Performance evaluation system and method for breathing machine

Abstract

The invention discloses a performance evaluation system and method for a breathing machine, and particularly relates to the field of data processing, the system comprises a sensor module, a control module, an execution module and a performance evaluation module; the sensor module is used for collecting blood oxygen saturation and respiratory rate of a patient; the control module intelligently selects a ventilation mode and dynamically adjusts the inspiration and expiration time ratio on the basis of blood oxygen saturation and respiratory frequency data and a biological feedback mechanism; the execution module adjusts the ratio of inspiration to expiration by controlling the opening and closing time of airflow; the performance evaluation module constructs a simulation model through MATLAB and Simulink, and evaluates the accuracy of the simulation model through comparison with hardware test data. The ventilation mode can be adaptively adjusted according to the state of the patient, and the comfort of the patient is improved through non-invasive monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and more particularly to a performance evaluation system and method for a ventilator. Background Art

[0002] With the continuous advancement of medical technology, the role of ventilators in the treatment of acute and chronic respiratory failure has become increasingly prominent.

[0003] Existing public literature 1 (Predictive value of MoCA score in weaning patients with mechanical ventilation in ICU, 2024) explores the predictive value of MoCA score in weaning patients with mechanical ventilation in ICU and finds the optimal critical value. Figure 2 General and clinical data (such as MoCA score, minute ventilation volume, Apache-II score, etc.) were collected within 24 hours of mechanical ventilation, and ventilation duration and weaning results were recorded before the spontaneous breathing trial. Binary logistic regression analysis was used to screen factors associated with weaning failure, and the predictive efficacy of the MoCA score was evaluated using a receiver operating characteristic (ROC) curve. However, mechanical ventilation cannot dynamically adjust ventilation parameters according to the patient's real-time physiological needs, leading to insufficient ventilation or ventilation problems.

[0004] Existing public literature 2 (Non-invasive ventilator fan drive circuit system design, 2021), see Figure 3 The designed fan driver module circuit was integrated into the ventilator system circuit and equipped with a test platform. In the design of the noninvasive ventilator fan driver circuit system, the driver circuit adopts a half-bridge topology, consisting of a half-bridge driver IR2013S and a MOSFET NTDV20N06L, to directly drive the fan motor. Pins 7 (H-OUT) and 5 (L-OUT) of the IR2013S control the switching states of the high-side and low-side MOSFETs, respectively. These outputs are controlled by the input signals at pins 35 (H-IN) and 3 (L-IN). The high-side output (H-OUT) and low-side output (L-OUT) are connected to the gates of the MOSFET pairs (Q3 and Q6, Q4 and Q7, and Q5 and Q8) through 47Ω current-limiting resistors to limit gate current and protect the devices. The drains of MOSFETs Q3, Q4, and Q5 are connected to the sources of Q6, Q7, and Q8, respectively, and then to pin 6 (VS) of the driver, forming a half-bridge structure to achieve precise control of the motor. However, the ventilator integrated with the fan drive module circuit lacks the automatic adjustment function based on the patient's respiratory status. Although it can monitor basic physiological parameters, its feedback mechanism is relatively simple and cannot achieve comprehensive optimization adjustment of ventilation mode and inspiration and expiration ratio.

[0005] Therefore, there is an urgent need for an evaluation system and method that can automatically select ventilation modes by collecting patients' physiological data in real time and combining feedback mechanisms to improve ventilator performance and patient comfort. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a performance evaluation system and method for a ventilator, which collects the patient's blood oxygen saturation and respiratory rate, intelligently selects the ventilation mode based on the biofeedback mechanism and dynamically adjusts the time ratio of inspiration and exhalation, combines the solenoid valve to control the airflow output, and then constructs a simulation model through MATLAB and Simulink to simulate the respiratory dynamics under real physiological conditions. The model is compared and verified with the hardware test data to evaluate the system performance, improve the reliability and accuracy of the system, and solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A performance evaluation system for a ventilator includes a sensor module, a control module, an execution module, and a performance evaluation module. The sensor module is used to collect a patient's blood oxygen saturation and respiratory rate. The control module intelligently selects a ventilation mode based on the blood oxygen saturation and respiratory rate data using a biofeedback mechanism, dynamically adjusting the ratio of inhalation to exhalation time. The execution module adjusts the inhalation to exhalation ratio by controlling the on and off time of the airflow. The performance evaluation module constructs a simulation model using MATLAB and Simulink, and evaluates its accuracy by comparing it with hardware test data.

