Ventilator control method and apparatus, ventilator, storage medium and program product

Abstract

This disclosure provides a ventilator control method and apparatus, a ventilator, a storage medium, and a program product. The ventilator control method includes: acquiring raw diaphragmatic electrophysiological signals of a subject; identifying ECG signals in the raw diaphragmatic electrophysiological signals; determining filter weights based on the deviation between the raw diaphragmatic electrophysiological signals and the ECG signals; filtering the ECG signals in the raw diaphragmatic electrophysiological signals using the filter to obtain a first intermediate signal; reconstructing the first intermediate signal to obtain a diaphragmatic electrophysiological signal to be processed; extracting the envelope curve of the diaphragmatic electrophysiological signal to be processed; and controlling the ventilator to switch between the expiratory and inspiratory phases based on the envelope curve.

Description

Technical Field

[0001] This disclosure relates to the field of medical device technology, and in particular to a ventilator control method and apparatus, a ventilator, a storage medium, and a program product. Background Technology

[0002] Non-invasive ventilators, with their advantages of being non-invasive, having fewer complications, and being suitable for home use, have been widely used in the treatment of various respiratory dysfunctions. Currently, traditional non-invasive ventilators mainly use pressure-triggered or flow-triggered methods. Summary of the Invention

[0003] The inventors noted that in related technologies, traditional non-invasive ventilators mainly use pressure-triggered or flow-triggered methods, which have the following obvious technical limitations.

[0004] 1. Poor patient-ventilator synchronization: Pressure / flow triggering depends on changes in airway pressure / flow during patient inspiration, with a triggering delay of more than 100ms, which can easily lead to increased breathing work, patient-ventilator asynchrony, and reduced treatment compliance.

[0005] 2. Challenges in acquiring diaphragmatic electrical signals: Diaphragmatic electrical signals (Edi) are direct neural signals that reflect the driving force of spontaneous breathing in patients. However, the surface electrode acquisition methods used in current diaphragmatic electrical signal acquisition have problems such as signal mixing, low signal-to-noise ratio, and susceptibility to interference from electrocardiogram (ECG) and chest wall electromyography, and cannot be stably used for ventilation triggering.

[0006] 3. Insufficient signal processing accuracy: Existing surface diaphragm electrical signal processing mostly uses a single filtering method, which is difficult to effectively remove ECG interference (diaphragm electrical signals and ECG signals have high frequency overlap). In addition, signal envelope extraction uses traditional methods such as rectification and low-pass filtering, which have problems such as large delay and poor dynamic response, and cannot accurately capture the patient's instantaneous respiratory effort.

[0007] 4. Insufficient neuro-mechanical coordination: Existing diaphragm-triggered ventilators have not established a dynamic mapping relationship between diaphragm electrical signals and mechanical ventilation parameters, and cannot adaptively adjust the ventilation support level according to the patient's real-time respiratory drive intensity, resulting in low assist efficiency.

[0008] Due to the aforementioned defects in existing diaphragm-triggered non-invasive ventilators, their clinical application is limited, failing to fully leverage the precise triggering advantage of diaphragm electrical signals and making it difficult to meet the personalized ventilation needs of patients with different body types and conditions.

[0009] Accordingly, this disclosure provides a ventilator control method that can effectively remove the influence of external interference on the diaphragm electrical signal, effectively improve the accuracy and synchronization of triggering, and solve the problems of large triggering delay and human-machine aggression in traditional methods.

[0010] In a first aspect of this disclosure, a ventilator control method is provided, comprising: acquiring raw diaphragmatic electrical signals of a subject; identifying an electrocardiogram (ECG) signal in the raw diaphragmatic electrical signals; determining the weights of a filter based on the deviation between the raw diaphragmatic electrical signals and the ECG signal; filtering the ECG signal in the raw diaphragmatic electrical signals using the filter to obtain a first intermediate signal; reconstructing the first intermediate signal to obtain a diaphragmatic electrical signal to be processed; extracting the envelope curve of the diaphragmatic electrical signal to be processed; and controlling the ventilator to switch between expiratory and inspiratory phases based on the envelope curve.

[0011] In some embodiments, determining the filter weights based on the deviation between the original diaphragm electroencephalogram (EEG) signal and the ECG signal includes: acquiring the sampled values ​​of the original diaphragm EEG signal from the (n-N+1)th sampling point to the nth sampling point, obtaining N first sampled values, where n is a natural number and N is a natural number greater than 1; acquiring the sampled values ​​of the ECG signal from the (n-N+1)th sampling point to the nth sampling point, obtaining N second sampled values; determining the signal error based on the sum of squared differences between the N first sampled values ​​and the N second sampled values; and determining the filter weights at the (n+1)th sampling point based on the signal error, the filter weights at the nth sampling point, and the filter input vector at the nth sampling point.

[0012] In some embodiments, determining the weight of the filter at the (n+1)th sampling point includes: calculating the product of the signal error, the input vector of the filter at the nth sampling point, and a predetermined parameter value to obtain an intermediate value; and calculating the sum of the weight of the filter at the nth sampling point and the intermediate value to obtain the weight of the filter at the (n+1)th sampling point.

[0013] In some embodiments, the signal reconstruction of the first intermediate signal includes: smoothing the first intermediate signal to obtain a second intermediate signal; and calibrating the amplitude of the second intermediate signal according to the acquisition gain of the original diaphragm electrical signal to obtain the diaphragm electrical signal to be processed.

[0014] In some embodiments, identifying the ECG signal in the original diaphragmatic electrical signal includes: preprocessing the original diaphragmatic electrical signal to obtain a third intermediate signal; and using an adaptive thresholding method to identify ECG signal features in the third intermediate signal to determine the occurrence time and waveform profile of the ECG signal.

[0015] In some embodiments, the ECG signal characteristics include at least one of the QRS complex, P wave, and T wave.

[0016] In some embodiments, the preprocessing of the original diaphragm electrical signal includes: removing the DC component and power frequency interference signal from the original diaphragm electrical signal to obtain a filtered signal; and filtering the filtered signal using a sliding window to obtain the third intermediate signal.

