Intelligent breathing machine and intelligent ventilation control system and method based on breathing sound monitoring

Abstract

The embodiment of the invention provides an intelligent breathing machine and an intelligent ventilation control system and method based on breathing sound monitoring. The system comprises a breath sound acquisition unit which comprises at least one trachea sound sensor and at least two chest wall sound sensors and is used for synchronously acquiring breath sound signals of a user; the preprocessing unit is used for preprocessing the breathing sound signals; the central processing unit is used for processing the preprocessed breathing sound signal to obtain the abnormal breathing sound type and confidence of the user; and the breathing machine controller is used for dynamically adjusting ventilation parameters of the breathing machine according to the abnormal breathing sound type and the confidence coefficient. The breathing sound signals of the patient are collected and analyzed in real time, ventilation parameters are dynamically adjusted, and the problems of signal lag and poor adaptability in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent medical devices, and particularly to an intelligent ventilator and an intelligent ventilation control system and method based on respiratory sound monitoring. Background Art

[0002] Mechanical ventilation refers to a medical intervention measure that uses medical devices such as ventilators to maintain the patency of a patient's airway and improve the patient's oxygenation effect, so as to reduce the patient's respiratory failure. The ventilation modes of existing ventilators mainly rely on preset parameters (such as tidal volume, respiratory rate) or patient physiological signals (such as airway pressure, blood oxygen saturation) for control, lacking real-time and direct perception of the pulmonary pathophysiological state. When a patient has airway spasm, atelectasis or secretion retention, medical staff must judge and manually adjust the ventilation parameters through intermittent auscultation or imaging examination, which has problems such as response lag, heavy workload and human-machine differences. Existing studies have shown that different abnormal breath sounds (wheezing, crackles, rhonchi, pleural friction rub) are closely related to parameters such as airway resistance and alveolar recruitment. If respiratory sounds can be collected and identified in real time and automatically fed back into the ventilator control logic, true "ventilation according to lung conditions" can be achieved.

[0003] For example, Patent CN118490944A discloses a ventilator and its ventilation mode control method, which calculates the drive ratio corresponding to the drive parameters between the ventilator and the user, sets the pressure adjustment parameters that change with the drive ratio, and then calculates the pressure adjustment amount corresponding to the next respiratory cycle according to the tidal volume deviation corresponding to the current respiratory cycle. In addition, Patent CN118022114A discloses a control method for the duration of mechanical ventilation, which sequentially performs blood oxygen saturation detection, inhaled oxygen concentration detection, pH value detection, spontaneous breathing detection, and tolerance monitoring when the ventilator operates in a preset ventilation mode according to preset ventilation parameters, and judges whether the condition for ending mechanical ventilation is reached. In addition, the literature "Research on Key Technologies of Intelligent Ventilation of Ventilators" points out that existing ventilators still have delays in synchronization and cannot respond to the patient's spontaneous breathing needs in real time.

[0004] However, the above-mentioned existing technologies have the following defects: 1. Signal lag: Existing ventilation technologies comprehensively adjust according to physiological parameters such as pulse oxygen saturation and carbon dioxide partial pressure and ventilator mechanical ventilation parameters. However, the human blood gas model corresponding to the former has a certain lag and cannot directly correspond to the state of the breathing process, which is likely to cause a delay in ventilation mode adjustment. In addition, due to the strong compensatory ability of the respiratory system, the oxygenation parameters cannot predict the change of the breathing state, and it is easy for the patient's breathing state to decline sharply and the ventilator ventilation mode fails to make compensation in advance.

[0005] 2. Insufficient adaptability: Currently, the preset parameters of the ventilator cannot dynamically match the instantaneous changes in the patient's respiratory state (such as coughing, wheezing, etc.), and the corresponding air pressure and flow adjustment modes cannot handle the abnormal sudden changes in respiratory parameters (such as events like patient movement, tracheal detachment, sudden onset of diseases, etc.), resulting in a conflict between mechanical ventilation and the actual respiratory state of the human body, causing discomfort and even air pressure damage. Summary of the Invention

[0006] The present invention aims to solve at least one of the technical problems existing in the prior art, and provides an intelligent ventilator, an intelligent ventilation control system and method based on respiratory sound monitoring.

