Multi-mode breathing assistance system and method based on surface diaphragm electromyography signal driving

Abstract

The disclosure provides a multi-mode respiratory assistance system and method driven by body surface diaphragm electromyography signals. The system comprises: a non-invasive body surface diaphragm electromyography signal acquisition device for obtaining a respiratory drive envelope signal; a ventilator control subsystem for generating a ventilator gas delivery instruction according to the obtained respiratory drive envelope signal and a preset ventilator control strategy; a neuromodulation high flow ventilation subsystem for calculating a target gas flow rate in real time according to the obtained respiratory drive envelope signal and a preset neuromodulation high flow control strategy, and outputting a target gas flow rate instruction; and a mode switching module for switching the system between a ventilator control mode and a neuromodulation high flow ventilation mode according to preset conditions or user instructions. The disclosure uses non-invasive body surface diaphragm electromyography signals to support seamless switching between the neuromodulation high flow ventilation mode and the ventilator control mode, and can flexibly adjust the ventilation strategy according to the user's condition, covering the whole course of management from mild to severe respiratory insufficiency.

Description

Technical Field

[0001] This disclosure relates to the field of medical device technology, and to a multi-mode respiratory assistance system and method driven by electrical signals from the diaphragm on the body surface. Background Technology

[0002] Mechanical ventilation is a core means of life support for critically ill patients. During assisted ventilation, the synchronicity between the ventilator and the patient's spontaneous breathing (i.e., patient-ventilator synchrony) directly affects the treatment effect and patient prognosis. Asynchrony can lead to adverse consequences such as increased respiratory work, diaphragmatic injury, prolonged ventilation time, and increased ICU stay.

[0003] Traditional ventilators sense a patient's inspiratory effort by detecting changes in airway pressure or flow, and then trigger ventilation. However, there is an inherent time delay between the nerve impulse sent from the patient's respiratory center and the generation of a detectable pressure or flow change in the airway. In addition, factors such as tubing resistance and intrinsic positive end-expiratory pressure (PEEPi) make it difficult for traditional triggering methods to achieve precise patient-ventilator synchronization.

[0004] Neurally Adjusted Ventilatory Assist (NAVA) technology uses a dedicated catheter (Edi catheter) with an integrated microelectrode array placed in the esophagus to directly collect electrical activity signals (EAdi) in the diaphragmatic crura region. This reflects the driving output of the respiratory center, enabling precise synchronization and proportional assistance between the ventilator and the patient's neuroventilatory effort. This technology has accumulated rich evidence-based support for improving patient-ventilator synchrony and implementing diaphragmatic protective ventilation.

[0005] However, the existing NAVA technology has the following inherent limitations: (1) Invasive operation: It requires the placement of a special Edi catheter through the nasal cavity to the lower esophagus. The operation is complicated and invasive to patients, which may cause complications such as nasopharyngeal discomfort and bleeding; (2) Dependence on special consumables: Edi catheters are special consumables for single use. They are expensive and are supplied by only a single manufacturer, which increases the cost of clinical use and supply chain risks; (3) Limited scope of application: For patients with esophageal lesions, severe coagulation dysfunction, skull base fractures, etc., the placement of Edi catheters is contraindicated or relatively contraindicated; (4) High operation threshold: The correct positioning of Edi catheters requires experience and skills. Improper positioning will seriously affect the signal quality, which limits the promotion and application of this technology in primary medical institutions.

[0006] Although the acquisition method of surface diaphragm electromyography (sEMGdi) is non-invasive, it has not been applied to real-time control of ventilators for a long time. The main reasons are: (1) The diaphragm electromyography signal acquired from the body surface is a far-field signal with a low signal amplitude (usually only 1 / 10 to 1 / 5 of the esophageal signal); (2) The body surface signal is severely contaminated by the electrocardiogram (ECG) signal, and the ECG amplitude is much larger than that of the diaphragm electromyography signal, making it difficult to effectively separate by conventional filtering; (3) The signal is also affected by crosstalk from other respiratory muscles (intercostal muscles, abdominal muscles, etc.), skin-electrode interface noise, motion artifacts, and other interference sources; (4) The above factors result in an extremely low signal-to-noise ratio of the surface diaphragm electromyography signal, which cannot meet the strict requirements of real-time closed-loop control of ventilators for signal reliability and timeliness.

[0007] In addition, high-flow nasal cannula (HFNC) has been widely used in clinical practice in recent years. HFNC continuously delivers a warmed and humidified air / oxygen mixture through a nasal cannula at a high flow rate (usually 20-60 L / min), creating a certain continuous positive pressure in the nasopharynx, which has clinical effects such as reducing the work of breathing, improving oxygenation, and flushing anatomical dead space. However, existing HFNC technology has the following core limitations: (i) Constant flow rate, unable to synchronize with respiratory neural drive: Existing HFNC devices output a constant flow rate throughout the respiratory cycle, and the pressure in the nasopharynx is basically fixed. It cannot provide maximum pressure support when the patient's inspiratory effort is strongest (the amplitude of neural drive is largest), nor can it automatically reduce the flow rate to reduce expiratory resistance during the expiratory phase. This "constant pressure" mode does not match the dynamic needs of human respiratory physiology, and the auxiliary efficiency needs to be improved. (ii) The auxiliary amount is not adjustable: Clinicians can only adjust the flow rate level as a whole, and cannot dynamically adjust the auxiliary intensity in real time according to the patient's current respiratory effort, lacking the ability to adjust it precisely and individually. (iii) Asynchrony between human and machine: Due to the constant flow rate, the nasal cavity is continuously subjected to a high positive pressure airflow during the patient's exhalation phase, which may cause discomfort and affect the coordination between human and machine.

[0008] Currently, there is no system or method that utilizes electrical signals from the diaphragm on the body surface to drive high-flow gas velocity in real time to achieve neurally regulated high-flow ventilation. Summary of the Invention

[0009] This disclosure provides a multi-mode respiratory support system and method based on diaphragmatic electrical signals driven by the body surface. It can non-invasively collect diaphragmatic electrical signals from the body surface and extract reliable respiratory drive envelope signals from them, so as to simultaneously realize the neural regulation control of the ventilator and the neural regulation control of high-flow nasal ventilation.

[0010] In a first aspect, embodiments of this disclosure provide a multi-mode respiratory support system based on surface diaphragm electrical signals. The multi-mode respiratory support system includes: a non-invasive surface diaphragm electrical signal acquisition device configured to acquire multi-channel surface bioelectrical signals and obtain a respiratory drive envelope signal based on the multi-channel surface bioelectrical signals; a ventilator control subsystem connected to the acquisition device, configured to generate a ventilator delivery command according to a preset ventilator control strategy based on the acquired respiratory drive envelope signal; a neurally regulated high-flow ventilation subsystem connected to the acquisition device, configured to calculate a target gas flow rate in real time according to a preset neurally regulated high-flow control strategy based on the acquired respiratory drive envelope signal, and output a target gas flow rate command; and a mode switching module configured to control the multi-mode respiratory support system to switch between a ventilator control mode and a neurally regulated high-flow ventilation mode according to preset conditions or user instructions; wherein the ventilator control subsystem and the neurally regulated high-flow ventilation subsystem share the same non-invasive surface diaphragm electrical signal acquisition device.

[0011] In this embodiment, a dual-mode shared architecture and seamless switching mechanism are achieved through a shared non-invasive surface diaphragm electrical signal acquisition device. It supports both closed airway positive pressure ventilation and open high-flow ventilation modes, and switches between them according to changes in the patient's condition.

[0012] In some implementations of the first aspect, the non-invasive surface diaphragm electrical signal acquisition device includes: a surface electrode array module, comprising a diaphragm electrical acquisition electrode array and a reference electrode; wherein the diaphragm electrical acquisition electrode array is attached to the surface projection area of ​​the diaphragm in the lower chest of the human body and is configured to acquire the multi-channel surface bioelectrical signals; the reference electrode is attached to a surface location away from the diaphragm region and is configured to provide a reference potential; a multi-channel bioelectrical signal acquisition module, connected to the surface electrode array module, is configured to preprocess the multi-channel surface bioelectrical signals according to the reference potential to generate multi-channel surface digital signals; and a signal processing module, configured to construct an electrocardiogram interference reference template and perform electrocardiogram interference elimination on the multi-channel surface digital signals, and then process the multi-channel surface digital signals after electrocardiogram interference elimination based on blind source separation and envelope detection technology to output a respiratory drive envelope signal.

[0013] In some implementations of the first aspect, the diaphragm electroacupuncture electrode array consists of multiple electrodes arranged on a flexible substrate according to a preset spatial configuration; wherein the spacing between the multiple electrodes is 10 mm to 30 mm; the preset spatial configuration includes: a ring array, a grid array, and / or a strip array along the rib direction; the flexible substrate is made of medical-grade silicone, polyurethane, or conductive fabric material to adapt to the surface curvature of different body types.

[0014] In some implementations of the first aspect, the signal processing module includes: an ECG interference cancellation unit, a multi-source signal separation unit, and an envelope extraction unit; wherein, the ECG interference cancellation unit is configured to construct an ECG interference reference template and estimate and remove the ECG interference component in the multi-channel body surface digital signal in real time using an adaptive noise cancellation algorithm; the multi-source signal separation unit is configured to perform multi-source signal separation on the multi-channel body surface digital signal after ECG interference component elimination using blind source separation technology to extract the optimal diaphragm electroindependent component; the envelope extraction unit is configured to perform envelope detection on the diaphragm electroindependent component to obtain a respiratory drive envelope signal reflecting the change of respiratory center drive intensity over time.

[0015] In some implementations of the first aspect, the ECG interference cancellation unit is configured to: perform principal component analysis on the multi-channel body surface digital signal, extract multiple candidate principal component signals, select a target ECG principal component signal from them, and perform polarity correction on the target ECG principal component signal to generate an ECG interference reference template and a corresponding first heartbeat peak position; perform peak detection on the multi-channel body surface digital signal to determine a second heartbeat peak position; extract heartbeat segments from the multi-channel body surface digital signal based on the second heartbeat peak position to construct a time window signal; reconstruct the ECG signal from the time window signal based on the ECG interference reference template and the first heartbeat peak position to generate a heartbeat-by-heart ECG estimation signal; and subtract the heartbeat-by-heart ECG estimation signal from the multi-channel body surface digital signal to remove ECG interference components.