[0017] In some embodiments, extracting the envelope curve of the diaphragm electrical signal to be processed includes: performing high-frequency filtering on the diaphragm electrical signal to be processed to obtain a fourth intermediate signal; using the TK energy operator to obtain multiple signal amplitudes of the fourth intermediate signal; smoothing the multiple signal amplitudes to obtain candidate envelope curves; and correcting the candidate envelope curves according to the basic respiratory state of the subject to be tested to obtain the envelope curve of the diaphragm electrical signal to be processed.

[0018] In some embodiments, controlling the ventilator to switch between the expiratory and inspiratory phases based on the envelope includes: real-time monitoring of the amplitude change of the envelope curve; and controlling the ventilator to enter the inspiratory phase when the envelope curve is in an upward phase and the amplitude of the envelope curve is greater than a first amplitude threshold.

[0019] In some embodiments, controlling the ventilator to enter the inspiratory phase includes: when the envelope curve is in the rising phase and the amplitude of the envelope curve is greater than the first amplitude threshold, detecting the slope of the envelope curve; when the slope of the envelope curve is greater than or equal to the slope threshold, controlling the ventilator to enter the inspiratory phase and delivering air at a first delivery intensity.

[0020] In some embodiments, controlling the ventilator to enter the inspiratory phase includes: controlling the ventilator to enter the inspiratory phase when the slope of the envelope curve is less than the slope threshold, and delivering air at a second delivery intensity, wherein the second delivery intensity is less than the first delivery intensity.

[0021] In some embodiments, after controlling the ventilator to enter the inspiratory phase, the method further includes: determining the pressure support level value of the inspiratory phase based on the amplitude of the envelope curve and the personalized adaptation coefficient of the subject; and adjusting the delivery intensity based on the pressure support level value of the inspiratory phase, wherein the delivery intensity is positively correlated with the pressure support level value of the inspiratory phase.

[0022] In some embodiments, determining the pressure support level value of the inspiratory phase includes: calculating the product of the amplitude of the envelope curve and the personalized fit coefficient of the test subject to obtain the pressure support level value of the inspiratory phase, wherein the personalized fit coefficient of the test subject is associated with at least one of the test subject's age, weight, and disease type.

[0023] In some embodiments, the first amplitude threshold is n times the amplitude of the diaphragm electrical signal envelope of the subject under test in the end-expiratory resting state, where n is a real number greater than 1.

[0024] In some embodiments, controlling the ventilator to switch between the expiratory and inspiratory phases according to the envelope curve includes: detecting the airway flow rate of the ventilator when the amplitude of the envelope curve is less than a second amplitude threshold, wherein the second amplitude threshold is less than a first amplitude threshold; and controlling the ventilator to enter the expiratory phase when the airway flow rate of the ventilator is less than a flow rate threshold.

[0025] In some embodiments, the second amplitude threshold is m times the peak envelope value during the current respiratory cycle, where m is a real number less than 1; the flow rate threshold is k times the peak inspiratory flow rate, where k is a real number less than 1.

[0026] In a second aspect of this disclosure, a ventilator control device is provided, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute instructions stored in the memory to implement the ventilator control method as described in any of the above embodiments.

[0027] In a third aspect of this disclosure, a ventilator is provided, comprising: a ventilator control device as described in any of the above embodiments; and a signal acquisition device configured to acquire raw diaphragmatic electrical signals of a subject.

[0028] In a fourth aspect of this disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method as described in any of the above embodiments.

[0029] In a fifth aspect of this disclosure, a computer program product is provided, including computer instructions, wherein the computer instructions, when executed by a processor, implement the method as described in any of the above embodiments.

[0030] Other features and advantages of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1This is a schematic flowchart of a ventilator control method according to an embodiment of the present disclosure;

[0033] Figure 2 This is a schematic diagram of the structure of a ventilator control device according to an embodiment of the present disclosure;

[0034] Figure 3 This is a schematic diagram of the structure of a ventilator according to an embodiment of the present disclosure. Detailed Implementation

[0035] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort are within the scope of protection of this disclosure.

[0036] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of this disclosure.

[0037] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0038] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0039] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0040] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0041] Figure 1 This is a schematic flowchart of a ventilator control method according to an embodiment of the present disclosure. In some embodiments, the following ventilator control method is executed by a ventilator control device, including steps 11-17.

[0042] In step 11, the raw diaphragm electrical signal of the subject to be tested is acquired.

[0043] In some embodiments, a signal acquisition device can be used to acquire the raw diaphragm electrical signal of the subject under test.

[0044] For example, the signal acquisition device includes a single-channel flexible surface electrode array. Each flexible surface electrode uses a stretchable polyimide (PI) flexible substrate, coated with a nano-silver-carbon nanotube composite conductive layer to form a single-channel electrode (5-8 mm in diameter). This electrode closely conforms to the thoracic contour, allowing for precise positioning of the diaphragm's surface projection area (7th-10th intercostal spaces). The electrodes employ differential acquisition to simultaneously acquire single-channel diaphragmatic electrical signals and skin impedance signals.

[0045] A single-channel flexible surface electrode array is encapsulated within a self-adhesive hydrogel patch. The hydrogel utilizes a polyacrylamide-sodium alginate composite system, integrating microfluidic breathable channels (channel diameter 50-100 micrometers). It combines high conductivity (conductivity ≥ 5000 S / cm), good biocompatibility (cytotoxicity ≤ 1), and breathability. The patch edges are coated with medical-grade pressure-sensitive adhesive, ensuring a tight fit to the skin after application. This effectively prevents signal drift caused by sweating, allowing for continuous use for up to 72 hours with a single application, reducing patient discomfort and the inconvenience of frequent replacements.