[0007] In a first aspect, an embodiment of the present invention provides an intelligent ventilation control system for a ventilator based on respiratory sound monitoring, including: A respiratory sound acquisition unit, including at least one tracheal sound sensor and at least two chest wall sound sensors, for synchronously acquiring the respiratory sound signals of the user; A preprocessing unit, for preprocessing the respiratory sound signals; A central processing unit, for processing the preprocessed respiratory sound signals to obtain the abnormal respiratory sound types and confidence levels of the user; A ventilator controller, for dynamically adjusting the ventilation parameters of the ventilator according to the abnormal respiratory sound types and confidence levels.

[0008] In some possible embodiments, the preprocessing unit includes an analog front end and an analog-to-digital conversion circuit, for performing band-pass filtering on the respiratory sound signals in the range of 50 Hz to 2000 Hz and digitally processing them at a sampling rate of not less than 4 kHz.

[0009] In some possible embodiments, the central processing unit is specifically configured to: Perform spectral analysis on the preprocessed respiratory sound signals based on a spectral analysis algorithm to obtain the time-frequency domain characteristic spectrogram corresponding to the respiratory sound signals; Perform unsupervised classification processing on the time-frequency domain characteristic spectrogram based on the FCM clustering algorithm to obtain an effective respiratory sound interval spectrogram; Identify the time-frequency characteristics corresponding to the effective respiratory sound interval spectrogram based on a time-frequency domain graph convolutional network GCN classification model, and output the abnormal respiratory sound types and confidence levels.

[0010] In some possible embodiments, the central processing unit is specifically further configured to: perform multi-label classification on the effective respiratory sound interval spectrogram based on the GCN classification model, and output the confidence levels including wheezing, moist rales, dry rales, pleural friction rubs and normal respiratory sounds.

[0011] In some possible embodiments, the ventilation parameters at least include: respiratory rate, inspiratory-expiratory ratio, positive end-expiratory pressure (PEEP), tidal volume, and peak flow rate.

[0012] In some possible embodiments, the types of abnormal breath sounds include wheezing, moist rales, dry rales, and pleural friction rub; the ventilator controller is further specifically configured to: When wheezing is detected, extend the expiratory time in a preset gradient and decrease the respiratory rate; When moist rales are detected, increase the PEEP in a fixed step until the intensity of the moist rales is lower than the threshold or reaches the safety upper limit; When dry rales are detected, trigger a suction alarm and optionally perform tidal volume impact ventilation; When pleural friction rub is detected, automatically decrease the tidal volume and limit the plateau pressure.

[0013] In some possible embodiments, the ventilator controller is further specifically configured to: Calculate the confidence level and duration of various types of abnormal breath sounds in real time; Based on a preset clinical hazard level weight table, perform a weighted severity score on multiple co-existing abnormal breath sounds; Determine the adjustment priority of the ventilation parameters according to the scoring result, and implement a hierarchical intervention strategy.

[0014] In some possible embodiments, it further includes a human-machine interaction unit; The human-machine interaction unit is used to display a real-time spectrogram, abnormal breath sound prompts, and parameter adjustment information.

[0015] In some possible embodiments, it further includes a safety monitoring unit; The safety monitoring unit is used to automatically deactivate the closed-loop regulation and alarm when a sensor failure or parameter exceedance is detected.

[0016] In a second aspect, an embodiment of the present invention provides a method for intelligent ventilation control of a ventilator based on breath sound monitoring, including the following steps: Synchronously collect the breath sound signals of the user through at least one tracheal sound sensor and at least two chest wall sound sensors; Preprocess the breath sound signals; Process the preprocessed breath sound signals to obtain the types and confidence levels of the abnormal breath sounds of the user; Dynamically adjust the ventilation parameters of the ventilator according to the types and confidence levels of the abnormal breath sounds.