[0009] The blood oxygen saturation level is an indicator for evaluating the patient's oxygenation status. Its normal range is above 94%. A level below 94% indicates the risk of hypoxemia. The respiratory rate is a parameter for evaluating the degree of shortness of breath. The respiratory rate of a healthy adult is 12-20 times / minute. A level above 15 times / minute indicates the risk of compensatory breathing. The biofeedback mechanism sets blood oxygen saturation <94% or respiratory rate >15bpm as the judgment criteria for initiating enhanced ventilation support: when blood oxygen saturation >94% The system recognizes that the patient is in a non-critical condition and adopts continuous positive pressure ventilation mode, that is, the ratio of inspiration and expiration is 1:2; when the blood oxygen saturation is ≤94% and the respiratory rate is >15 times / minute, the system recognizes that the patient is in a critical condition and continues to use I:E=1:2, reducing the total respiratory cycle time TCT by controlling the pulse width modulation signal frequency, that is, TCT is less than 4 seconds; when the blood oxygen saturation is ≤94% and the respiratory rate is 15 times / minute, the system recognizes that the patient is in a mild critical condition, and the ratio of inspiration and expiration is 1:4.

[0010] As a further solution of the present invention, the sensor module is used to collect the patient's blood oxygen saturation and respiratory rate, including the following specific contents: the sensor module includes a pulse oximeter and a bioelectric signal acquisition unit, the pulse oximeter is used to non-invasively monitor the patient's blood oxygen saturation (SpO2), and the bioelectric signal acquisition unit is used to collect respiratory waveform data to extract the respiratory rate (RR). The pulse oximeter is a high-precision device based on optical sensing technology, which calculates the SpO2 level through photoplethysmography (PPG). The bioelectric signal acquisition unit uses impedance spirometry to record the breathing pattern by measuring the impedance changes of the chest caused by respiratory activity; the bioelectric signal acquisition unit uses an analog front-end (AFE) chip to capture the tiny electrical signals caused by changes in chest volume, converts them into digital signals, and extracts the respiratory rate (RR) through a signal processing algorithm.

[0011] As a further solution of the present invention, the control module intelligently selects the ventilation mode based on the blood oxygen saturation and respiratory rate data and the biofeedback mechanism, and dynamically adjusts the ratio of inspiration and expiration time, including the following specific contents: the control module first filters the blood oxygen saturation (SpO2) and respiratory rate (RR) data collected by the sensor module to remove noise interference. The blood oxygen saturation is collected by the pulse oximeter, which may produce baseline drift or high-frequency noise due to ambient light, patient movement or poor skin contact. Therefore, the control module uses a moving average filter to smooth the photoplethysmography (PPG) signal with a sampling frequency of 125Hz; the respiratory rate data is collected by the bioelectric signal acquisition unit through impedance spirometry, recording the impedance changes of the chest caused by breathing. Since the respiratory waveform may be interfered with by muscle activity or other electrical signals, the control module uses a low-pass filter (LPF) to remove high-frequency noise and combines it with a high-pass filter (HPF) to eliminate baseline drift, thereby extracting a clear respiratory signal.

[0012] During the data analysis stage, the control module uses the biofeedback mechanism to determine the patient's respiratory status and automatically adjust the ventilation mode. The blood oxygen saturation (SpO2) level is an indicator for evaluating the patient's oxygenation status. Its normal range is above 94%. Below 94% indicates the risk of hypoxemia. The respiratory rate (RR) is a parameter for evaluating the degree of shortness of breath. The respiratory rate of a healthy adult is 12-20 times / minute (bpm). Above 15bpm may indicate dyspnea or compensatory breathing. The biofeedback mechanism sets SpO2 < 94% or RR > 15bpm as the judgment criteria for starting enhanced ventilation support. The control module performs logical judgment by comparing the real-time SpO2 and RR values ​​with the threshold value: when SpO2>94%, the system identifies the patient as being in a non-critical state (mode 1) and adopts the continuous positive airway pressure (CPAP) mode with I:E=1:2, that is, the ratio of inspiration and expiration is 1:2 to maintain a stable airway pressure; when SpO2≤94% and RR>15bpm, the system identifies the patient as being in a critical state (mode 2), continues to use I:E=1:2, reduces the total respiratory cycle time TCT by controlling the pulse width modulation signal frequency, that is, TCT is less than 4 seconds, and increases the respiratory rate to cope with the rapid breathing demand. The calculation formula of the total respiratory cycle time is TCT=T i +T e , where T i is the inspiratory time, T e is the exhalation time; when SpO2≤94% and RR≤15bpm, the system identifies the patient as being in a mildly critical condition (mode 3), switches to I:E=1:4, prolongs the exhalation time, and at the same time increases the total respiratory cycle time TCT by controlling the pulse width modulation signal frequency, that is, TCT is greater than or equal to 4 seconds, to reduce airway pressure and avoid complications caused by hyperventilation.