[0016] In some implementations of the first aspect, the multi-source signal separation unit is configured to: perform digital bandpass filtering on the multi-channel body surface digital signal after eliminating ECG interference components to suppress low-frequency motion artifacts and high-frequency noise; perform blind source separation on the filtered multi-channel body surface digital signal using fast independent component analysis to obtain multi-channel diaphragm electro-independent components; extract multi-dimensional signal features of each channel diaphragm electro-independent component, and determine the optimal diaphragm electro-independent component based on the multi-dimensional signal features through a preset decision logic or a pre-trained classifier; wherein the multi-dimensional signal features include one or more of the following: spectral center frequency features, respiratory cycle correlation features, and spatial distribution features.

[0017] In some implementations of the first aspect, the envelope extraction unit is configured to: rectify the optimal diaphragm electroindependent component to obtain an instantaneous amplitude sequence of the optimal diaphragm electroindependent component; perform low-pass filtering on the instantaneous amplitude sequence to extract an initial envelope signal; perform baseline estimation and removal on the initial envelope signal to eliminate baseline drift components in the initial envelope signal to obtain a drift-free envelope signal; and perform normalization processing on the drift-free envelope signal to output a normalized respiratory drive envelope signal.

[0018] In this embodiment, a surface electrode array is used to acquire diaphragmatic electrical signals, eliminating the need for an indwelling esophageal catheter and removing the risk of complications (nasopharyngeal injury, bleeding, infection, etc.) associated with invasive procedures, thus significantly improving patient comfort. By acquiring multi-channel far-field diaphragmatic electrical signals through a flexible surface electrode array, and utilizing adaptive ECG elimination combined with blind source separation technology, respiratory drive envelope signals that can be used for real-time closed-loop control are extracted from a background of strong noise, overcoming the technical shortcomings of existing neuromodulation-assisted ventilation techniques that rely on invasive esophageal catheters.

[0019] In some implementations of the first aspect, the ventilator control subsystem includes: a ventilation control module configured to receive the respiratory drive envelope signal, generate a ventilator delivery control command according to a preset ventilator control strategy, and control the pneumatic unit of the ventilator to deliver supportive positive pressure ventilation through the ventilator delivery control command.

[0020] In some implementations of the first aspect, the ventilator control strategy includes: when the respiratory drive envelope signal rises above an inspiratory trigger threshold, generating and outputting an inspiratory trigger command to control the ventilator to begin delivering air; during the inspiratory phase, determining the instantaneous output pressure or flow level of the ventilator based on the product of the real-time amplitude of the respiratory drive envelope signal and a preset gain coefficient; and when the respiratory drive envelope signal falls below an expiratory trigger threshold, generating and outputting an inspiratory-expiratory switching command to control the ventilator to terminate inspiratory delivery.

[0021] In some implementations of the first aspect, the ventilation control module further includes: a signal quality assessment unit; the signal quality assessment unit is configured to assess the reliability of the respiratory drive envelope signal in real time and calculate a signal quality index; when the signal quality index is continuously lower than a signal quality threshold within a preset respiratory cycle, the ventilation triggering mode of the ventilator is switched from a control mode based on the respiratory drive envelope signal to a pressure triggering mode and / or a flow triggering mode.

[0022] In this embodiment, the processed respiratory drive envelope signal of the diaphragm electrophysiology on the body surface is used as a direct control variable for ventilator triggering, proportional assist and inspiratory-expiratory switching, so as to achieve precise synchronization between ventilator output and patient neurorespiration.

[0023] In some implementations of the first aspect, the neurally regulated high-flow ventilation subsystem includes: a high-flow rate control module configured to receive the respiratory drive envelope signal, calculate the target gas flow rate in real time according to a preset neurally regulated high-flow control strategy, and output a target gas flow rate command, thereby controlling the high-flow delivery unit to deliver high-flow gas through the target gas flow rate command; an open gas delivery module including an airway pressure sensor configured to monitor airway pressure in real time; and the high-flow rate control module further configured to adjust the target gas flow rate command according to the airway pressure.

[0024] In some implementations of the first aspect, the high-flow-rate control module is configured to: control the output gas flow rate of the high-flow-rate delivery unit to maintain a baseline airflow; acquire the respiratory drive envelope signal and calculate the target gas flow rate in real time according to a preset neural-regulated high-flow-rate control strategy, including: comparing the respiratory drive envelope signal with a preset trigger threshold; when the respiratory drive envelope signal exceeds the trigger threshold, increasing the output gas flow rate of the high-flow-rate delivery unit according to a gas flow rate adjustment formula, and generating the target gas flow rate command that matches the current respiratory drive envelope signal; when the respiratory drive envelope signal drops below a preset switching threshold, controlling the output gas flow rate of the high-flow-rate delivery unit to smoothly return to the baseline airflow, and generating the target gas flow rate command.

[0025] In some implementations of the first aspect, the high flow rate control module further includes a pressure closed-loop control unit; the pressure closed-loop control unit is configured to generate a target pressure waveform based on the respiratory drive envelope signal; calculate a correction amount for the target gas flow rate in real time based on the deviation between the target pressure waveform and the airway pressure; and adjust the target gas flow rate command based on the correction amount so that the airway pressure tracks the target pressure waveform.

[0026] In this embodiment, the respiratory drive envelope signal of the diaphragm electrophysiology on the body surface is used to regulate the gas flow rate through the open nasal interface in real time, forming a dynamic positive pressure in the nasopharynx that is synchronized with the patient's neural drive, thereby upgrading constant flow high flow oxygen therapy to an assisted ventilation mode synchronized with the respiratory rhythm.

[0027] Secondly, embodiments of this disclosure provide a multi-mode respiratory support method based on surface diaphragm electrical signals. The multi-mode respiratory support method includes: acquiring a respiratory drive envelope signal using a non-invasive surface diaphragm electrical signal acquisition device; controlling the multi-mode respiratory support system to switch between a ventilator control mode and a neurally regulated high-flow ventilation mode according to preset conditions or user instructions; wherein in the ventilator control mode, a ventilator delivery command is generated according to a preset ventilator control strategy based on the acquired respiratory drive envelope signal; and in the neurally regulated high-flow ventilation mode, a target gas flow rate is calculated in real time according to a preset neurally regulated high-flow control strategy based on the acquired respiratory drive envelope signal, and a target gas flow rate command is output.

[0028] Thirdly, embodiments of this disclosure provide an electronic device. The electronic device includes: a memory configured to store an executable program; and a processor configured to execute the program, causing the electronic device to perform the method described in the second aspect.

[0029] In this embodiment of the disclosure, the electromyographic signals of the diaphragm are collected non-invasively through a surface electrode array, eliminating the need for nasal Edi catheter placement and completely avoiding the risk of complications associated with esophageal catheter placement. At the same time, this disclosure does not rely on expensive disposable Edi catheter consumables, significantly reducing clinical usage costs and supply chain risks, and greatly lowering the operational threshold, which is conducive to its promotion and application in primary healthcare institutions.

[0030] In this embodiment of the disclosure, for the low amplitude and extremely low signal-to-noise ratio of the diaphragm electrical signal on the body surface, this disclosure adopts a two-stage blind source separation architecture of "principal component analysis (PCA) + independent component analysis (ICA)". In this architecture, PCA is used to extract the principal component signal of the electrocardiogram and construct the electrocardiogram interference reference template, while ICA is used to separate multiple independent diaphragm electrical components from the multi-channel mixed signal. Then, the optimal independent diaphragm electrical components are automatically selected through multi-dimensional signal features (frequency features, periodic features, spatial distribution features). Through the above-mentioned series architecture of "coarse extraction + fine separation + intelligent screening", the disclosure effectively extracts pure independent diaphragm electrical components from the strong noise background, and for the first time improves the signal-to-noise ratio of the diaphragm electrical signal on the body surface to a level that can meet the requirements of real-time closed-loop control of ventilators.

[0031] In this embodiment of the disclosure, addressing the technical challenge that the amplitude of ECG interference in the body surface signal is much greater than that in the diaphragm electrical signal, making it difficult for conventional filtering to effectively separate them, an adaptive interference suppression strategy of "ECG principal component extraction → polarity correction → heartbeat peak localization → heartbeat-by-heart ECG reconstruction → template cancellation" is adopted. The candidate principal component signals output by principal component analysis are screened by correlation coefficient to accurately locate the target ECG principal component signal, which is then subtracted from the original signal after polarity correction and adaptive scaling of heartbeat amplitude. Compared with filtering methods with fixed parameters, this strategy can adapt to individual differences in ECG morphology and amplitude among different patients, achieving a better balance between the thoroughness of ECG interference suppression and the fidelity of the diaphragm electrical signal.

[0032] In this embodiment, the respiratory drive envelope signal is extracted from the selected optimal diaphragm electrical independent component. After rectification, low-pass filtering, baseline estimation and removal, and normalization, a standardized respiratory drive envelope signal is obtained. Compared with traditional triggering methods that rely on changes in airway pressure or flow, this invention directly acquires the diaphragm electrical signal reflecting the output of the respiratory central nervous system, fundamentally eliminating the inherent time delay between the generation of a nerve impulse and the production of a detectable physical signal in the airway. Based on this, this disclosure adopts a control strategy of "baseline minimum flow rate + threshold triggering + proportional increase + fall back to baseline," enabling the respiratory support intensity to follow the changes in the patient's respiratory drive intensity in real time, achieving true neural-ventilator precision synchronization and effectively avoiding the discomfort caused by continuous positive pressure impact during the expiratory phase in traditional fixed flow rate modes.

[0033] In this embodiment, a three-layer architecture is constructed, consisting of a ventilator control subsystem, a neurally modulated high-flow ventilation subsystem, and a multi-mode respiratory support system. The ventilator control system generates delivery control commands based on the respiratory drive envelope signal to achieve neurally modulated assisted ventilation (NAVA) mode. The neurally modulated high-flow ventilation subsystem calculates and outputs target gas flow rate commands in real time based on the same respiratory drive envelope signal, achieving dynamic synchronization between nasal high-flow oxygen therapy and the patient's neural respiratory drive for the first time. The multi-mode respiratory support system integrates the above two modes and supports intelligent switching based on signal quality index. This architecture allows clinicians to flexibly select or switch ventilation modes on the same device platform according to the severity of the patient's condition and the stage of treatment, achieving a continuous spectrum of respiratory support from non-invasive high-flow support to invasive ventilator support.