[0017] In a third aspect, an embodiment of the present invention provides an intelligent ventilator, including the intelligent ventilation control system of the ventilator based on breath sound monitoring described above.

[0018] In a fourth aspect, an embodiment of the present invention provides an electronic device, including: One or more processors; A storage unit for storing one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement the intelligent ventilation control method of a ventilator based on breath sound monitoring described above.

[0019] In a fifth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it can implement the intelligent ventilation control method of a ventilator based on breath sound monitoring described above.

[0020] The beneficial effects of the intelligent ventilator and the intelligent ventilation control system and method based on breath sound monitoring according to the embodiments of the present invention are as follows: 1. Improvement in evaluation accuracy: A breath dynamics evaluation model with strong generalization ability is constructed based on the breath sound signal spectrum feature analysis technology, realizing non-invasive high-precision evaluation of the respiratory system and being able to directly feedback the ventilation quality of the respiratory system; 2. Improvement in time accuracy: The breath dynamics evaluation model based on breath sound can synchronously respond to the instantaneous changes of physiological characteristics such as lung volume and airway pressure, with a faster response speed compared to traditional methods; 3. Strong dynamic adaptability: The breath pattern classifier based on a deep convolutional neural network can accurately identify normal breath patterns and abnormal breath events (including coughing, wheezing, apnea, etc.), and give dynamic adjustment suggestions according to normal and abnormal breath patterns; 4. Strong anti-noise ability: It is relatively less affected by non-ideal factors such as environmental noise under the condition of non-invasive physiological monitoring, and has higher monitoring accuracy and lower operation difficulty compared to methods such as carbon dioxide partial pressure detection. Description of the Drawings

[0021] Figure 1 It is a schematic structural diagram of an exemplary electronic device for implementing the intelligent ventilation control method of a ventilator based on breath sound monitoring according to an embodiment of the present invention; Figure 2 It is a conceptual schematic diagram of the intelligent ventilation control of a ventilator based on breath sound monitoring according to another embodiment of the present invention; Figure 3 It is a schematic structural diagram of the intelligent ventilation control system of a ventilator based on breath sound monitoring according to another embodiment of the present invention; Figure 4 It is a schematic structural diagram of the intelligent ventilation control system of a ventilator based on breath sound monitoring according to another embodiment of the present invention; Figure 5Flowchart of breath sound acquisition and signal processing for another embodiment of the present invention; Figure 6 Example spectrogram of abnormal breath sound recognition result for another embodiment of the present invention; Figure 7 Schematic diagram of closed-loop parameter adjustment logic state machine for another embodiment of the present invention; Figure 8 Flowchart of intelligent ventilation control method for a ventilator based on breath sound monitoring for another embodiment of the present invention; Figure 9 Flowchart of intelligent ventilation control method for a ventilator based on breath sound monitoring for another embodiment of the present invention; Figure 10 Schematic diagram of the structure of an intelligent ventilator for another embodiment of the present invention. Detailed implementation manners

[0022] Next, the technical solutions of the present invention will be clearly and completely described in conjunction with the accompanying drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0023] Figure 1 Schematic diagram of the structure of an example electronic device for implementing an intelligent ventilation control method for a ventilator based on breath sound monitoring according to an embodiment of the present invention. As Figure 1 shown, the electronic device 100 includes one or more processors 110, one or more storage devices 120, one or more input devices 130, one or more output devices 140, etc., and these components are interconnected through a bus system 150 and / or other forms of connection mechanisms. It should be noted that Figure 1 the components and structure of the shown electronic device are exemplary and not restrictive. According to needs, the electronic device may also have other components and structures.

[0024] The processor 110 may be a central processing unit (CPU), or may be composed of multiple processing cores, or have other forms of processing units with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 100 to perform desired functions.

[0025] The storage device 120 may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage media, and the processor may run the program instructions to implement the client functions (implemented by the processor) and / or other desired functions in the embodiments of the present disclosure described below. Various application programs and various data may also be stored in the computer-readable storage media. For example, various data used and / or generated by the application programs, etc.