[0046] In some embodiments, the signal acquisition device includes an adaptive electrode position calibration unit, which is electrically connected to the single-channel electrode and incorporates an impedance measurement chip and a machine learning calibration model. During operation, it measures the contact impedance between the single-channel electrode and the skin in real time (normal contact impedance is less than or equal to 5 ohms). When the impedance exceeds the limit, it is determined to be a poor electrode contact, and a calibration prompt is issued. At the same time, the signal characteristics (amplitude, frequency, waveform similarity) of the single-channel acquisition are analyzed by a machine learning model, and the acquisition gain is automatically adjusted to adapt to patients of different body types (adults, children) and different chest morphologies. Electrode position calibration can be completed without manual intervention, ensuring the stability of the acquired signal. Moreover, the calibration process is completed online in real time, without affecting signal acquisition and subsequent processing.

[0047] In some embodiments, the signal acquisition device includes an ultra-low power signal preamplifier, which is integrated inside the electrode patch, close to the electrode acquisition end, and employs a low-power instrumentation amplifier (operating current less than or equal to 10 kW). The system integrates a right-leg drive circuit and a shielding drive circuit to form a dual anti-interference mechanism. The right-leg drive circuit is used to suppress common-mode interference (such as power frequency interference), while the shielding drive circuit is used to reduce noise superposition during long-distance transmission. The preamplifier converts weak diaphragm electrical signals (amplitude 10-1000) into signals that are weak in amplitude. The signal is amplified to 1-5V and pre-filtered (cutoff frequency of 10-500Hz). The pre-processed signal is then transmitted to the ventilator control device in real time with a transmission delay of less than or equal to 3ms, which meets the pre-processing requirements for real-time online processing, reduces noise interference during signal transmission, and improves the signal-to-noise ratio.

[0048] It should be noted that the closer the preamplifier is to the signal source, the better the system performance. This can significantly reduce interference introduced by the wires and reduce the requirements for subsequent wires and shielding, making it suitable for wearable, long wire, and multi-channel scenarios.

[0049] In step 12, the ECG signal in the raw diaphragm electrical signal is identified.

[0050] In some embodiments, the step of identifying the ECG signal in the raw diaphragm electrical signal includes steps S101-S102.

[0051] S101. Preprocess the original diaphragm electrical signal to obtain the intermediate signal.

[0052] In some embodiments, the DC component and power frequency interference signal are removed from the original diaphragm electrical signal to obtain a filtered signal. Then, a sliding window is used to filter the filtered signal to obtain an intermediate signal.

[0053] It should be noted that by removing the DC component and applying power frequency filtering (50Hz / 60Hz notch filtering) to the original diaphragm electrical signal, DC drift and power frequency interference can be eliminated, resulting in a pre-processed filtered signal. Furthermore, using sliding window denoising (e.g., a window size of 3-5 sampling points) can suppress small-amplitude motion artifacts, thus preserving the core characteristics of the diaphragm electrical signal and ECG signal.

[0054] S102. Use the adaptive threshold method to identify ECG signal characteristics in the intermediate signal in order to determine the occurrence time and waveform profile of the ECG signal.

[0055] In some embodiments, ECG signal characteristics include at least one of QRS complex, P wave, and T wave.

[0056] In some embodiments, an ECG signal feature template library is first constructed (containing typical time-domain and frequency-domain features of QRS complexes, P waves, and T waves to adapt to single-channel signal feature extraction). Next, an adaptive thresholding method is used to identify QRS complexes (the most recognizable feature in ECG signals) in single-channel mixed signals. Through triple determination using time-domain peak value, rise slope, and waveform width, interference from similar waveforms in the diaphragmatic electroencephalogram (EEG) is eliminated, accurately locating the occurrence time and waveform contour of the ECG signal. The identification accuracy can be greater than or equal to 98%.

[0057] It should be noted that the QRS complex, P wave, and T wave are the three main waveforms in an electrocardiogram (ECG), representing different stages of cardiac electrical activity.

[0058] P wave: Represents atrial depolarization, that is, the electrical excitation process before the left and right atria contract.

[0059] QRS complex: Represents ventricular depolarization, that is, the electrical excitation process before the left and right ventricles systole, and is the most prominent wave group on an electrocardiogram.

[0060] T wave: Represents ventricular repolarization, which is the electrical activity process of the ventricle returning from a systolic state to a resting state at the end of expiration.

[0061] In step 13, the filter weights are determined based on the deviation between the original diaphragm electrical signal and the ECG signal.

[0062] In some embodiments, during the period when the ECG signal is present, the filter has a first weight. During the period when the ECG signal is absent, the filter has a second weight, which is less than the first weight.

[0063] In other words, the filter strength is increased during the period when ECG signals are present, and decreased during the period when ECG signals are absent. This allows for adaptive adjustment of the filter strength to accommodate individual differences in ECG signals among different patients.

[0064] In some embodiments, the step of determining the filter weights based on the deviation between the original diaphragm electrical signal and the ECG signal includes steps S201-S204.

[0065] S201. Obtain the sampled values ​​of the original diaphragm electrical signal from the n-N+1th sampling point to the nth sampling point, and obtain N first sampled values, where n is a natural number and N is a natural number greater than 1.

[0066] S202. Obtain the sampled values ​​of the ECG signal from the n-N+1th sampling point to the nth sampling point to obtain N second sampled values.

[0067] S203. Determine the signal error based on the sum of the squared differences between the N first sampled values ​​and the N second sampled values.

[0068] For example, signal error As shown in formula (1).

[0069] (1)

[0070] In formula (1), This represents the original diaphragm electrical signal at the k-th sampling point. Let N be the ECG signal at the kth sampling point, and N be the size of the sliding calculation window, for example, N can be 3-5.

[0071] S204. Based on the signal error, the weight of the filter at the nth sampling point, and the input vector of the filter at the nth sampling point, determine the weight of the filter at the (n+1)th sampling point.

[0072] In some embodiments, the signal error, the product of the filter's input vector at the nth sampling point and the predetermined parameter value is first calculated to obtain an intermediate value. Next, the sum of the filter's weight at the nth sampling point and the intermediate value is calculated to obtain the filter's weight at the (n+1)th sampling point.

[0073] For example, the weight of the filter at the (n+1)th sampling point is shown in formula (2).