[0034] In this embodiment of the present disclosure, with the nasal pressure sensor provided, a pressure closed-loop control mode is further provided: a target pressure waveform is generated based on the respiratory drive envelope signal, and the target gas flow rate is adjusted in real time by a PID controller, so that the measured intranasal pressure accurately tracks the target pressure waveform. This mode effectively compensates for the problem of inconsistent pressure transmission caused by the difference in nasal resistance among different patients, and further improves the auxiliary accuracy and cross-patient consistency. Attached Figure Description

[0035] Figure 1 This illustration shows the intended application scenario of the multi-mode respiratory assistance system and method driven by electrical signals from the diaphragm on the body surface, as provided in the embodiments of this disclosure.

[0036] Figure 2 The diagram shown is a structural schematic of a non-invasive diaphragm electrosignal acquisition device provided in an embodiment of this disclosure.

[0037] Figure 3 The diagram shows an attachment schematic of a body surface electrode array module provided in an embodiment of this disclosure.

[0038] Figure 4 The diagram shows a flowchart illustrating how the ECG interference cancellation unit provided in this embodiment estimates and removes ECG interference components from multi-channel body surface digital signals in real time.

[0039] Figure 5 The diagram shows a flowchart illustrating the process of a multi-source signal separation unit extracting the optimal diaphragm electrical independent component, as provided in an embodiment of this disclosure.

[0040] Figure 6 The diagram shown is a schematic of the optimal independent diaphragm electrical components extracted after processing by the multi-source signal separation unit, as provided in an embodiment of this disclosure.

[0041] Figure 7 The diagram shows a flowchart illustrating how the envelope extraction unit provided in this embodiment acquires a respiratory drive envelope signal that reflects the change in respiratory center drive intensity over time.

[0042] Figure 8 The diagram shows a flowchart of a non-invasive method for acquiring electrical signals from the diaphragm surface provided in an embodiment of this disclosure.

[0043] Figure 9 The diagram shown is a structural schematic of a ventilator control subsystem provided in an embodiment of this disclosure.

[0044] Figure 10 The diagram shown is a schematic of the timing of the respiratory drive envelope signal and the delivery pressure provided in the embodiments of this disclosure.

[0045] Figure 11 The diagram shown is a flowchart illustrating the ventilator control method provided in an embodiment of this disclosure.

[0046] Figure 12 The diagram shown is a structural schematic of a neurally modulated high-flow ventilation subsystem provided in an embodiment of this disclosure.

[0047] Figure 13 The diagram shown is a schematic representation of the neurally regulated high-flow ventilation timing provided in an embodiment of this disclosure.

[0048] Figure 14 The diagram shown is a flowchart of a neurally modulated high-flow ventilation method provided in an embodiment of this disclosure.

[0049] Figure 15 The diagram shown is a schematic representation of a multi-mode respiratory support system based on diaphragmatic electrical signals from the body surface, as provided in an embodiment of this disclosure.

[0050] Figure 16 The diagram shows a flowchart of a multimodal respiratory assistance method based on diaphragmatic electrical signals from the body surface, as provided in an embodiment of this disclosure.

[0051] Figure 17 The diagram shown is a structural schematic of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0052] The following specific examples illustrate the implementation of this disclosure. Those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. This disclosure can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0053] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this disclosure. Therefore, the illustrations only show the components related to this disclosure and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0054] In this disclosure, the terms "exemplary" or "for example" indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this disclosure should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0055] In this disclosure, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0056] Figure 1 This is a schematic diagram illustrating an application scenario of a multi-mode respiratory support system driven by electrical signals from the diaphragm, according to an embodiment of this disclosure. Figure 1 As shown, the multi-mode respiratory support system 1 includes: a non-invasive surface diaphragm electrical signal acquisition device 100, a first control module 200, a second control module 300, a mode switching module 400, a ventilator pneumatic unit 500, and a high-flow delivery unit 600.

[0057] The non-invasive diaphragm electrophysiological signal acquisition device 100 is configured to acquire multi-channel surface bioelectrical signals through a diaphragm electrophysiological acquisition electrode array attached to the surface projection area of ​​the diaphragm in the lower chest of the human body, and to provide a reference potential by a reference electrode set away from the diaphragm; to preprocess the multi-channel surface bioelectrical signals to generate multi-channel surface digital signals; to construct an electrocardiogram interference reference template and to eliminate electrocardiogram interference from the multi-channel surface digital signals; and then to process the multi-channel surface digital signals after eliminating electrocardiogram interference based on blind source separation and envelope detection technology to output a respiratory drive envelope signal.

[0058] The first control module 200 is configured to generate a ventilator delivery command according to a preset ventilator control strategy based on the acquired respiratory drive envelope signal.

[0059] The second control module 300 is configured to calculate the target gas flow rate in real time according to the acquired respiratory drive envelope signal and a preset neural regulation high flow control strategy, and output the target gas flow rate command.

[0060] The mode switching module 400 is configured to control the multi-mode respiratory support system 1 to switch between ventilator control mode and neurally modulated high-flow ventilation mode according to preset conditions or user instructions.

[0061] For example, after the system starts, the mode switching module 400 controls the system to enter the ventilator control mode according to user instructions. The first control module 200 compares the received respiratory drive envelope signal with the inspiratory trigger threshold / expiratory trigger threshold. When the respiratory drive envelope signal rises above the inspiratory trigger threshold, an inspiratory trigger command is generated and output to control the ventilator pneumatic unit 500 to start delivering air. During the inspiratory phase, the instantaneous output pressure or flow level of the ventilator is determined based on the product of the real-time amplitude of the respiratory drive envelope signal and the preset gain coefficient. When the respiratory drive envelope signal falls below the expiratory trigger threshold, an inspiratory-expiratory switching command is generated and output to control the ventilator pneumatic unit 500 to terminate inspiratory delivery.

[0062] During system operation, according to user instructions, the mode switching module 400 controls the gradual reduction of the output of the ventilator pneumatic unit 500, while simultaneously activating the high-flow-rate delivery unit 600 to deliver gas at the baseline airflow, ensuring a smooth switching process. After entering the neurally regulated high-flow-rate ventilation mode, the second control module 300 controls the output gas flow rate of the high-flow-rate delivery unit 600 to maintain the baseline airflow, and then compares the respiratory drive envelope signal with a preset trigger threshold. When the respiratory drive envelope signal exceeds the trigger threshold, the output gas flow rate of the high-flow-rate delivery unit 600 is increased according to the gas flow rate adjustment formula, and a target gas flow rate command matching the current respiratory drive envelope signal is generated. When the respiratory drive envelope signal drops below the preset switching threshold, the output gas flow rate of the high-flow-rate delivery unit 600 is smoothly reduced back to the baseline airflow, and a target gas flow rate command is generated.

[0063] The technical solutions in the embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0064] Figure 2 This is a schematic diagram illustrating the structure of a non-invasive diaphragm electrosignal acquisition device provided according to an embodiment of this disclosure. Figure 2 As shown, the non-invasive diaphragm electro-signal acquisition device 100 includes: a body surface electrode array module 110, a multi-channel bioelectrical signal acquisition module 120, and a signal processing module 130.

[0065] The body surface electrode array module 110 includes a diaphragm electroacupuncture electrode array 111 and a reference electrode 112. The diaphragm electroacupuncture electrode array 111 is attached to the diaphragm projection area on the lower chest of the human body and is configured to collect multi-channel bioelectrical signals from the body surface. The reference electrode 112 is attached to a body surface location away from the diaphragm area and is configured to provide a reference potential.

[0066] A multi-channel bioelectrical signal acquisition module 120, connected to a diaphragm electroacupuncture electrode array 111, is configured to preprocess multi-channel surface bioelectrical signals to generate multi-channel surface digital signals; and

[0067] The signal processing module 130 is configured to construct an ECG interference reference template and perform ECG interference cancellation on the multi-channel body surface digital signal. Then, based on blind source separation and envelope detection technology, it processes the ECG-cancelled multi-channel body surface digital signal and outputs a respiratory drive envelope signal.

[0068] In this embodiment, as Figure 3 As shown, the diaphragm electroacupuncture electrode array 111 consists of N electrodes arranged in a preset spatial configuration on a flexible substrate, where N is a positive integer greater than or equal to 4. Optionally, the diaphragm electroacupuncture electrode array 111 uses 16 electrodes arranged in a 4x4 grid array, with an electrode diameter of 8 mm, a row spacing of 20 mm, and a column spacing of 20 mm. When worn, the array is attached to the 7th to 10th intercostal region of the patient's right lower chest, i.e., the main surface projection position of the diaphragm dome.

[0069] The reference electrode 112 uses two standard patch electrodes placed at the sternum position to serve as the reference electrode for the diaphragm electroacupuncture electrode array 111, providing a reference potential.

[0070] The flexible substrate is made of medical-grade silicone, polyurethane, or conductive fabric, and is stretchable and breathable, adaptable to the surface curvature of patients of different body types. In alternative embodiments, the number of electrodes can be 8 to 64, the electrode spacing can be 10 mm to 30 mm, and the arrangement can be a ring array, a strip array along the rib direction, or a combination of the above.

[0071] In this embodiment, the multi-channel bioelectrical signal acquisition module 120 is configured correspondingly to the aforementioned diaphragm electroacupuncture electrode array 111, and includes a 16-channel diaphragm electroacupuncture signal acquisition front-end. Each channel is configured with a sampling rate of 1000Hz, a resolution of 24 bits, an input impedance greater than 10GΩ, a common-mode rejection ratio greater than 100dB, and an input reference noise less than 10μVrms. Within a bandwidth of 0.5Hz to 500Hz, each channel is equipped with an analog bandpass pre-filter of 0.5Hz to 1000Hz. The digital signal after analog-to-digital conversion is transmitted to the signal processing module 130 through a serial interface.

[0072] In this embodiment, the multi-channel bioelectric signal acquisition module 120 includes a signal conditioning circuit (not shown in the figure).

[0073] The 16 electrodes in the diaphragm electroacupuncture electrode array 111 ( Figure 3As shown, each electrode is connected to the signal conditioning circuit via a flexible wire. Each electrode is attached to a flexible substrate, and a corresponding conductive line is arranged on the flexible substrate. The body surface potential collected by each electrode is transmitted to the signal conditioning circuit through the corresponding conductive line.

[0074] The reference electrode 112 is connected to the reference input terminal of each differential amplifier unit in the signal conditioning circuit via an independent wire, providing a reference potential.

[0075] Specifically, the signal conditioning circuit has multiple differential amplification units, each corresponding to a signal channel.

[0076] The 16 electrodes in the diaphragm electroacupuncture electrode array 111 transmit the unipolar surface potentials (i.e., the absolute potentials of each electrode relative to the floating ground of the human body) they have acquired to the differential amplification units. At the same time, the reference potential provided by the reference electrode is synchronously transmitted to each differential amplification unit.

