increment, response delay offset, and contact offset are calculated using the effective auxiliary area, target auxiliary area, reference response delay, and reference contact impedance. Safety gating is determined based on contact offset and recovery hysteresis. When the conditions for continued training are met, the dominant cause is separated by combining phase collaborative debt increment and site decay, and a unique control action is output. The control parameters for the next respiratory cycle are written based on the unique control action, and low-disturbance pre-detection is performed when isolation is restored.

2. The control method based on respiratory rehabilitation equipment as described in claim 1, characterized in that, The specific steps for constructing the basic respiratory phase sequence, site phase response baseline, and reference airflow curve library are as follows: Time-scale alignment and period segmentation were performed on respiratory-related physiological signals and stimulus-coupled signals to obtain the basic respiratory phase sequence; Time normalization is performed on the steady-state respiratory cycle to generate a reference airflow curve library; Low-intensity probe stimulation was applied to each stimulation site under different respiratory phase windows to obtain the auxiliary gain per unit probe charge, reference response delay, and reference contact impedance, and to construct a site phase response baseline.

3. The control method based on respiratory rehabilitation equipment as described in claim 1 or 2, characterized in that, The method for determining the fixed charge budget training prescription for the current training phase includes: Based on the baseline of site phase response, effective stimulation site phase window combinations are screened to determine the target stimulation phase window and the initial stimulation site group; The initial stimulus advance is determined based on the target stimulus phase window and the reference response delay; The total periodic stimulus charge is determined based on the restored safety baseline. The target auxiliary area sequence is generated based on the ratio between the total periodic stimulus charge and the probe charge.

4. The control method based on respiratory rehabilitation equipment as described in claim 1, characterized in that, The effective auxiliary area within the phase window of the target stimulus is calculated as follows: At the start of the current respiratory cycle, the starting point of the current respiratory cycle is determined based on the zero-crossing moment when the airflow signal changes from expiration to inspiration; The trigger time of the current cycle is determined based on the start time of the target stimulus phase window and the stimulus lead, and a stimulus pulse sequence is output to the starting stimulus site group when the trigger time of the current cycle is reached. Collect the real-time airflow curve of the current respiratory cycle and read the corresponding reference airflow curve from the reference airflow curve library; Using the start and end times of the target stimulus phase window as statistical boundaries, the positive difference between the real-time airflow curve and the current respiratory cycle reference airflow curve within the target stimulus phase window is integrated to obtain the effective auxiliary area within the target stimulus phase window.

5. The control method based on respiratory rehabilitation equipment as described in claim 1, characterized in that, The fixed charge budget training prescription includes: Phase allowable boundary, contact allowable boundary, site decay boundary, recovery allowable boundary, migration trigger boundary, and candidate site cache validity period.

6. The control method based on respiratory rehabilitation equipment as described in claim 5, characterized in that, The calculation of phase coordination debt increment, response delay offset, and contact offset includes: The moment when the airflow growth rate first reaches a local peak after the stimulus is determined based on the real-time airflow curve. The time difference between the moment when the airflow growth rate first reaches its local peak and the moment of stimulus triggering is defined as the actual response delay of the current respiratory cycle. The difference between the actual response delay and the reference response delay of the current respiratory cycle is normalized relative to the duration of the current respiratory cycle to obtain the response delay offset. The average value of the contact impedance curve of the current respiratory cycle is calculated within the stable sampling segment before the stimulus is triggered, and the average value is determined as the real-time contact impedance of the current respiratory cycle. The difference between the real-time contact impedance and the reference contact impedance during the current respiratory cycle is normalized relative to the reference contact impedance to obtain the contact offset. The auxiliary benefit per unit stimulus charge in the current respiratory cycle is determined based on the ratio between the effective auxiliary area within the target stimulus phase window and the total stimulus charge in the current respiratory cycle. The site decay is then obtained by combining the auxiliary benefit per unit probe charge of the reference unit. Based on the difference between the target auxiliary area and the effective auxiliary area, and combined with the response delay offset and contact offset, the phase coordination debt increment of the current respiratory cycle is calculated.

7. The control method based on respiratory rehabilitation equipment as described in claim 6, characterized in that, The unique output control action includes: When the contact offset exceeds the contact allowable boundary, or the recovery hysteresis exceeds the recovery allowable boundary, output recovery isolation is the only control action; When the effective auxiliary area reaches the target auxiliary area, or the phase coordination debt increment is zero, the output maintains the current prescription as the only control action; When the effective auxiliary area does not reach the target auxiliary area and the phase coordination debt increment is greater than zero, the dominant cause separation is performed based on the response delay offset and the site decay.

8. The control method based on respiratory rehabilitation equipment as described in claim 7, characterized in that, The process of separating the dominant causes includes: When the response delay offset is greater than the phase allowable boundary and the site decay is not greater than the site decay boundary, the dominant cause is determined to be the phase offset, and the phase correction is output as the only control action. When the response delay offset is greater than the phase allowable boundary and the site decay is greater than the site decay boundary, or when the response delay offset is not greater than the phase allowable boundary and the site decay is in the boundary close band, the dominant cause is determined to be mixed offset, and the output double-series pulse rearrangement is used as the only control action. When the response delay offset is not greater than the phase allowable boundary and the site decay is greater than the site decay boundary, the dominant cause is determined to be site decay, and the candidate stimulus site group is judged to have substitution value based on the candidate site response parameter set cache. When the candidate stimulus site group has alternative value, output site migration is the only control action; otherwise, output restoration of isolation is the only control action.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the control method based on the respiratory rehabilitation device according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the control method based on the respiratory rehabilitation device according to any one of claims 1 to 8.

Images