[0074] (2)

[0075] In formula (2), The weight of the filter at the nth sampling point. For signal error, Let n be the input vector of the filter at the nth sampling point. This is a predetermined parameter value, also known as the step size factor, with a value range of 0.001-0.01, which can be dynamically adapted according to the signal and noise intensity.

[0076] In step 14, the ECG signal in the original diaphragm electrical signal is filtered using a filter to obtain the first intermediate signal.

[0077] In step 15, the first intermediate signal is reconstructed to obtain the diaphragm electrical signal to be processed.

[0078] In some embodiments, the step of reconstructing the first intermediate signal includes steps S301-S302.

[0079] S301. Smooth the first intermediate signal to obtain the second intermediate signal.

[0080] It should be noted that smoothing can effectively correct signal distortions generated during filtering and restore the original waveform characteristics of the diaphragm electrical signal.

[0081] S302. Based on the acquisition gain of the original diaphragm electrical signal, the amplitude of the second intermediate signal is calibrated to obtain the diaphragm electrical signal to be processed.

[0082] It should be noted that the amplitude of the diaphragm electrical signal is calibrated based on the acquisition gain of the single-channel signal to ensure that the signal amplitude is consistent with the patient's actual respiratory effort, and finally outputs a high-purity single-channel diaphragm electrical signal (signal-to-noise ratio greater than or equal to 35dB, which is more than 20dB higher than the original signal).

[0083] In the above embodiments, by employing a combination of "precise ECG feature recognition + adaptive composite filtering," the core problem of incomplete removal of ECG interference and the resulting distortion of diaphragmatic electrical signals under single-channel conditions is solved. It requires no multi-channel signal support, has a simple structure, low computational load (single sampling point calculation time less than or equal to 0.1ms), adapts to the simplified hardware requirements of single-channel acquisition, and possesses patient-specific adaptability, ensuring the stability and thoroughness of interference removal. Furthermore, it supports real-time online processing throughout the process, with a processing latency of less than or equal to 20ms, without affecting the real-time performance of subsequent envelope extraction and trigger control.

[0084] It should be noted that other alternative methods can also be used to filter ECG signals.

[0085] Alternative method 1: Wavelet transform + adaptive threshold filtering (suitable for scenarios with strong ECG interference)

[0086] 1. Core Principle: A 5-level wavelet decomposition is performed on the single-channel raw diaphragm electrical signal (Edi + ECG + interference) using the db4 wavelet basis, decomposing the signal into wavelet coefficients of different frequency bands. Based on the frequency band differences between the ECG signal (mainly distributed in 0.05-100Hz) and the Edi signal (10-500Hz), an adaptive threshold is set to suppress the low-frequency wavelet coefficients containing ECG interference, while retaining the high-frequency wavelet coefficients corresponding to the Edi signal. Finally, the single-channel Edi signal after removing ECG interference is reconstructed through inverse wavelet transform.

[0087] 2. Implementation Details: Wavelet decomposition and inverse transform calculation for a single sampling point takes less than or equal to 0.15ms, and processing latency is less than or equal to 25ms; the adaptive threshold is dynamically adjusted based on the ECG amplitude of each frame of signal, and the ECG interference stripping rate is greater than or equal to 97%. After stripping, the signal-to-noise ratio is greater than or equal to 33dB; there is no need for precise identification of QRS groups, simplifying the feature identification process and providing stronger resistance to sudden ECG interference.

[0088] 3. Advantages: Strong anti-interference ability, suitable for scenarios with large fluctuations in ECG interference amplitude (such as patients with arrhythmia), and does not require complex ECG feature recognition.

[0089] Alternative Method 2: Adaptive Notch Filtering + Moving Average Denoising (Suitable for low-cost, low-power scenarios)

[0090] 1. Core Principle: A three-channel adaptive notch filter is designed to target the main frequency components of the ECG signal (50Hz power frequency interference + ECG fundamental and harmonics), targeting 50Hz, 10Hz (ECG fundamental), and 20Hz (ECG second harmonic), respectively, dynamically adjusting the notch depth (0-40dB). The single-channel raw diaphragm electrical signal is sequentially subjected to the three-channel notch filter to initially remove the main ECG interference. Then, noise reduction using a moving average of 10 sampling points is applied to suppress residual small-amplitude ECG interference and motion artifacts, outputting a high-purity Edi signal.

[0091] 2. Implementation Details: The calculation time for a single sampling point of the notch filter is less than or equal to 0.08ms, and the processing delay is less than or equal to 18ms; the notch depth is dynamically adjusted according to the amplitude of ECG interference in the signal to avoid excessive notching causing distortion of the Edi signal; the signal-to-noise ratio after stripping is greater than or equal to 32dB, and the ECG interference stripping rate is greater than or equal to 96%. It has a simple structure and can be adapted to low-power hardware platforms.

[0092] 3. Advantages: low computational load, low power consumption, simple hardware implementation, low cost, and short processing latency.

[0093] In step 16, the envelope curve of the diaphragm electrical signal to be processed is extracted.

[0094] In some embodiments, the step of extracting the envelope curve of the diaphragm electrical signal to be processed includes steps S401-S404.

[0095] S401. Perform high-frequency filtering on the diaphragm electrical signal to be processed to obtain an intermediate signal.

[0096] It should be noted that the diaphragm electrical signal after removing ECG interference is subjected to high-frequency filtering (cutoff frequency of 500Hz) to eliminate residual high-frequency interference and ensure signal smoothness.

[0097] S402. Use the TK (Teager-Kaiser) energy operator to obtain multiple signal amplitudes of the fourth intermediate signal.

[0098] It should be noted that this operator only requires 3 consecutive sampling points to complete the calculation, with low computational load and high time resolution (less than or equal to 10ms), solving the problems of large delay (greater than or equal to 50ms) and poor dynamic response in the traditional rectification + low-pass filter envelope extraction.

[0099] S403. Smooth the amplitudes of multiple signals to obtain candidate envelope curves;

[0100] Instantaneous fluctuations can be eliminated by smoothing the instantaneous energy values ​​calculated by the TK energy operator using a moving average (with a sliding window size of 5-10 sampling points).