[0077] In any differential amplification unit, the reference potential is first subtracted from the unipolar surface potentials supplied by the two electrodes, and then the two differences are compared and amplified. Since the same reference potential is subtracted from each of the two unipolar surface potentials, the reference potentials cancel each other out when the two differences are subtracted.

[0078] Therefore, the single-channel bioelectrical signal output after differential amplification completely eliminates the reference potential component, reflecting only the surface potential difference between the corresponding attachment sites of the two electrodes. This mechanism allows a common reference potential provided by the reference electrode to simultaneously serve all differential channels, achieving uniform suppression of common-mode interference.

[0079] In this embodiment, the signal processing module 130 includes: an ECG interference cancellation unit 131, a multi-source signal separation unit 132, and an envelope extraction unit 133; wherein, the ECG interference cancellation unit 131 is configured to construct an ECG interference reference template and estimate and remove ECG interference components from the multi-channel body surface digital signal in real time using an adaptive noise cancellation algorithm; the multi-source signal separation unit 132 is configured to perform multi-source signal separation on the multi-channel body surface digital signal to remove ECG interference components using blind source separation technology and extract the optimal diaphragm electroindependent component; the envelope extraction unit 133 is configured to perform envelope detection on the diaphragm electroindependent component to obtain a respiratory drive envelope signal reflecting the change of respiratory center drive intensity over time.

[0080] The ECG interference cancellation unit 131 estimates and removes ECG interference components from the multi-channel body surface digital signal in real time using the following adaptive noise cancellation algorithm.

[0081] like Figure 4 As shown, it includes:

[0082] Step S1311: Perform principal component analysis on the multi-channel digital body surface signal, extract multiple candidate principal component signals, select the target ECG principal component signal from them, and perform polarity correction on the target ECG principal component signal to generate an ECG interference reference template and the corresponding first heartbeat peak position.

[0083] Step S1312: Perform peak detection on the multi-channel body surface digital signal to determine the position of the second heartbeat peak.

[0084] Step S1313: Based on the peak position of the second heartbeat, extract the heartbeat segment from the multi-channel body surface digital signal to construct a time window signal.

[0085] Step S1314: Based on the ECG interference reference template and the peak position of the first heartbeat, the ECG signal of the time window is reconstructed to generate the ECG estimation signal for each heartbeat.

[0086] Step S1315: Subtract the ECG estimation signal from the multi-channel body surface digital signal to remove ECG interference components.

[0087] In step S1311, the CancPCA algorithm is used to extract the ECG template. Based on principal component analysis, multiple candidate principal component signals related to ECG are separated from the multi-channel digital body surface signal. The variance contribution rate, time-domain waveform morphology and spectral characteristics of each candidate principal component are comprehensively judged, and the principal component with the highest proportion of ECG artifact energy is selected as the target ECG principal component signal. Then, the polarity of the target ECG principal component signal is corrected. The polarity of the principal component is automatically corrected according to the time-domain waveform polarity to ensure that it is consistent with the polarity of the ECG QRS complex. The corrected ECG interference reference template and the corresponding first heartbeat peak position are output.

[0088] In steps S1312 to S1313, peak detection is first performed on the multi-channel body surface digital signal to determine the position of the second heartbeat peak. Then, based on the detected position of the second heartbeat peak, heartbeat segments are extracted from the multi-channel body surface digital signal to obtain the time window signal of each heartbeat.

[0089] In steps S1314 to S1315, after determining the ECG interference reference template and the peak position of the first heartbeat, in order to accurately match the actual shape of each heartbeat in the amplitude and time axis in the multi-channel body surface digital signal, this embodiment adopts the least squares adaptive matching method to perform independent amplitude scaling and time delay correction on each heartbeat segment.

[0090] Specifically, let the heartbeat segment extracted from the multi-channel digital signal of the body surface be... ECG interference reference template is ,in This is the sampling point index. It's used to reconstruct the heartbeat fragment. The ECG estimation signal that best matches the actual ECG interference component is found in this step by solving the following least-squares optimization problem to find the optimal amplitude scaling factor. and time delay ;

[0091]

[0092] in, The amplitude scaling factor to be solved is used to eliminate the attenuation or amplification of the ECG signal amplitude caused by differences in electrode distance and skin impedance. The time delay to be solved is used to correct for small shifts in the ECG morphology on the time axis caused by differences in signal transmission path or sampling phase.

[0093] In one specific implementation, the above optimization problem is solved using a two-step method:

[0094] First, use the ECG interference reference template. Filming with Heart Perform time-domain cross-correlation calculations to search for the delay with the largest absolute value of the cross-correlation coefficient. To achieve precise alignment of the ECG interference reference template and the heartbeat segments on the time axis;

[0095] Then, within the determined delay amount Next, calculate the amplitude scaling factor. That is, the optimal amplitude matching coefficient can be obtained directly through the least squares closed solution.

[0096] In obtaining the first Optimal amplitude scaling factor for individual shots and delay Then, based on the ECG interference reference template Reconstruct the heartbeat fragment One-to-one correspondence of electrocardiogram estimation signals .

[0097] The system iterates through all detected heartbeat segments, generating corresponding beat-by-beat ECG estimation signals sequentially. Finally, the beat-by-beat ECG estimation signals are subtracted point-by-point from the multi-channel digital body surface signal, thus completing the adaptive removal of ECG interference components.

[0098] By correcting the ECG interference reference template and reconstructing the heartbeat as described above, the ECG artifacts can be precisely suppressed while preserving the diaphragmatic electrical signal components. The elimination effect on strong interference near the R wave is better than the traditional fixed template method, and the overall ECG interference suppression can reach more than 30dB.

[0099] In other implementations, one or a combination of normalized least mean square algorithm, recursive least squares algorithm, or affine projection algorithm can be used to eliminate ECG interference components. Adaptive filters can employ a variable step-size strategy, increasing the step-size factor during the duration of the ECG QRS complex to accelerate the tracking and elimination of ECG components, and restoring the normal step-size between QRS complexes.

[0100] Even after the removal of ECG interference components, the multichannel digital signal of the body surface still contains a mixture of various components such as diaphragmatic electromyography, intercostal electromyography, abdominal electromyography, and motion artifacts.

[0101] The multi-source signal separation unit 132 extracts the optimal independent electrical components of the diaphragm through the following process. For example... Figure 5 As shown, it includes:

[0102] Step S1321: Perform digital bandpass filtering on the multi-channel body surface digital signal to eliminate ECG interference components, thereby suppressing low-frequency motion artifacts and high-frequency noise.

[0103] Step S1322: Fast independent component analysis is used to perform blind source separation on the filtered multi-channel body surface digital signal to obtain multi-channel diaphragm electrical independent components.

[0104] Step S1323: Extract the multi-dimensional signal features of the independent diaphragm electrical components of each channel, and determine the optimal independent diaphragm electrical components based on the multi-dimensional signal features through a preset decision logic or a pre-trained classifier.

[0105] In step S1321, the passband of the digital bandpass filter is set to 20Hz to 500Hz to suppress low-frequency motion artifacts and high-frequency noise.

[0106] In step S1322, the fast independent component analysis algorithm is used to perform blind source separation on the filtered multi-channel digital body surface signal. The number of separated components is set to K, typically ranging from 4 to 32.

[0107] In step S1323, the multi-dimensional signal features include one or more of the following: spectral center frequency features, respiratory cycle correlation features, and spatial distribution features.

[0108] Because the order of the independent diaphragmatic electrical components output by independent component analysis is random, it is necessary to automatically identify the optimal independent diaphragmatic electrical component representing diaphragmatic electrical activity from multiple candidate components. In this embodiment, the screening of independent diaphragmatic electrical components for each channel involves constructing and calculating the following three types of discriminant features:

[0109] Spectral center frequency characteristics: Fourier transform is performed on the time-domain signal of the independent electrical components of the diaphragm in each channel to obtain its power spectral density. Calculate its spectral center frequency (SC) within the effective frequency band:

[0110]

[0111] in, and These are the lower and upper limits of integration (e.g., 20Hz and 250Hz). The main energy of the diaphragmatic electromyography (EMG) signal on the body surface is concentrated in the 60Hz–120Hz frequency band; while the main energy of ECG interference is concentrated in the low-frequency band of 0.5Hz–40Hz, and high-frequency noise is distributed above 250Hz. Therefore, when the spectral center frequency of a candidate component is within the preset range of 60Hz–120Hz, the probability of it belonging to the diaphragmatic EMG component increases significantly.

[0112] Respiratory cycle-related characteristics: The diaphragm's independent electrical component originates from diaphragmatic electrical activity, and its instantaneous energy envelope should exhibit periodic fluctuations synchronized with the patient's spontaneous breathing rhythm—that is, energy increases during inspiration and decreases during expiration. Specifically, firstly, the diaphragm's independent electrical component... Perform a Hilbert transform and take its modulus to obtain the instantaneous energy envelope. (Alternatively, the envelope can be obtained using "rectification + low-pass filtering"), and then the autocorrelation function of the energy envelope is calculated. :

[0113]

[0114] The envelope signal of the diaphragm's independent electrical components exhibits significant respiratory rhythm periodicity, with its autocorrelation function corresponding to a delay time of 2–5 seconds in the respiratory cycle. The autocorrelation function should exhibit a distinct peak value, which is significantly larger than the autocorrelation value at adjacent delays. When the ratio of the peak value to the second-highest value is greater than a preset threshold (e.g., 1.5), the diaphragm electrical independent component is determined. It exhibits rhythmic characteristics related to the respiratory cycle in the time domain.

[0115] Spatial distribution characteristics: Since the surface electrode array is attached to the diaphragm projection area in a matrix manner, the spatial distribution of different physiological source signals on the electrode array has significant differences: the equivalent dipole source of the diaphragm electrical independent component is located in the diaphragm region (anatomically lower), and the spatial weighted distribution is concentrated in the middle and lower part of the electrode array; the cardiac interference source is located in the viscera (anatomically left and upper), and the spatial weighted distribution is concentrated in the upper left part of the electrode array.

[0116] Based on the corresponding column vectors in the inverse matrix (i.e., the mixing matrix) of the separation matrix output by independent component analysis, the spatial projection weights of each independent diaphragmatic electrical component are calculated. Specifically, the column vectors are rearranged into a spatial weight matrix according to the physical topology of the body surface electrode array, and its energy distribution centroid is calculated. When the energy distribution centroid is located within the preset diaphragmatic projection region of the electrode array (such as the lower middle part of the array), it is determined that the component has a typical spatial distribution pattern of diaphragmatic electrical signals.