[0026] The input device 130 may be a device used by a user to input instructions, and may include one or more of a keyboard, a mouse, a microphone, a touch screen, etc.

[0027] The output device 140 may output various information (such as images or sounds) to the outside (such as the user), and may include one or more of a display, a speaker, etc.

[0028] Figure 2 It is a conceptual schematic diagram of a ventilator intelligent ventilation control based on breath sound monitoring according to another embodiment of the present invention. Figure 3 It is a structural schematic diagram of a ventilator intelligent ventilation control system based on breath sound monitoring according to another embodiment of the present invention. Figure 4 It is a structural schematic diagram of a ventilator intelligent ventilation control system based on breath sound monitoring according to another embodiment of the present invention.

[0029] As Figures 2 to 4 shown, the embodiment of the present invention relates to a ventilator intelligent ventilation control system based on breath sound monitoring, including: a breath sound acquisition unit 301, a preprocessing unit 302, a central processing unit 303, and a ventilator controller 304.

[0030] Specifically, as Figure 3 and Figure 4 shown, the breath sound acquisition unit 301 includes at least one tracheal sound sensor and at least two chest wall sound sensors, and is used to synchronously acquire the breath sound signals of the user. The preprocessing unit 302 is used to preprocess the breath sound signals; the central processing unit 303 is used to process the preprocessed breath sound signals to obtain the abnormal breath sound types and confidence levels of the user; the ventilator controller 304 is used to dynamically adjust the ventilation parameters of the ventilator according to the abnormal breath sound types and confidence levels.

[0031] The beneficial effects of the intelligent ventilation control system of the ventilator based on breath sound monitoring according to the embodiments of the present invention are as follows: 1. Improvement in evaluation accuracy: A respiratory dynamics evaluation model with strong generalization ability is constructed based on the spectral feature analysis technology of breath sound signals, realizing non-invasive and high-precision evaluation of the respiratory system, and being able to directly feedback the ventilation quality of the respiratory system; 2. Improvement in time accuracy: The respiratory dynamics evaluation model based on breath sound can synchronously respond to the instantaneous changes of physiological characteristics such as lung volume and airway pressure, and the response speed is improved compared with traditional methods; 3. Strong dynamic adaptability: The respiratory pattern classifier based on the deep convolutional neural network can accurately identify normal respiratory patterns and abnormal respiratory events (including coughing, wheezing, apnea, etc.), and give dynamic adjustment opinions according to normal and abnormal respiratory patterns; 4. Strong anti-noise ability: It is relatively less affected by non-ideal factors such as environmental noise under the condition of non-invasive physiological monitoring, and has higher monitoring accuracy and lower operation difficulty compared with methods such as carbon dioxide partial pressure detection.

[0032] Exemplarily, as Figure 3 and Figure 4 shown, the preprocessing unit 302 includes an analog front end and an analog-to-digital conversion circuit, which is used to perform band-pass filtering on the breath sound signal in the range of 50 Hz to 2000 Hz and digitize it at a sampling rate of not less than 4 kHz. Specifically, in this embodiment, a high-sensitivity acceleration contact microphone sensor is used, and the MCU compensation algorithm is used to make up for the bandwidth defect of the accelerometer and the tape coupling strength defect, and the wide-band real breath sound signal of the patient is collected in real time; the breath sound signal is time-synchronized with the airway pressure / flow signal.

[0033] Exemplarily, in combination with Figure 3 , Figure 4 and Figure 5 , Figure 5 is the flowchart of breath sound acquisition and signal processing of another embodiment of the present invention. The central processor 303 is specifically used for: performing spectral analysis on the preprocessed breath sound signal based on the spectral analysis algorithm to obtain the time-frequency domain characteristic spectrogram corresponding to the breath sound signal; performing unsupervised classification processing on the time-frequency domain characteristic spectrogram based on the FCM clustering algorithm to obtain an effective breath sound interval spectrogram; identifying the time-frequency characteristics corresponding to the effective breath sound interval spectrogram based on the time-frequency domain graph convolutional network GCN classification model, and outputting the type and confidence of abnormal breath sounds. Among them, an exemplary spectrogram of the identified result of abnormal breath sounds is as Figure 6 shown.