[0101] S404. Based on the basic respiratory state of the subject, the candidate envelope curve is corrected to obtain the envelope curve of the diaphragm electrical signal to be processed.

[0102] It should be noted that by introducing an amplitude correction mechanism, the calibration coefficient of the envelope amplitude is dynamically adjusted according to the patient's basic respiratory state (end-expiratory resting state, activity state), thus avoiding envelope amplitude deviation caused by individual patient differences.

[0103] This allows for the generation of smooth and precise diaphragmatic electrical signal envelope curves, which reflect the patient's instantaneous respiratory drive intensity in real time.

[0104] In the above embodiments, the TK energy operator is used to achieve rapid extraction of the diaphragm electrical signal envelope, which greatly reduces the extraction delay and requires very little computation (only 3 consecutive sampling points are needed, and the calculation time per cycle is less than or equal to 0.05ms). The entire process supports real-time online processing, and the extraction delay is less than or equal to 10ms. At the same time, through the smoothing correction mechanism, the stability and accuracy of the envelope curve are ensured, which can accurately capture the patient's respiratory effort at the beginning of inspiration and provide support for reducing the trigger delay (less than or equal to 40ms).

[0105] It should be noted that other alternative methods can also be used for envelope extraction.

[0106] Alternative Method 1: Hilbert Transform + Low-Pass Filter Envelope Extraction (suitable for scenarios prioritizing envelope stability)

[0107] 1. Core Principle: Perform Hilbert transform on the single-channel Edi signal after removing ECG interference to obtain the analytic signal; calculate the amplitude of the analytic signal to obtain the original envelope curve of the Edi signal; since the original envelope has high-frequency fluctuations, a low-pass filter (first-order Butterworth filter) with a cutoff frequency of 10Hz is used to smooth the original envelope, eliminate fluctuations, and output a stable envelope curve.

[0108] 2. Implementation Details: The Hilbert transform single-cycle calculation time is less than or equal to 0.1ms, the low-pass filtering time is less than or equal to 0.03ms, and the total extraction delay is less than or equal to 15ms; the envelope extraction accuracy is greater than or equal to 98%. The smoothed envelope curve has no obvious distortion and can accurately reflect the changing trend of the patient's respiratory drive intensity; no moving average processing is required, and the smoothing effect is more stable.

[0109] 3. Advantages: Strong envelope stability, no instantaneous fluctuations, suitable for patients with slow respiratory rhythms (such as COPD patients in the end-expiratory resting state), and high accuracy of envelope amplitude.

[0110] Alternative Method 2: Peak Detection + Linear Interpolation Envelope Extraction (Suitable for low-cost, fast-triggering scenarios)

[0111] 1. Core Principle: For the preprocessed single-channel Edi signal, an adaptive peak detection algorithm (threshold is 1.2 times the average amplitude of the current signal) is used to detect local peak points in each frame of the signal in real time; the detected peak points are filtered to remove false peaks (such as small amplitude peaks caused by interference); a linear interpolation method is used to connect adjacent valid peak points to obtain the envelope curve of the Edi signal; finally, the smoothness of the envelope is optimized by the moving average of 5 sampling points.

[0112] 2. Implementation Details: Peak detection single-sampling point calculation time is less than or equal to 0.03ms, linear interpolation time is less than or equal to 0.02ms, and total extraction delay is less than or equal to 8ms; envelope extraction accuracy is greater than or equal to 97%. It has a fast extraction speed and minimal computational load, making it suitable for high sampling frequency scenarios (greater than or equal to 1000Hz); the peak detection threshold can be dynamically adjusted to adapt to the differences in Edi signal amplitude among different patients.

[0113] 3. Advantages: Fastest extraction speed, least computational load, lowest latency, and adaptable to rapid triggering requirements.

[0114] In step 17, the ventilator is controlled to switch between the expiratory and inspiratory phases based on the envelope curve.

[0115] In some embodiments, the step of controlling the ventilator to switch between the expiratory and inspiratory phases according to the envelope curve includes steps S501-S504.

[0116] S501: Real-time monitoring of the amplitude changes of the envelope curve.

[0117] S502. When the envelope curve is in the rising phase and the amplitude of the envelope curve is greater than the first amplitude threshold, control the ventilator to enter the inspiratory phase.

[0118] It should be noted that when the envelope amplitude rises and exceeds the first amplitude threshold, it is determined that the subject has started inhaling, thereby controlling the ventilator to enter the inspiratory phase. Here, the first amplitude threshold can be referred to as the absolute threshold.

[0119] In some embodiments, the first amplitude threshold is n times the amplitude of the diaphragm electrical signal envelope of the subject at rest at the end of expiration, where n is a real number greater than 1.

[0120] For example, the first amplitude threshold is 1.5 to 2 times the amplitude of the diaphragmatic electrical signal envelope of the subject at rest at the end of expiration.

[0121] In some embodiments, when the envelope curve is in its rising phase and its amplitude is greater than a first amplitude threshold, the slope of the envelope curve is detected. If the slope of the envelope curve is greater than or equal to the slope threshold, the ventilator is controlled to enter the inspiratory phase and inhalation is performed at a first inhalation intensity.

[0122] It should be noted that when entering the inspiratory phase, if the slope of the envelope curve is large (for example, the slope of the curve is greater than or equal to 0.5V / s), it indicates that the inspiratory demand of the test object is large, so gas is delivered at a greater intensity.

[0123] Furthermore, when the slope of the envelope curve is less than the slope threshold, the ventilator is controlled to enter the inspiratory phase and inhaled at a second inhalation intensity, wherein the second inhalation intensity is less than the first inhalation intensity.

[0124] It should be noted that if the slope of the envelope curve is small when entering the inspiratory phase (e.g., less than 0.5 V / s), it indicates that the inspiratory demand of the test object is small, and therefore a smaller gas delivery intensity is used. This allows for precise gas delivery.

[0125] S503. When the amplitude of the envelope curve is less than the second amplitude threshold, the airway flow of the ventilator is detected, wherein the second amplitude threshold is less than the first amplitude threshold.