[0117] After calculating the three characteristic indicators mentioned above, a final selection is made using either a weighted scoring method or a threshold determination method:

[0118] Weighted scoring method: Assign weights to the spectral center frequency, respiratory cycle correlation, and spatial distribution centroid, calculate the comprehensive quality score of each diaphragmatic electroindependent component, and select the diaphragmatic electroindependent component with the highest score as the optimal diaphragmatic electroindependent component.

[0119] Threshold determination method: The diaphragm electrical independent component must simultaneously meet the following three conditions to be identified as the optimal diaphragm electrical independent component:

[0120] (1) Center frequency of the spectrum Hz;

[0121] (2) The peak value ratio of energy envelope autocorrelation is greater than the preset threshold (e.g., 1.5).

[0122] (3) The center of gravity of energy distribution is located in the lower half of the electrode array.

[0123] like Figure 6 The diagram shown is a schematic representation of the optimal independent electrical components of the diaphragm extracted after processing by the multi-source signal separation unit 132. This is the filtered multi-channel digital signal of the body surface. To remove ECG interference components from multi-channel body surface digital signals, The diaphragm's electrical components are independent components. This is the final selected optimal independent component of diaphragmatic electrical activity.

[0124] In some implementations, nonnegative matrix factorization or convolutional blind source separation can also be used to obtain the optimal independent components of the diaphragm electrical system.

[0125] The envelope extraction unit 133 acquires the respiratory drive envelope signal, which reflects the change in the intensity of respiratory center drive over time, through the following process. For example... Figure 7 As shown, it includes:

[0126] Step S1331: Rectify the optimal diaphragm electrical independent component to obtain the instantaneous amplitude sequence of the optimal diaphragm electrical independent component.

[0127] Step S1332: Perform low-pass filtering on the instantaneous amplitude sequence to extract the initial envelope signal.

[0128] Step S1333: Baseline estimation and removal are performed on the initial envelope signal to eliminate the baseline drift component in the initial envelope signal and obtain the drift-free envelope signal.

[0129] Step S1334: Normalize the drift-free envelope signal and output the normalized respiratory drive envelope signal.

[0130] In step S1331, the optimal diaphragm electrical independent components are... Perform full-wave rectification and take the absolute value to obtain the instantaneous amplitude sequence. .

[0131] Full-wave rectification is used to separate the optimal diaphragmatic electrical components. The negative phase component is flipped to the positive phase, so that subsequent envelope detection can completely preserve its energy envelope information.

[0132] In other implementations, half-wave rectification is used, which retains only the positive amplitude and discards the negative amplitude. Since the diaphragm electrical independent component exhibits bidirectional fluctuation characteristics during the inspiration phase, full-wave rectification can utilize the positive and negative phase information of the diaphragm electrical independent component, making the energy of the rectified diaphragm electrical independent component more abundant and the sensitivity of envelope detection higher. Therefore, full-wave rectification is the preferred solution.

[0133] In step S1332, the rectified instantaneous amplitude sequence is... Low-pass filtering is performed to extract the low-frequency modulation components and obtain the initial envelope signal. .

[0134] In this embodiment, a fourth-order Butterworth low-pass filter is used for low-pass filtering. The filter cutoff frequency is set to 2Hz to 5Hz. The selection of the cutoff frequency is based on the physiological range of respiratory rate—the normal resting respiratory rate of a person is usually 12 to 20 breaths / min (corresponding to 0.2 to 0.33Hz), and in extreme cases it can reach 60 breaths / min (corresponding to 1Hz). Setting the cutoff frequency to 2 to 5Hz can not only retain all the effective components within the respiratory rate range, but also effectively filter out high-frequency electromyographic noise components in the diaphragm electrical signal that are higher than the respiratory rate.

[0135] The transfer function of the Butterworth filter is:

[0136]

[0137] in, This is the cutoff angular frequency (corresponding to 2~5Hz). The fourth-order Butterworth filter has the flattest amplitude-frequency response characteristic within the passband, effectively preserving the waveform shape of the envelope signal without introducing passband ripple distortion.

[0138] In other implementations, envelope detection is performed by analyzing the optimal diaphragmatic electrical independent components. This is achieved using the Hilbert transform. The Hilbert transform is suitable for applications requiring higher accuracy in envelope detection, but it involves relatively large computational costs.

[0139] In step S1333, during the acquisition of electrical signals from the diaphragm on the body surface, the initial envelope signal is affected by factors such as changes in electrode-skin interface impedance caused by respiratory movements, changes in patient position, and amplifier temperature drift. The baseline often contains a slowly varying low-frequency drift component, which can affect the accurate assessment of subsequent respiratory drive intensity.

[0140] In this embodiment, the sliding percentile method is used for baseline estimation. Specifically, the sliding time window length is set to 8 seconds (covering approximately 2-3 respiratory cycles), and at each sampling time... , take window Envelope signal within The 10th percentile is used as the baseline estimate at that time. :

[0141]

[0142] The reason for choosing the 10th percentile instead of the minimum or mean is that the 10th percentile is not sensitive to sudden motion artifacts and ECG spikes, has better stability than the minimum, and reflects the true baseline level of the envelope signal "bottom" better than the mean, thus more accurately tracking the long-term drift trend of the envelope signal.

[0143] Subsequently, from the initial envelope signal Subtract the corresponding baseline estimate point by point Obtain the drift envelope signal :

[0144]

[0145] When the drift envelope signal When a negative value is found, it is set to zero to ensure that the signal amplitude in subsequent processing is non-negative.

[0146] In other implementations, baseline estimation uses the moving average method (taking the arithmetic mean of the signals within the window) or the polynomial fitting method (applying a low-order polynomial to the signals within the window and taking the fitted value as the baseline estimate). However, compared to the percentile method, these two methods are less robust to interference.

[0147] In step S1334, to eliminate individual differences in signal amplitude among different patients and under different electrode attachment conditions, and to enable the subsequent ventilator control module and high-flow rate control module to use uniform gain parameters, this embodiment preferably uses a drift-de-envelope signal. Normalization is performed.

[0148] In this embodiment, the drift-free envelope signal is calculated. The moving average of the peak value within a recent time window, which is 30 to 60 seconds in length, is the moving average of the peak value (maximum value) of the signal within the window.

[0149] After normalization, the respiratory drive envelope signal The amplitude range is dynamically constrained to the 0~1 interval, making it comparable between different patients and facilitating the use of a unified gain parameter for control of the ventilator control subsystem and the neuromodulated high-flow ventilation subsystem.

[0150] In other implementations, the normalization process uses a max-min normalization method, where the maximum and minimum values ​​are taken from the global extrema within the current signal segment. However, max-min normalization is too sensitive to extreme values. In contrast, normalization based on the peak moving average has better robustness and is the preferred solution in this embodiment.

[0151] It should be noted that if the subsequent control module has adaptive gain adjustment capability, the normalization process can be omitted, and the breathing drive envelope signal can be directly output as a drift-free envelope signal. .

[0152] After processing through steps S1331 to S1334, the envelope extraction unit 133 outputs the normalized respiratory drive envelope signal. It has the following waveform characteristics: during the patient's inspiration phase, the amplitude increases due to the enhanced electrical activity of the diaphragm; during the patient's expiration phase, the amplitude decreases due to the weakened electrical activity of the diaphragm, and the amplitude falls back to near the baseline.

[0153] The waveform morphology is highly consistent with the envelope morphology of the clinically recognized esophageal EAdi signal, which can accurately reflect the neural drive output of the patient's respiratory center and meet the requirements of real-time closed-loop control of ventilators for signal reliability and timeliness.

[0154] This disclosure provides a non-invasive method for acquiring electrical signals from the diaphragm on the body surface, applied to the non-invasive electrical signal acquisition device for the diaphragm on the body surface described in the above embodiments. For example... Figure 8 As shown, it includes the following steps:

[0155] Step S101: Multi-channel surface bioelectric signals are acquired by an array of diaphragm electroacquisition electrodes attached to the surface projection area of ​​the diaphragm in the lower chest of the human body, and a reference potential is provided by a reference electrode located away from the diaphragm.

[0156] Step S102: Preprocess the multi-channel bioelectrical signals on the body surface according to the reference potential to generate multi-channel digital signals on the body surface.

[0157] Step S103: Construct an ECG interference reference template and perform ECG interference cancellation on the multi-channel body surface digital signal. Then, process the multi-channel body surface digital signal after ECG interference cancellation based on blind source separation and envelope detection technology to output the respiratory drive envelope signal.

[0158] The specific implementation methods of steps S101 to S103 above have been described in detail in the system embodiment, and will not be repeated here to avoid repetition.

[0159] Figure 9 This is a schematic diagram illustrating the structure of a ventilator control subsystem provided according to an embodiment of the present disclosure. For example... Figure 9 As shown, the ventilator control subsystem 900 includes a ventilation control module 910.

[0160] The ventilation control module 910 is configured to receive the respiratory drive envelope signal acquired by the non-invasive diaphragm electrical signal acquisition device 100, and generate a ventilator delivery control command according to the preset ventilator control strategy, and control the pneumatic unit of the ventilator to deliver supportive positive pressure ventilation through the ventilator delivery control command.

[0161] In this embodiment, the non-invasive diaphragm electro-signal acquisition device 100 and its specific structure and working principle have been described in detail in the above embodiments, and will not be repeated here.

[0162] The ventilation control module 910 acquires the normalized respiratory drive envelope signal. The system generates ventilator delivery control commands according to a preset ventilator control strategy. This ventilator control strategy includes the following sub-steps:

[0163] (a) Trigger detection: The ventilation control module 910 monitors the respiratory drive envelope signal in real time. The amplitude of the respiratory drive envelope signal is compared with the inspiratory trigger threshold: when the respiratory drive envelope signal... When the amplitude of the ventilator continuously increases and exceeds the inspiratory trigger threshold, it is determined that the patient is making an inspiratory effort, and the ventilator is triggered to start delivering air.

[0164] (b) Assisted level calculation: During the inspiratory phase, the ventilation control module 910 calculates the respiratory drive envelope signal. The current amplitude is used to calculate the level of support for this breath according to the proportional support principle, and to determine the instantaneous output pressure or flow level of the ventilator; specifically, the level of support (such as pressure support value or tidal volume) is related to the respiratory drive envelope signal. The amplitude is directly proportional to:

[0165] (c) Generation of ventilation command: The ventilation control module 910 generates a ventilator ventilation control command based on the level of support and preset ventilation parameters (such as inspiratory time, expiratory trigger sensitivity, etc.). The ventilation control command includes a target pressure value or a target tidal volume value, as well as ventilation duration parameters.