[0034] In some embodiments, the central processing unit 303 implements signal filtering and performs spectral analysis algorithms such as short-time Fourier transform, wavelet transform, and Mel frequency spectrum transform; uses FCM clustering to identify the effective signal interval, and uses the AIC criterion for accurate picking to achieve real-time signal monitoring with low algorithm complexity; uses a graph convolutional network (GCN) classification model in the time-frequency domain to identify respiratory sound features (such as inspiratory phase, expiratory phase, abnormal sounds), and extracts key parameters such as respiratory rate, respiratory intensity, and airway resistance.

[0035] Exemplarily, such as Figure 3 and Figure 4 shown, the central processing unit 303 is specifically further configured to: perform multi-label classification on the spectrogram of the effective respiratory sound interval based on the GCN classification model, and output the confidence levels including wheezing, crackles, rhonchi, pleural friction rub, and normal breath sounds.

[0036] In some embodiments, the ventilation parameters at least include: respiratory rate, inspiratory-expiratory ratio, positive end-expiratory pressure PEEP, tidal volume, and peak flow rate.

[0037] Exemplarily, such as Figure 3 and Figure 4 shown, the ventilator controller 304 is specifically further configured to: when wheezing is detected, extend the expiratory time and decrease the respiratory rate according to a preset gradient; when crackles are detected, increase the PEEP by a fixed step until the intensity of the crackles is lower than the threshold or reaches the safety upper limit; when rhonchi are detected, trigger a suction alarm and optionally perform tidal volume impact ventilation; when pleural friction rub is detected, automatically decrease the tidal volume and limit the plateau pressure.

[0038] In some embodiments, in combination with Figure 7 , Figure 7 is a schematic diagram of the closed-loop parameter adjustment logic state machine of another embodiment of the present invention. The ventilator controller 304 dynamically switches the ventilation mode (such as assist-control ventilation, pressure support ventilation) according to the respiratory sound features, and adjusts parameters such as tidal volume and inspiratory-expiratory ratio in real time through a closed-loop feedback system to ensure man-machine synchronization and quickly adjust the ventilation pressure in real time for monitoring sudden abnormal times.

[0039] Wheezing: Extend expiration, relatively decrease the respiratory rate, shorten the inspiratory time, and the ratio can be adjusted to 1:3 - 1:4; Crackles: Gradually increase the positive end-expiratory pressure (PEEP) in stages, +2 cm H2O each time until the crackles are significantly weakened; Rhonchi: Issue a suction alarm and select airway impact ventilation; Pleural friction rub: Automatically decrease the tidal volume, limit the plateau pressure, and increase the synchronous trigger sensitivity.

[0040] Exemplarily, such as Figure 3 and Figure 4As shown, the ventilator controller 304 is further specifically configured to: calculate the confidence and duration of various abnormal breath sounds in real time; perform a weighted severity score on multiple co-existing abnormal breath sounds based on a preset clinical hazard level weight table; determine the ventilation parameter adjustment priority according to the scoring result, and execute a hierarchical intervention strategy.

[0041] Exemplarily, as Figure 3 and Figure 4 shown, the system further includes a human-machine interaction unit 305; the human-machine interaction unit 305 is used to display real-time spectrograms, abnormal breath sound prompts, and parameter adjustment information. Specifically, the human-machine interaction unit 305 can implement real-time spectrograms, color marking of abnormal types; adjustment suggestion pop-up windows and confirmation logic; one-key switching between automatic / manual, log and historical trend tracking, etc.

[0042] Exemplarily, as Figure 3 and Figure 4 shown, it further includes a safety monitoring unit 306; the safety monitoring unit 306 is used to automatically deactivate the closed-loop adjustment and alarm when a sensor failure or parameter over-limit is detected.

[0043] Based on the same inventive concept, an embodiment of the present invention provides a method for intelligent ventilation control of a ventilator based on breath sound monitoring. This control method can be applied to the control system described above. For details, reference can be made to the relevant descriptions above and will not be elaborated here.