[0126] In some embodiments, the second amplitude threshold is m times the peak value of the envelope during the current respiratory cycle, where m is a real number less than 1. This second amplitude threshold can be referred to as the relative threshold.

[0127] For example, the second amplitude threshold is 0.3-0.5 times the peak envelope value during the current respiratory cycle;

[0128] S504. When the airway flow rate of the ventilator is less than the flow rate threshold, control the ventilator to enter the expiratory phase.

[0129] In some embodiments, the flow threshold is k times the peak inspiratory flow rate, where k is a real number less than 1.

[0130] For example, the flow threshold is 0.25 times the peak inspiratory flow rate.

[0131] It should be noted that by using the second amplitude threshold and the flow threshold for dual verification, the increased breathing work caused by switching too early or too late is avoided, thereby further improving human-machine synchronization.

[0132] In some embodiments, after controlling the ventilator to enter the inspiratory phase, the pressure support level for the inspiratory phase can be determined based on the amplitude of the envelope curve and the individualized fit coefficient of the test subject. Next, the delivery intensity is adjusted according to the pressure support level for the inspiratory phase, wherein the delivery intensity is positively correlated with the pressure support level for the inspiratory phase.

[0133] For example, the product of the amplitude of the envelope curve and the individualized fit coefficient of the test subject is used to obtain the pressure support level value of the inspiratory phase, wherein the individualized fit coefficient of the test subject is associated with at least one of the test subject's age, weight, and disease type.

[0134] For example, the pressure support level of the inspiratory phase. As shown in formula (3).

[0135] (3)

[0136] In formula (3), The amplitude of the envelope curve. The personalized adaptation coefficient for the object under test. When the envelope amplitude When the breathing effort increases, Synchronous increase enhances ventilation support intensity; when the envelope amplitude When breathing effort decreases, The diaphragm decreases synchronously, thereby avoiding disuse atrophy of the diaphragm caused by over-extension.

[0137] For example, The value is 0.02, and the unit is cmH2O / V.

[0138] It should be noted here that the pressure support level of the inspiratory phase... It is the difference between IPAP (Inspiratory Positive Airway Pressure) and EPAP (Expiratory Positive Airway Pressure).

[0139] In the above embodiments, by establishing a dynamic mapping relationship between diaphragmatic electrical signals and ventilation parameters, "on-demand assistance and seamless adaptation" are achieved. Simultaneously, dual-threshold triggering and dual verification during expiratory switching improve the accuracy and synchronization of triggering, solving the problems of large triggering delays and human-machine asynchrony associated with traditional methods. The algorithm logic is concise, with triggering decision time less than or equal to 5ms, supporting real-time online computation. It can dynamically adjust the triggering timing and ventilation parameters based on the instantaneous changes in the single-channel diaphragmatic electrical signal envelope, adapting to the real-time fluctuations of the patient's respiratory rhythm.

[0140] The ventilator control method provided in the above embodiments of this disclosure can effectively remove the influence of external interference on the diaphragm electrical signal, effectively improve the accuracy and synchronization of triggering, and solve the problems of large triggering delay and human-machine aggression in traditional methods.

[0141] Figure 2This is a schematic diagram of the structure of a ventilator control device according to an embodiment of the present disclosure.

[0142] like Figure 2 As shown, the ventilator control device 20 can be represented in the form of a general-purpose computing device. The ventilator control device 20 includes a memory 21, a processor 22, and a bus 23 connecting different system components.

[0143] The memory 21 may include, for example, system memory, non-volatile storage media, etc. The system memory may store, for example, an operating system, application programs, a boot loader, and other programs. The system memory may include volatile storage media, such as random access memory (RAM) and / or cache memory. The non-volatile storage media may store, for example, instructions for a corresponding embodiment of at least one ventilator control method being executed. Non-volatile storage media include, but are not limited to, disk storage, optical storage, flash memory, etc.

[0144] Processor 22 can be implemented using a general-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA) or other programmable logic devices, discrete hardware components such as discrete gates or transistors. Accordingly, each module, such as the acquisition module, calculation module, and adjustment module, can be implemented by executing instructions in the central processing unit (CPU) running memory to perform the corresponding steps, or by implementing dedicated circuits that perform the corresponding steps.

[0145] For example, processor 22 is configured as a memory-based instruction execution implementation such as Figure 1 The method involved in any of the embodiments.

[0146] Bus 23 can use any of the various bus architectures. For example, bus architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MCA) bus, and the Peripheral Component Interconnect (PCI) bus.

[0147] The interfaces 24, 25, and 26 of the ventilator control device 20, as well as the memory 21 and processor 22, can be connected via bus 23. Input / output interface 24 provides a connection interface for input / output devices such as displays, mice, and keyboards. Network interface 25 provides a connection interface for various networked devices. Storage interface 26 provides a connection interface for external storage devices such as floppy disks, USB flash drives, and SD cards.

[0148] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations thereof, can be implemented by computer-readable program instructions.

[0149] These computer-readable program instructions are provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable device to produce a machine, such that execution of the instructions by the processor produces means for implementing the functions specified in one or more boxes of the flowchart and / or block diagram.

[0150] These computer-readable program instructions may also be stored in a computer-readable storage medium. These instructions cause a computer to work in a particular manner to produce an article of manufacture, including instructions that implement the functions specified in one or more boxes in a flowchart and / or block diagram.

[0151] This disclosure may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0152] This disclosure also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement... Figure 1 The method involved in any of the embodiments.

[0153] This disclosure also provides a computer program product, including computer instructions, wherein the computer instructions, when executed by a processor, implement as follows: Figure 1 The method involved in any of the embodiments.

[0154] Figure 3 This is a schematic diagram of the structure of a ventilator according to an embodiment of this disclosure. Figure 3 As shown, the ventilator 30 includes a signal acquisition device 31 and a ventilator control device 32. The ventilator control device 32 is... Figure 2 The ventilator control device shown in any of the embodiments.

[0155] The signal acquisition device 31 is configured to acquire the raw diaphragm electrical signal of the subject under test.

[0156] The following specific examples illustrate this disclosed solution.