[0166] (d) Expiratory phase management: The ventilation control module 910 monitors the respiratory drive envelope signal in real time. The amplitude of the respiratory drive envelope signal is compared with the expiratory trigger threshold: when the respiratory drive envelope signal... When the amplitude drops below the expiratory trigger threshold, the ventilation control module 910 generates an inspiratory-expiratory switching command, controls the ventilator to stop delivering air and switch to the expiratory phase, allowing the patient to exhale passively.

[0167] Compared to traditional ventilator triggering methods that rely on changes in airway pressure or flow to detect a patient's inspiratory intent, this embodiment directly acquires diaphragmatic electrical signals reflecting the output of the respiratory center's nervous system, fundamentally eliminating the inherent time delay between the generation of a nerve impulse and the production of a detectable physical signal in the airway. The triggering delay in this embodiment is primarily determined by the signal processing time (typically less than 50ms), which is significantly less than the time delay of traditional pressure / flow triggering methods (typically 100ms to 300ms), thus significantly improving patient-ventilator synchronization.

[0168] like Figure 10 The diagram shows the timing of the respiratory drive envelope signal and the delivery pressure, from top to bottom: the original diaphragmatic electrical independent component, the respiratory drive envelope signal, and the closed-loop neural regulation positive pressure ventilation pressure response.

[0169] This disclosure provides a ventilator control method, applied to the ventilator control subsystem described in the above embodiments. For example... Figure 11 As shown, it includes the following steps:

[0170] Step S901: Acquire respiratory drive envelope signals using a non-invasive surface diaphragm electrical signal acquisition device.

[0171] Step S902: Receive the respiratory drive envelope signal and generate a ventilator delivery control command according to the preset ventilator control strategy. Control the pneumatic unit of the ventilator to deliver supportive positive pressure ventilation through the ventilator delivery control command.

[0172] The specific implementation methods of steps S901 to S902 have been described in detail in the system embodiments and will not be repeated here to avoid repetition.

[0173] Figure 12 This is a schematic diagram illustrating the structure of a neurally modulated high-flow ventilation subsystem provided according to an embodiment of the present disclosure. Figure 12 As shown, the neurally regulated high-flow ventilation subsystem 1200 includes a high-flow rate control module 1210 and an open-type gas delivery module 1220.

[0174] The high flow rate control module 1210 is configured to receive the respiratory drive envelope signal acquired by the non-invasive diaphragm electrical signal acquisition device 100, calculate the target gas flow rate in real time according to the preset neural regulation high flow rate control strategy, and output the target gas flow rate command to control the high flow rate delivery unit to deliver high flow rate gas through the target gas flow rate command.

[0175] The open-type gas delivery module 1220 includes an airway pressure sensor configured to monitor airway pressure in real time.

[0176] The high flow rate control module 1210 is also configured to adjust the target gas flow rate command based on the airway pressure.

[0177] The high-flow-rate control module 1210's hardware core is a precision proportional electromagnetic flow valve (range 0~80L / min, response time ≤20ms, control accuracy ±0.5L / min), controlled by an embedded real-time controller (main frequency ≥1GHz, real-time operating system ensuring control cycle ≤5ms) based on the breathing drive envelope signal. Real-time driven.

[0178] The open-type gas delivery module 1220 uses a single tube to connect to the nasal tube, which comes in three sizes: S, M, and L, to accommodate different adult nasal cavity sizes.

[0179] In this embodiment, the non-invasive diaphragm electro-signal acquisition device 100 and its specific structure and working principle have been described in detail in the above embodiments, and will not be repeated here.

[0180] The high-flow-rate control module 1210 acquires the normalized respiratory drive envelope signal. The target gas flow rate command is generated according to a preset neural-modulated high-flow control strategy. The neural-modulated high-flow control strategy includes the following sub-steps:

[0181] (a) Maintaining a baseline minimum flow rate: Throughout the respiratory cycle, the high-flow rate control module 1210 controls the high-flow delivery unit to maintain a baseline airflow rate of minimum flow rate at its output. For example, the minimum flow rate is set to 10 L / min to 20 L / min to maintain continuous flushing and basal oxygenation of the patient's airway during both the inspiratory and expiratory phases, while maintaining a certain continuous positive pressure in the nasopharynx.

[0182] (b) Real-time monitoring and threshold comparison: The high flow rate control module 1210 acquires the respiratory drive envelope signal in real time. And compare its amplitude with the preset trigger threshold.

[0183] (c) Proportional assisted ventilation: When the respiratory drive envelope signal When the amplitude exceeds the preset trigger threshold (indicating that the patient is inhaling), the high flow rate control module 1210 triggers the high flow rate control module based on the respiratory drive envelope signal. The amplitude and gas flow rate adjustment formulas determine the target gas flow rate proportionally:

[0184]

[0185] in, Minimum flow rate (e.g., 10 L / min). This is a preset flow rate ratio coefficient (unit: L / min, value range: 10 to 40). Set a preset trigger threshold (e.g., 0.1 to 0.2).

[0186] (d) Return to baseline: when the respiratory-driven envelope signal When the amplitude drops below the preset switching threshold (indicating the end of the inspiratory phase and the start of the expiratory phase), the high flow rate control module 1210 re-determines the target gas flow rate as the minimum flow rate and generates a corresponding target gas flow rate command to control the output gas flow rate of the high flow rate delivery unit to smoothly drop back to the minimum flow rate.

[0187] (e) Flow rate smoothing and ramp control: To avoid nasal discomfort caused by sudden changes in flow rate, the target gas flow rate is controlled. Apply the following treatment:

[0188] Rise slope limit: The rate of increase of the flow velocity shall not exceed ΔQ_rise_max (preferably 100L / min / s);

[0189] Descending ramp limit: The rate of flow velocity descent shall not exceed ΔQ_fall_max (preferably 50L / min / s);

[0190] First-order low-pass filtering (time constant τ=20ms) eliminates high-frequency jitter.

[0191] (f) Airway pressure formation mechanism: Due to the non-sealed connection of the gas delivery interface, airway pressure... The target gas flow rate With the patient's nasal resistance The nonlinear function. In the getter phase:

[0192]

[0193] in, , This is an individual-specific coefficient related to the patient's nasal cavity anatomy. Typical values ​​are 0.01~0.05 cmH2O·min / L. The typical value is 1.5~2.0.

[0194] In this embodiment, the high-flow-rate control module 1210 does not need to actively distinguish between the inspiratory and expiratory phases. Instead, it automatically tracks the phase by adjusting the amplitude of the respiratory drive envelope signal. When the envelope signal amplitude exceeds the trigger threshold, the flow rate is automatically increased (corresponding to the inspiratory phase), and when the envelope signal amplitude falls below the switching threshold, the flow rate is automatically decreased (corresponding to the expiratory phase). This "signal amplitude-driven" control method enables real-time, continuous, and proportional responses to the patient's respiratory effort.

[0195] In other embodiments, the high-flow-rate control module 1210 may also be based on the respiratory drive envelope signal. The system actively determines the patient's inspiratory and expiratory phases: the inspiratory phase begins when the amplitude of the envelope signal rises above the trigger threshold, and the expiratory phase begins when the amplitude of the envelope signal falls below the switching threshold. Based on this, the high-flow delivery unit is controlled to output gas at a first flow rate (higher than the minimum flow rate) during the inspiratory phase, and at a second flow rate (i.e., the minimum flow rate) during the expiratory phase.

[0196] In this embodiment, with the open gas delivery module 1220 equipped with an intranasal pressure sensor, the high flow rate control module 1210 also includes a pressure closed-loop control unit 1211, which further improves assistive accuracy and cross-patient consistency by enabling the pressure closed-loop control mode.

[0197] The intranasal pressure sensor is a miniature pressure sensor placed inside the nasal cannula near the entrance of the patient's nasal cavity to collect pressure signals within the patient's nasopharynx in real time.

[0198] In the pressure closed-loop control mode, the pressure closed-loop control unit 1211 first determines the pressure based on the respiratory drive envelope signal. Generate target pressure waveform The shape of the target pressure waveform is proportional to the wave formation of the respiratory drive envelope signal:

[0199]

[0200] Wherein, K_p is a preset pressure proportionality coefficient (unit: cmH2O, value range: 2 to 10). This represents the current amplitude of the normalized respiratory drive envelope signal.

[0201] Subsequently, the pressure closed-loop control unit 1211 acquires the measured intranasal pressure signal collected by the intranasal pressure sensor. And calculate the target pressure waveform. Compared with the measured intranasal pressure signal Deviation between .

[0202] The pressure closed-loop control unit 1211 calculates the correction amount of the target gas flow rate in real time through a PID controller. According to the correction amount Adjust the target gas flow rate To make the measured intranasal pressure signal Accurately track target pressure waveform Through the aforementioned closed-loop pressure control, the system can effectively compensate for the inconsistency in pressure transmission caused by differences in nasal resistance among different patients, ensuring that the actual pressure delivered to the patient's nasopharynx is consistent with the ideal pressure calculated based on the diaphragm's electrical signals.

[0203] In some implementations, the neurally modulated high-flow ventilation subsystem 1200 also includes a gas heating and humidification module 1230: heating and humidifying the output gas to maintain the output gas temperature at 37℃±1℃ and relative humidity ≥95% in order to protect the patient's respiratory mucosal barrier function.

[0204] In some implementations, the neurally modulated high-flow ventilation subsystem 1200 continuously records the respiratory drive envelope signal reflecting the patient's respiratory drive demand. The peak value and respiratory rate are used to generate a corresponding trend graph, which is used to assess the patient based on the respiratory drive envelope signal. The peak value decreases over time, while the minimum flow rate... If the flow rate remains constant, it indicates that the patient's respiratory function is gradually improving. This can be used as a reference to gradually reduce the flow rate proportional coefficient. .

[0205] Figure 13 This is a schematic diagram of the timing of neurally regulated high-flow ventilation. From top to bottom, it shows: the original diaphragmatic electrical independent component, the respiratory drive envelope signal, the neurally regulated high-flow ventilation response, and the airway pressure response.

[0206] This disclosure provides a neurally modulated high-flow-rate ventilation method, applied to the neurally modulated high-flow-rate ventilation subsystem described in the above embodiments. For example... Figure 14 As shown, it includes the following steps:

[0207] Step S1201: Acquire respiratory drive envelope signals using a non-invasive surface diaphragm electrical signal acquisition device.