[0044] Figure 8 is a flowchart of the method for intelligent ventilation control of a ventilator based on breath sound monitoring according to another embodiment of the present invention. Figure 9 is a flowchart of the method for intelligent ventilation control of a ventilator based on breath sound monitoring according to another embodiment of the present invention. As Figure 8 and Figure 9 shown, an embodiment of the present invention relates to a method for intelligent ventilation control of a ventilator based on breath sound monitoring. This method includes the following steps S801 to step S804: Step S801, synchronously collect the breath sound signals of the user through at least one tracheal sound sensor and at least two chest wall sound sensors.

[0045] Step S802, preprocess the breath sound signals.

[0046] Step S803, process the preprocessed breath sound signals to obtain the abnormal breath sound types and confidence levels of the user.

[0047] Step S804, dynamically adjust the ventilation parameters of the ventilator according to the abnormal breath sound types and confidence levels.

[0048] The beneficial effects of the intelligent ventilation control method of the ventilator based on breath sound monitoring in the embodiments of the present invention are as follows: 1. Improved evaluation accuracy: A respiratory dynamics evaluation model with strong generalization ability is constructed based on the spectral feature analysis technology of breath sound signals, realizing non-invasive high-precision evaluation of the respiratory system and being able to directly feedback the ventilation quality of the respiratory system; 2. Improved time accuracy: The respiratory dynamics evaluation model based on breath sound can synchronously respond to the instantaneous changes of physiological characteristics such as lung volume and airway pressure, with a faster response speed than traditional methods; 3. Strong dynamic adaptability: The respiratory pattern classifier based on deep convolutional neural network can accurately identify normal respiratory patterns and abnormal respiratory events (including cough, wheezing, apnea, etc.), and give dynamic adjustment suggestions according to normal and abnormal respiratory patterns; 4. Strong anti-noise ability: It is relatively less affected by non-ideal factors such as environmental noise under the condition of non-invasive physiological monitoring, and has higher monitoring accuracy and lower operation difficulty compared with methods such as carbon dioxide partial pressure detection.

[0049] In some embodiments, the parameter adjustment includes: wheezing triggering extended exhalation and reduced respiratory rate, crackles triggering increased PEEP, rhonchi triggering sputum suction reminder, and pleural friction rub triggering reduced tidal volume.

[0050] In some embodiments, the adjustment amplitude of all parameters is limited by a safety threshold and recorded in an unchangeable log. In addition, in some embodiments, a function of one-key switching between automatic mode and manual mode is provided.

[0051] In some embodiments, a real-time spectrogram, abnormal determination, and parameter adjustment suggestions are also displayed through a human-machine interface. And the processing delay of the central processing unit does not exceed 50 ms, so as to complete the acquisition-determination-control closed loop within a single respiratory cycle.

[0052] Based on the same inventive concept, as Figure 10 shown, the embodiments of the present invention also provide an intelligent ventilator, which includes the intelligent ventilation control system of the ventilator based on breath sound monitoring described above. For details, reference can be made to the relevant descriptions above and will not be elaborated here.

[0053] Based on the same inventive concept, the embodiments of the present invention provide a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it can implement the offline page development method based on the low-code development platform described above.

[0054] Among them, the computer-readable medium can be included in the devices, equipment, and systems of the present disclosure, or can exist alone.

[0055] Among them, the computer-readable storage medium can be any tangible medium that contains or stores a program. It can be an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device. More specific examples include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, an optical fiber, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0056] Among them, the computer-readable storage medium can also include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Specific examples include, but are not limited to, electromagnetic signals, optical signals, or any suitable combination thereof.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent ventilation control system for a ventilator based on respiratory sound monitoring, characterized in that, Comprising: A breath sound acquisition unit, including at least one tracheal sound sensor and at least two chest wall sound sensors, for synchronously acquiring the breath sound signals of a user; A preprocessing unit, for preprocessing the breath sound signals; A central processing unit, for processing the preprocessed breath sound signals to obtain the abnormal breath sound type and confidence level of the user; A ventilator controller, for dynamically adjusting the ventilation parameters of the ventilator according to the abnormal breath sound type and confidence level.