[0157] In some embodiments, the parameters used in this disclosure are as follows.

[0158] 1. Signal acquisition device

[0159] 1.1) Electrode: Single-channel flexible electrode with a stretchable PI material as the substrate and a nano-silver-carbon nanotube composite layer as the conductive layer. The electrode diameter is 6mm. Differential acquisition is used to accurately locate the projection area of ​​the diaphragm on the body surface.

[0160] 1.2) Electrode patch: Hydrogel conductivity 6000 S / cm, microfluidic channel diameter 80 The adhesive area is 5cm*8cm, and a single wear can last for 72 hours.

[0161] 1.3) Preamplifier: Low-power instrumentation amplifier, operating current 10 The gain is adjustable (1000-10000 times), the cutoff frequency is 10-500Hz, and it integrates the right leg drive and shielding drive circuits.

[0162] 1.4) Adaptation calibration: Impedance measurement range 0.1-10 The machine learning model uses a CNN model, and the calibration time is less than or equal to 5 seconds.

[0163] 2. Ventilator control device

[0164] 2.1) Sampling frequency: 1000Hz for diaphragm electrical signal sampling, and 500Hz for airway pressure and flow signal sampling.

[0165] 2.2) Single-channel ECG adaptive filtering algorithm: Combining adaptive IIR notch filtering with LMS adaptive filtering, the QRS group identification accuracy is greater than or equal to 98%. The LMS filter weight adjustment step size is 0.01, and the signal-to-noise ratio after interference stripping is greater than or equal to 38dB.

[0166] 2.3) TK energy operator: sliding window size of 8 sampling points, envelope extraction delay of 8ms, amplitude correction coefficient of 0.8-1.2.

[0167] 2.4) Dual threshold triggering: The absolute threshold is 1.8 times the resting envelope amplitude, and the relative threshold is 0.4 times the current peak value.

[0168] 2.5) IPAP adjustment range 4-25cmH2O, EPAP adjustment range 2-10cmH2O.

[0169] In some embodiments, the ventilator operates as follows.

[0170] 1. Wearing and initialization: The patient attaches the electrode patch to the 7th-10th intercostal space of the thoracic cavity (the surface projection area of ​​the diaphragm), starts the ventilator, and the adaptive electrode position calibration unit measures the contact impedance between the single-channel electrode and the skin in real time, automatically adjusts the acquisition gain, and completes parameter initialization (calibration time is less than or equal to 5 seconds, which does not affect the start-up and use).

[0171] 2. Signal Acquisition: A single-channel electrode acquires diaphragmatic electrical and skin impedance signals. A preamplifier amplifies the diaphragmatic electrical signal to 3V, performs preliminary filtering, and then transmits it to the control device in real time (transmission delay less than or equal to 3ms). Airway pressure and flow sensors simultaneously acquire airway pressure and flow signals and transmit them to the control device.

[0172] 3. Control and processing: The original single-channel diaphragm electrical signal is subjected to DC component removal and 50Hz notch filtering. Through ECG feature recognition + adaptive composite filtering algorithm, ECG interference is accurately removed to obtain a high-purity diaphragm electrical signal. Then, the envelope curve is extracted by TK energy operator and transmitted to the fusion trigger control module.

[0173] 4. Fusion Trigger: When the envelope amplitude exceeds the absolute threshold (1.8 times the resting amplitude), the inspiratory phase is triggered, and the IPAP is dynamically adjusted according to the envelope amplitude; when the envelope amplitude drops to the current peak value of 40... And the flow rate drops to the peak inspiratory flow rate of 25. At this time, the expiratory phase is triggered.

[0174] 5. Abnormal Handling: When the electrode patch falls off and the diaphragm electrical signal is lost, the control device switches to pressure trigger mode and issues an audible and visual alarm to prompt the patient to adjust the electrode.

[0175] 6. Monitoring and Feedback: The human-computer interaction module displays the raw electrical signals of the diaphragm, envelope curve, ventilation parameters and trigger status in real time. Medical staff can adjust parameters such as K value and dual threshold through the touch screen, and patients can view treatment data.

[0176] Implementation effect verification.

[0177] The ventilator in this embodiment was used on 50 adult COPD patients for 7 consecutive days, and the verification results are as follows.

[0178] 1. Signal quality: The average signal-to-noise ratio of the diaphragm electrical signal was 39.2 dB, and the ECG interference removal rate was greater than or equal to 98%. There was no obvious signal drift.

[0179] 2. Synchronization: The average trigger delay was 8.5ms, and the incidence of patient-machine asynchrony was 4%. Compared to traditional pressure-triggered ventilators (with an adverse reaction rate of 28%), () decreased significantly.

[0180] 3. Reliability: False trigger rate 0.8 Missed trigger rate 0.5 No ventilation interruption event occurred.

[0181] 4. Wearing experience: 86 The patient reported that the pads were comfortable to wear, with no obvious skin irritation, and that they did not need to be replaced within 72 hours.

[0182] 5. Treatment effect: Patients' respiratory rate and heart rate decreased by an average of 12. 10 Arterial blood oxygen saturation increased by an average of 5%. The treatment effect is significant.

[0183] In some embodiments, the functional units described above may be implemented as general-purpose processors, programmable logic controllers (PLCs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or any suitable combination thereof for performing the functions described herein.

[0184] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0185] The description in this disclosure is provided for illustrative and descriptive purposes only and is not intended to be exhaustive or to limit the disclosure to its forms. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of this disclosure and to enable those skilled in the art to understand this disclosure and to design various embodiments with various modifications suitable for a particular purpose.

Claims

1. A ventilator control method, comprising: Acquire raw diaphragmatic electrical signals from the subject under test; Identify the ECG signal in the raw diaphragm electrical signal; The filter weights are determined based on the deviation between the original diaphragm electrical signal and the ECG signal. The ECG signal in the original diaphragm electrical signal is filtered using the filter to obtain a first intermediate signal; The first intermediate signal is reconstructed to obtain the diaphragm electrical signal to be processed. Extract the envelope curve of the diaphragm electrical signal to be processed; Based on the envelope curve, the ventilator is controlled to switch between the expiratory and inspiratory phases.