[0208] Step S1202: Using the respiratory drive envelope signal and airway pressure, the target gas flow rate is calculated in real time according to the preset neural regulation high flow control strategy, and the target gas flow rate command is output. The high flow delivery unit is controlled to deliver high flow gas through the target gas flow rate command.

[0209] The specific implementation methods of the above steps S1201 to S1202 have been described in detail in the system embodiment, and will not be repeated here to avoid repetition.

[0210] Figure 15 This is a schematic diagram illustrating the structure of a multimodal respiratory support system based on electrical signals from the diaphragm surface, according to an embodiment of this disclosure. Figure 15 As shown, the multi-mode respiratory support system 1 includes: a non-invasive surface diaphragm electrical signal acquisition device 100, a ventilator control subsystem 900, a neurally modulated high-flow ventilation subsystem 1200, and a mode switching module 400.

[0211] A non-invasive surface diaphragm electrical signal acquisition device 100 is configured to acquire respiratory drive envelope signals.

[0212] Ventilator control subsystem 900 (e.g.) Figure 9 As shown, it is connected to a non-invasive diaphragm electrical signal acquisition device 100 and is configured to generate a ventilator delivery command according to a preset ventilator control strategy based on the acquired respiratory drive envelope signal.

[0213] The neurally modulated high-flow ventilation subsystem 1200 (e.g.) Figure 12 As shown), it is connected to a non-invasive surface diaphragm electrical signal acquisition device 100, and is configured to calculate the target gas flow rate in real time according to a preset neurally modulated high-flow control strategy based on the acquired respiratory drive envelope signal, and output the target gas flow rate command; and

[0214] The mode switching module 400 is configured to control the multi-mode respiratory support system 1 to switch between ventilator control mode and neurally modulated high-flow ventilation mode according to preset conditions or user instructions.

[0215] The ventilator control subsystem 900 and the neuromodulated high-flow ventilation subsystem 1200 share the same non-invasive surface diaphragm electrical signal acquisition device 100.

[0216] In this embodiment, the mode switching module 400 is configured to control the multi-mode respiratory support system 1 to switch between ventilator control mode and neurally modulated high-flow ventilation mode according to preset conditions or user instructions.

[0217] In ventilator control mode, the ventilator control subsystem 900 generates ventilator delivery control commands based on the respiratory drive envelope signal, and controls the ventilator pneumatic unit to deliver supportive positive pressure ventilation to the patient.

[0218] In the neurally regulated high-flow ventilation mode, the neurally regulated high-flow ventilation subsystem 1200 generates a target gas flow rate command based on the respiratory drive envelope signal, and controls the high-flow delivery unit to deliver high-flow gas synchronized with the respiratory drive to the patient.

[0219] In one implementation, the mode switching module 400 automatically performs mode switching based on the signal quality index (SQI) obtained from the signal quality assessment.

[0220] When the SQI is higher than or equal to a preset threshold (e.g., 0.6), the mode switching module 400 maintains or switches to a neural modulation mode based on diaphragmatic electrical signals (ventilator control mode or neural modulation high flow ventilation mode).

[0221] When the SQI is lower than the preset threshold, the mode switching module 400 switches the control mode from the neural modulation mode based on diaphragm electrical signals to the traditional control mode based on pressure or flow triggering.

[0222] Switching to traditional control mode when the SQI is below the threshold ensures that respiratory support is not interrupted even when the body surface signal is disturbed (such as electrode detachment, motion artifacts, environmental noise, etc.).

[0223] In another implementation, the mode switching module 400 switches modes based on mode selection commands input by the user through a human-machine interface. The human-machine interface includes a touch screen display for displaying real-time information such as the current operating mode, breathing drive envelope signal waveform, and signal quality index, and for receiving the user's mode selection operation.

[0224] In this embodiment, the ventilator control subsystem 900 and the neuromodulated high-flow ventilation subsystem 1200 share the same non-invasive surface diaphragm electrical signal acquisition device 100. Specifically, the output of the acquisition device is divided into two paths: one connected to the input of the ventilator control subsystem 900, and the other connected to the input of the neuromodulated high-flow ventilation subsystem 1200. The non-invasive surface diaphragm electrical signal acquisition device 100 simultaneously provides the same respiratory drive envelope signal to both control modules.

[0225] The technical solution of using the same acquisition device avoids patient discomfort and increased costs caused by repeatedly setting up non-invasive surface diaphragm electrical signal acquisition devices 100. At the same time, it ensures that the two control modes are controlled based on the exact same respiratory drive signal, and there is no problem of inconsistent signal sources when switching modes, thus ensuring the smoothness of the switching process and the continuity of control.

[0226] This disclosure provides a neurally modulated high-flow-rate ventilation method, applied to the neurally modulated high-flow-rate ventilation subsystem 1200 in the above embodiments. For example... Figure 16 As shown, it includes the following steps:

[0227] Step S1501: Acquire respiratory drive envelope signals using a non-invasive surface diaphragm electrical signal acquisition device.

[0228] Step S1502: Based on preset conditions or user instructions, control the multi-mode respiratory support system to switch between ventilator control mode and neurally modulated high-flow ventilation mode.

[0229] The specific implementation methods of the above steps S1501 to S1502 have been described in detail in the system embodiment, and will not be repeated here to avoid repetition.

[0230] The multimodal respiratory support system 1 described in this embodiment can cover the following typical application scenarios in clinical practice:

[0231] a. After the endotracheal tube is removed, mechanically ventilated patients are immediately switched to the neuromodulated high-flow ventilation mode. The surface electrode array does not need to be reattached. The multi-mode respiratory support system 1 automatically switches the ventilation control from the ventilator control mode to the neuromodulated high-flow ventilation mode. This switching process is usually completed within 30 seconds, which can quickly achieve a seamless transition of respiratory support after endotracheal extubation.

[0232] b. For patients who require neurally modulated high-flow ventilation as first-line respiratory support upon hospitalization, it is necessary to monitor the respiratory drive envelope signal simultaneously. The trend is that when patients continue respiratory support in a neuromodulated high-flow-rate ventilation mode for 4-6 hours, the respiratory drive envelope signal... When the peak value still exceeds the preset trigger threshold, the multi-mode respiratory support system 1 issues an early warning to prompt clinicians to consider upgrading to non-invasive or invasive mechanical ventilation.

[0233] c. When the patient wears the surface electrode array and high-flow nasal cannula at night, the multimodal respiratory support system 1 operates in a neuromodulated high-flow ventilation mode, and when a respiratory drive envelope signal is detected... If the central apnea disappears for more than a preset time, the multi-mode respiratory support system 1 automatically switches to a backup ventilation mode with a fixed frequency and flow rate, based on the respiratory drive envelope signal. After recovery, it automatically switches back to neuromodulated high-flow ventilation mode.

[0234] In a typical clinical application scenario, the workflow of the multimodal respiratory support system 1 is as follows:

[0235] (1) Attach the electrode array to a specific location on the patient’s body surface and start the multi-mode respiratory support system 1.

[0236] (2) The non-invasive diaphragm electrical signal acquisition device 100 acquires multi-channel body surface biological signals in real time. After a series of processing such as signal conditioning, analog-to-digital conversion, principal component analysis dimensionality reduction, independent component analysis separation, automatic screening of diaphragm electrical components, adaptive suppression of electrocardiogram interference, envelope detection, and normalization, the normalized respiratory drive envelope signal is output.

[0237] (3) Calculate the signal quality index (SQI) in real time.

[0238] (4) When SQI ≥ preset threshold, mode switching module 400 enables ventilator control mode or neurally modulated high-flow ventilation mode according to the current preset mode or user selection. In ventilator control mode, ventilator control subsystem 900 generates gas delivery control command in real time according to respiratory drive envelope signal; in high-flow ventilation mode, neurally modulated high-flow ventilation subsystem 1200 generates target gas flow rate command in real time according to respiratory drive envelope signal.

[0239] (5) When SQI < preset threshold, the mode switching module 400 automatically switches the control mode to the traditional pressure / flow trigger mode, which is triggered by the ventilator or high flow device according to changes in airway pressure or flow to ensure the continuity of treatment.

[0240] (6) When the SQI recovers to above the preset threshold, the mode switching module 400 automatically switches the control mode back to the neural modulation mode based on the diaphragm electrical signal, and restores precise neural drive synchronous control.

[0241] The scope of protection of the multi-mode respiratory assistance method based on the electrical signal of the diaphragm on the body surface provided in this disclosure is not limited to the execution order of the steps listed in this embodiment. Any solution that adds, subtracts, or replaces steps in the prior art based on the principles of this disclosure is included within the scope of protection of this disclosure.

[0242] Those skilled in the art will recognize that the various illustrative logical blocks and steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this disclosure.

[0243] In the several embodiments provided in this disclosure, it should be understood that the disclosed systems, electronic devices, and methods can be implemented in other ways. For example, the electronic device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interface; the indirect coupling or communication connection of electronic devices or units may be electrical, mechanical, or other forms.

[0244] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0245] In addition, the functional units in the various embodiments of this disclosure can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0246] This disclosure also provides an electronic device. Figure 17 This is a schematic block diagram of an electronic device provided in an embodiment of this disclosure. Figure 17 As shown, electronic device 17 includes at least one processor 171, memory 172, at least one network interface 173, and user interface 175. The various components in electronic device 17 are coupled together via a bus system 174. It is understood that the bus system 174 is used to implement communication between these components. In addition to a data bus, the bus system 174 also includes a power bus, a control bus, and a status signal bus. However, for clarity, ... Figure 17 The general will label all buses as bus systems.

[0247] User interface 175 may include a monitor, keyboard, mouse, trackball, clicker, button, touchpad, or touchscreen.

[0248] It is understood that memory 172 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM) or programmable read-only memory (PROM), used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM) and synchronous static random access memory (SSRAM). The memories described in the embodiments of this disclosure are intended to include, but are not limited to, these and any other suitable categories of memory.

[0249] The memory 172 in this embodiment is used to store various types of data to support the operation of the electronic device 17. Examples of this data include any executable program for operation on the electronic device 17, such as the operating system 1721 and application programs 1722; the operating system 1721 contains various system programs, such as the framework layer, core library layer, driver layer, etc., for implementing various basic services and handling hardware-based tasks. The application program 1722 may contain various applications, such as a media player, browser, etc., for implementing various application services. The multimodal respiratory assistance method based on diaphragmatic electrical signals provided in this embodiment can be included in the application program 1722.