2. The intelligent ventilation control system of a ventilator based on breath sound monitoring according to claim 1, wherein The preprocessing unit includes an analog front end and an analog-to-digital conversion circuit, for performing band-pass filtering on the breath sound signals in the range of 50 Hz to 2000 Hz and digitally processing them at a sampling rate not lower than 4 kHz.

3. The intelligent ventilation control system of a ventilator based on breath sound monitoring according to claim 1, wherein The central processing unit is specifically used for: Performing spectral analysis on the preprocessed breath sound signals based on a spectral analysis algorithm to obtain the time-frequency domain characteristic spectrogram corresponding to the breath sound signals; Performing unsupervised classification processing on the time-frequency domain characteristic spectrogram based on the FCM clustering algorithm to obtain an effective breath sound interval spectrogram; Identifying the time-frequency characteristics corresponding to the effective breath sound interval spectrogram based on a time-frequency domain graph convolutional network GCN classification model, and outputting the abnormal breath sound type and confidence level.

4. The intelligent ventilation control system of a ventilator based on breath sound monitoring according to claim 3, wherein, The central processing unit is specifically further used for: performing multi-label classification on the effective breath sound interval spectrogram based on the GCN classification model, and outputting the confidence levels including wheezing, crackles, rhonchi, pleural friction rub, and normal breath sounds.

5. The intelligent ventilation control system of a ventilator based on breath sound monitoring according to any one of claims 1 to 4, characterized in that, The ventilation parameters at least include: respiratory rate, inspiratory-expiratory ratio, positive end-expiratory pressure PEEP, tidal volume, and peak flow rate.

6. The intelligent ventilation control system of a ventilator based on breath sound monitoring according to any one of claims 1 to 4, characterized in that, The abnormal breath sound type includes wheezing, crackles, rhonchi, pleural friction rub; the ventilator controller is specifically further used for: When detecting wheezing, extending the expiratory time according to a preset gradient and reducing the respiratory rate; When detecting crackles, increasing the PEEP by a fixed step until the crackle intensity is lower than the threshold or reaches the safety upper limit; When detecting rhonchi, triggering a sputum suction alarm and optionally performing tidal volume impact ventilation; When detecting pleural friction rub, automatically reducing the tidal volume and limiting the plateau pressure.

7. The intelligent ventilation control system of a ventilator based on breath sound monitoring according to any one of claims 1 to 4, characterized in that, The ventilator controller is specifically further used for: Calculating the confidence level and duration of various abnormal breath sounds in real time; Based on a preset clinical hazard level weight table, performing a weighted severity score on multiple co-existing abnormal breath sounds; Determining the ventilation parameter adjustment priority according to the scoring result, and implementing a hierarchical intervention strategy.

8. The intelligent ventilation control system of a ventilator based on breath sound monitoring according to any one of claims 1 to 4, characterized in that It further includes a human-computer interaction unit; The human-computer interaction unit is used for displaying the real-time spectrogram, abnormal breath sound prompt, and parameter adjustment information.

9. The intelligent ventilation control system of a ventilator based on breath sound monitoring according to any one of claims 1 to 4, characterized in that, It further includes a safety monitoring unit; The safety monitoring unit is used for automatically deactivating the closed-loop regulation and alarming when detecting sensor failure or parameter overrun.

10. An intelligent ventilation control method for a ventilator based on breath sound monitoring, characterized in that, Including the following steps: Synchronously acquiring the breath sound signals of a user through at least one tracheal sound sensor and at least two chest wall sound sensors; Preprocessing the breath sound signals; Processing the preprocessed breath sound signals to obtain the abnormal breath sound type and confidence level of the user; Dynamically adjusting the ventilation parameters of the ventilator according to the abnormal breath sound type and confidence level.

11. An intelligent ventilator, characterized in that, Including the intelligent ventilation control system for a ventilator based on breath sound monitoring according to any one of claims 1 to 9.

Images