2. The ventilator control method according to claim 1, wherein, The step of determining the filter weights based on the deviation between the original diaphragm electrical signal and the ECG signal includes: The original diaphragm electrical signal is obtained from the sampling points n-N+1 to the nth sampling point to obtain N first sampling values, where n is a natural number and N is a natural number greater than 1; The sampled values ​​of the ECG signal from the (n-N+1)th sampling point to the nth sampling point are obtained to obtain N second sampled values; The signal error is determined based on the sum of the squared differences between the N first sampled values ​​and the N second sampled values; The weight of the filter at the (n+1)th sampling point is determined based on the signal error, the weight of the filter at the nth sampling point, and the input vector of the filter at the nth sampling point.

3. The ventilator control method according to claim 2, wherein, Determining the weights of the filter at the (n+1)th sampling point includes: The intermediate value is obtained by multiplying the signal error, the input vector of the filter at the nth sampling point, and the predetermined parameter value. The weight of the filter at the nth sampling point is calculated as the sum of the intermediate value, and the weight of the filter at the (n+1)th sampling point is obtained.

4. The ventilator control method according to claim 1, wherein, The signal reconstruction of the first intermediate signal includes: The first intermediate signal is smoothed to obtain the second intermediate signal; Based on the acquisition gain of the original diaphragm electrical signal, the amplitude of the second intermediate signal is calibrated to obtain the diaphragm electrical signal to be processed.

5. The ventilator control method according to claim 1, wherein, The process of identifying the ECG signal in the original diaphragmatic electrical signal includes: The original diaphragm electrical signal is preprocessed to obtain a third intermediate signal; The ECG signal characteristics in the third intermediate signal are identified using an adaptive thresholding method to determine the occurrence time and waveform profile of the ECG signal.

6. The ventilator control method according to claim 5, wherein, The ECG signal characteristics include at least one of the QRS complex, P wave, and T wave.

7. The ventilator control method according to claim 5, wherein, The preprocessing of the raw diaphragm electrical signal includes: The DC component and power frequency interference signal are removed from the original diaphragm electrical signal to obtain the filtered signal; The filtered signal is filtered using a sliding window to obtain the third intermediate signal.

8. The ventilator control method according to claim 1, wherein, The extraction of the envelope curve of the diaphragm electrical signal to be processed includes: The diaphragm electrical signal to be processed is subjected to high-frequency filtering to obtain a fourth intermediate signal; Multiple signal amplitudes of the fourth intermediate signal are obtained using the TK energy operator; The amplitudes of the multiple signals are smoothed to obtain candidate envelope curves; Based on the baseline respiratory state of the subject under test, the candidate envelope curve is corrected to obtain the envelope curve of the diaphragm electrical signal to be processed.

9. The ventilator control method according to any one of claims 1-8, wherein, The step of controlling the ventilator to switch between the expiratory and inspiratory phases based on the envelope includes: Real-time monitoring of the amplitude changes of the envelope curve; When the envelope curve is in the rising phase and the amplitude of the envelope curve is greater than the first amplitude threshold, the ventilator is controlled to enter the inspiratory phase.

10. The ventilator control method according to claim 9, wherein, The control of the ventilator to enter the inspiratory phase includes: When the envelope curve is in the rising phase and the amplitude of the envelope curve is greater than the first amplitude threshold, the slope of the envelope curve is detected. If the slope of the envelope curve is greater than or equal to the slope threshold, the ventilator is controlled to enter the inspiratory phase and inhalation is performed at a first inhalation intensity.

11. The ventilator control method according to claim 10, wherein, The control of the ventilator to enter the inspiratory phase includes: If the slope of the envelope curve is less than the slope threshold, the ventilator is controlled to enter the inspiratory phase and delivers air at a second delivery intensity, wherein the second delivery intensity is less than the first delivery intensity.

12. The ventilator control method according to claim 9, wherein, After controlling the ventilator to enter the inspiratory phase, the method further includes: The pressure support level of the inspiratory phase is determined based on the amplitude of the envelope curve and the personalized adaptation coefficient of the test object. The air delivery intensity is adjusted based on the pressure support level of the inspiratory phase, wherein the air delivery intensity is positively correlated with the pressure support level of the inspiratory phase.

13. The ventilator control method according to claim 12, wherein, Determining the pressure support level value for the inspiratory phase includes: The product of the amplitude of the envelope curve and the personalized fit coefficient of the test subject is calculated to obtain the pressure support level value of the inspiratory phase, wherein the personalized fit coefficient of the test subject is associated with at least one of the test subject's age, weight, and disease type.

14. The ventilator control method according to claim 9, wherein, The first amplitude threshold is n times the amplitude of the diaphragm electrical signal envelope of the subject under test in the end-expiratory resting state, where n is a real number greater than 1.

15. The ventilator control method according to claim 9, wherein, The step of controlling the ventilator to switch between the expiratory and inspiratory phases based on the envelope curve includes: If the amplitude of the envelope curve is less than a second amplitude threshold, the airway flow of the ventilator is detected, wherein the second amplitude threshold is less than the first amplitude threshold. When the airway flow rate of the ventilator is less than the flow rate threshold, the ventilator is controlled to enter the expiratory phase.

16. The ventilator control method according to claim 15, wherein, The second amplitude threshold is m times the peak value of the envelope during the current respiratory cycle, where m is a real number less than 1; The flow threshold is k times the peak inhalation flow rate, where k is a real number less than 1.

17. A ventilator control device, comprising: Memory; A processor, coupled to a memory, is configured to implement the ventilator control method as described in any one of claims 1-16 based on the execution of instructions stored in the memory.

18. A ventilator, comprising: The ventilator control device as described in claim 17; The signal acquisition device is configured to acquire the raw diaphragmatic electrical signal of the subject under test.

19. A computer-readable storage medium, wherein, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the ventilator control method as described in any one of claims 1-16.

20. A computer program product comprising computer instructions, wherein the computer instructions, when executed by a processor, implement the ventilator control method as described in any one of claims 1-16.

Images