[0250] The methods disclosed in the above embodiments of this disclosure can be applied to processor 171, or implemented by processor 171. Processor 171 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in processor 171 or by instructions in the form of software. The processor 171 may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Processor 171 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this disclosure. General-purpose processor 171 may be a microprocessor or any conventional processor, etc. The steps of the accessory optimization method provided in the embodiments of this disclosure can be directly reflected as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium, which is located in memory. The processor reads the information in the memory and combines it with its hardware to complete the steps of the aforementioned method.

[0251] In an exemplary embodiment, the electronic device 17 may be used by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), or complex programmable logic devices (CPLDs) to perform the aforementioned method.

[0252] The above are merely specific embodiments of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure. Therefore, the scope of protection of this disclosure should be determined by the scope of the claims.

[0253] The above embodiments are merely illustrative of the principles and effects of this disclosure and are not intended to limit this disclosure. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this disclosure. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this disclosure should still be covered by the claims of this disclosure.

Claims

1. A multi-mode respiratory support system based on surface diaphragm electrical signals, characterized in that, include: A non-invasive diaphragmatic electroelectroelectric signal acquisition device is configured to acquire multi-channel bioelectric signals from the body surface and obtain respiratory drive envelope signals based on the multi-channel bioelectric signals from the body surface. The ventilator control subsystem, connected to the acquisition device, is configured to generate a ventilator delivery command according to a preset ventilator control strategy based on the acquired respiratory drive envelope signal. A neurally regulated high-flow ventilation subsystem, connected to the acquisition device, is configured to calculate the target gas flow rate in real time according to a preset neurally regulated high-flow control strategy based on the acquired respiratory drive envelope signal, and output the target gas flow rate command; and The mode switching module is configured to control the multi-mode respiratory support system to switch between ventilator control mode and neurally modulated high-flow ventilation mode according to preset conditions or user instructions. The ventilator control subsystem and the neuromodulated high-flow ventilation subsystem share the same non-invasive surface diaphragm electrical signal acquisition device.

2. The multi-mode respiratory support system according to claim 1, characterized in that, The non-invasive surface diaphragm electrical signal acquisition device includes: The body surface electrode array module includes a diaphragm electroacupuncture electrode array and a reference electrode; The diaphragm electroacupuncture electrode array is attached to the diaphragm surface projection area in the lower chest of the human body and is configured to collect the multi-channel surface bioelectric signals. The reference electrode is attached to a body surface location away from the diaphragm region and is configured to provide a reference potential; A multi-channel bioelectric signal acquisition module, connected to the body surface electrode array module, is configured to preprocess the multi-channel body surface bioelectric signals according to the reference potential to generate multi-channel body surface digital signals; and The signal processing module is configured to construct an ECG interference reference template and perform ECG interference cancellation on the multi-channel body surface digital signal. Then, based on blind source separation and envelope detection technology, the multi-channel body surface digital signal after ECG interference cancellation is processed to output a respiratory drive envelope signal.

3. The multi-mode respiratory support system according to claim 2, characterized in that, The diaphragm electroacupuncture electrode array consists of multiple electrodes arranged in a preset spatial configuration on a flexible substrate; wherein... The spacing between the plurality of electrodes is 10 mm to 30 mm; The preset spatial configuration includes: a ring array, a grid array, and / or a strip array along the rib direction; The flexible substrate is made of medical-grade silicone, polyurethane, or conductive fabric to adapt to the surface curvature of different body types.

4. The multi-mode respiratory support system according to claim 2, characterized in that, The signal processing module includes: an ECG interference cancellation unit, a multi-source signal separation unit, and an envelope extraction unit; wherein... The ECG interference cancellation unit is configured to construct an ECG interference reference template and estimate and remove ECG interference components in the multi-channel body surface digital signal in real time using an adaptive noise cancellation algorithm. The multi-source signal separation unit is configured to use blind source separation technology to perform multi-source signal separation on the multi-channel body surface digital signal after eliminating electrocardiogram interference components, and extract the optimal diaphragm electroindependent component. The envelope extraction unit is configured to perform envelope detection on the diaphragm electrical independent component to obtain a respiratory drive envelope signal that reflects the change of respiratory center drive intensity over time.

5. The multi-mode respiratory support system according to claim 4, characterized in that, The ECG interference cancellation unit is configured as follows: Principal component analysis is performed on the multi-channel digital body surface signal to extract multiple candidate principal component signals. The target ECG principal component signal is then selected from these candidates. Polarity correction is performed on the target ECG principal component signal to generate an ECG interference reference template and the corresponding first heartbeat peak position. as well as Peak detection is performed on the multi-channel digital signal of the body surface to determine the position of the second heartbeat peak; Based on the second heartbeat peak position, a heartbeat segment is extracted from the multi-channel body surface digital signal to construct a time window signal; Based on the ECG interference reference template and the peak position of the first heartbeat, the ECG signal of the time window is reconstructed to generate a heartbeat-by-heart ECG estimation signal; The ECG estimation signal obtained by each heartbeat is subtracted from the multi-channel digital surface signal to remove ECG interference components.

6. The multi-mode respiratory support system according to claim 4, characterized in that, The multi-source signal separation unit is configured as follows: The multi-channel digital surface signal, after eliminating ECG interference components, is subjected to digital bandpass filtering to suppress low-frequency motion artifacts and high-frequency noise; Fast independent component analysis was used to perform blind source separation on the filtered multi-channel digital body surface signal to obtain multi-channel diaphragmatic electroindependent components; Multidimensional signal features of the independent diaphragmatic electrophysiological components of each channel are extracted, and the optimal independent diaphragmatic electrophysiological components are determined based on the multidimensional signal features through a preset decision logic or a pre-trained classifier. The multidimensional signal features include one or more of the following: spectral center frequency features, respiratory cycle correlation features, and spatial distribution features.

7. The multi-mode respiratory support system according to claim 4, characterized in that, The envelope extraction unit is configured as follows: The optimal diaphragm electrical independent component is rectified to obtain the instantaneous amplitude sequence of the optimal diaphragm electrical independent component; The instantaneous amplitude sequence is low-pass filtered to extract the initial envelope signal; Baseline estimation and removal are performed on the initial envelope signal to eliminate the baseline drift component in the initial envelope signal, resulting in a drift-free envelope signal; The drift-free envelope signal is normalized to output the normalized respiratory drive envelope signal.

8. The multi-mode respiratory support system according to claim 1, characterized in that, The ventilator control subsystem includes: a ventilation control module; The ventilation control module is configured to receive the respiratory drive envelope signal and generate a ventilator delivery control command according to a preset ventilator control strategy, and control the pneumatic unit of the ventilator to deliver supportive positive pressure ventilation through the ventilator delivery control command.

9. The multi-mode respiratory support system according to claim 8, characterized in that, The ventilator control strategy includes: When the respiratory drive envelope signal rises above the inspiratory trigger threshold, an inspiratory trigger command is generated and output to control the ventilator to start delivering air. During the inspiratory phase, the instantaneous output pressure or flow level of the ventilator is determined by multiplying the real-time amplitude of the respiratory drive envelope signal with a preset gain coefficient. When the respiratory drive envelope signal drops below the expiratory trigger threshold, an inspiratory-expiratory switching command is generated and output to control the ventilator to terminate inspiratory delivery.

10. The multi-mode respiratory support system according to claim 8, characterized in that, The ventilation control module also includes: a signal quality assessment unit; The signal quality assessment unit is configured to assess the reliability of the respiratory drive envelope signal in real time and calculate the signal quality index. If the signal quality index remains below the signal quality threshold for an extended period of time, the ventilator's delivery trigger mode will be switched from the control mode based on the respiratory drive envelope signal to a pressure trigger mode and / or a flow trigger mode.

11. The multi-mode respiratory support system according to claim 1, characterized in that, The neurally modulated high-flow ventilation subsystem includes: a high-flow rate control module and an open-type gas delivery module; The high flow rate control module is configured to receive the respiratory drive envelope signal, calculate the target gas flow rate in real time according to the preset neural regulation high flow rate control strategy, and output the target gas flow rate command, thereby controlling the high flow rate delivery unit to deliver high flow rate gas through the target gas flow rate command. The open gas delivery module includes an airway pressure sensor configured to monitor airway pressure in real time. The high-flow-rate control module is also configured to adjust the target gas flow rate command based on the airway pressure.

12. The multi-mode respiratory support system according to claim 11, characterized in that, The high-flow-rate control module is configured as follows: The output gas flow rate of the high-flow-rate delivery unit is controlled to maintain the baseline airflow. Acquiring the respiratory drive envelope signal and calculating the target gas flow rate in real time according to a preset neural-regulated high-flow control strategy includes: comparing the respiratory drive envelope signal with a preset trigger threshold; when the respiratory drive envelope signal exceeds the trigger threshold, increasing the output gas flow rate of the high-flow delivery unit according to a gas flow rate regulation formula, and generating the target gas flow rate command that matches the current respiratory drive envelope signal; when the respiratory drive envelope signal drops below a preset switching threshold, controlling the output gas flow rate of the high-flow delivery unit to smoothly return to the baseline airflow, and generating the target gas flow rate command.

13. The multi-mode respiratory support system according to claim 12, characterized in that, The high-flow-rate control module also includes a pressure closed-loop control unit; The pressure closed-loop control unit is configured to generate a target pressure waveform based on the respiratory drive envelope signal; The correction amount for the target gas flow rate is calculated in real time based on the deviation between the target pressure waveform and the airway pressure. The target gas flow rate command is adjusted according to the correction amount so that the airway pressure tracks the target pressure waveform.

14. A multimodal respiratory assistance method based on surface diaphragmatic electrical signals, characterized in that, include: Respiratory drive envelope signals were acquired using a non-invasive surface diaphragm electrical signal acquisition device. Based on preset conditions or user instructions, control the multi-mode respiratory support system to switch between ventilator control mode and neurally modulated high-flow ventilation mode; In the ventilator control mode, based on the acquired respiratory drive envelope signal, a ventilator delivery command is generated through the ventilator control subsystem according to a preset ventilator control strategy. In the neurally regulated high-flow ventilation mode, based on the acquired respiratory drive envelope signal, the target gas flow rate is calculated in real time through the neurally regulated high-flow ventilation subsystem according to a preset neurally regulated high-flow control strategy, and a target gas flow rate command is output.

15. An electronic device, characterized in that, The electronic device includes: The memory is configured to store executable programs; A processor is configured to execute the program to cause the electronic device to perform the method of claim 14.

Images