[0013] As a further solution of the present invention, the performance evaluation module constructs a simulation model through MATLAB and Simulink, and evaluates its accuracy by comparing with hardware test data, including the following specific contents: designing a simulation model based on MATLAB and Simulink, and generating simulated pulse oximeter and bioelectric signal acquisition unit output signals through MATLAB functions, which represent the patient's SpO2 and RR data respectively. Then, a control module is built in Simulink to implement feedback logic in the form of an algorithm block. The control module decides the ventilation mode and the inspiration and expiration time ratio (I:E ratio) based on the input SpO2 and RR values. The execution module adjusts the airflow output by simulating the switching behavior of the solenoid valve, which is specifically manifested as changes in the pressure signal to simulate the actual ventilation process. A multi-compartment lung equivalent circuit model is also introduced to simulate the dynamic changes in pressure, flow and volume of the lungs, so that the simulation environment is closer to actual physiological conditions, such as Figure 5As shown in the figure, when the inspiration-expiration ratio (I:E) is 1:2, the pressure signal waveform generated by the simulation model reflects the dynamic characteristics of the ventilator, in which the pressure rises rapidly to the peak value during the inspiration phase and gradually drops to the baseline level during the expiration phase.

[0014] The input parameters of the simulation experiment include SpO2 and RR, where the SpO2 range is set to 88% to 99% and the RR range is 12 to 30 times / minute, corresponding to three different conditions: non-critical state, critical state and mild critical state. The setting of the experimental scenario is based on the respiratory state that actual patients may encounter. For example, when SpO2>94%, it simulates mild respiratory discomfort and the system should select mode 1 (I:E=1:2); when SpO2≤94% and RR>15 times / minute, it simulates acute respiratory distress and the system should switch to mode 2 (I:E=1:2); and when SpO2≤94% and RR≤15 times / minute, mode 3 (I:E=1:4) is selected. In Simulink, the corresponding pressure signal is generated by the MATLAB function. For example, taking the RR of 20 times / minute as an example, 8 complete respiratory cycles are designed, including the inspiratory time (T i ) 1 second, exhalation time (T e ) waveform for 2 seconds, PEEP (positive end expiratory pressure) was set to 5 cmH2O, and peak inspiratory flow rate (PIFR) was set to 100 L / min to achieve a tidal volume of 500 mL. In the Mode 1 scenario (SpO2>94%), the simulation model shows that the system can stably output a breathing pattern of I:E=1:2, and the pressure signal presents a regular 1-second inhalation and 2-second exhalation cycle. The flow and volume waveforms are consistent with expectations, indicating that the system can maintain comfortable ventilation support in mild dyspnea. In the Mode 2 scenario (SpO2≤94%, RR>15 times / minute), the simulation results show that the system quickly switches to a high-frequency breathing mode, providing more breaths per minute (for example, 20 times / minute) to cope with acute hypoxia. The pressure waveform shows that the inspiratory peak reaches 25cmH2O, which meets the requirements of lung-protective ventilation. In the Mode 3 scenario (SpO2≤94%, RR≤15 times / minute), the system is adjusted to I:E=1:4, and the exhalation time is extended to 4 seconds. The simulation waveform shows a lower respiratory rate (12 times / minute), which is suitable for patients with more stable hypoxic conditions and avoids complications caused by hyperventilation.

[0015] To evaluate the accuracy of the simulation model, the model output was compared with data from a hardware prototype tested under identical conditions. The hardware was tested using real-world SpO2 and RR data collected from 12 healthy volunteers with an average age of 33.9 years and 32 subjects ranging in age from 19 to 92 years. The simulation model extracted RR from the PPG signals using a discrete wavelet transform (DWT) algorithm (using the sym5 wavelet with a 9-level decomposition). The results showed high consistency with the RR measured by the hardware using the biopotential signal acquisition unit.

[0016] Through the above simulation evaluation, the system can automatically select the appropriate ventilation mode and adjust the I:E ratio based on real-time changes in SpO2 and RR to meet the needs of different respiratory states. For example, in acute hypoxia (mode 2), the system increases the respiratory rate to improve oxygenation; in stable hypoxia (mode 3), it prolongs the exhalation time to reduce the risk of lung damage.

[0017] A performance evaluation method for a ventilator, comprising the following specific steps:

[0018] In step S1 , the patient's blood oxygen saturation (SpO2) is monitored by a pulse oximeter using photoplethysmography, and a bioelectric signal acquisition unit uses impedance spirometry to acquire a respiratory waveform and respiratory rate (RR).

[0019] Step S2: Filter the collected blood oxygen saturation and respiratory rate data to remove noise. Based on the biofeedback mechanism, preset thresholds (SpO2 < 94% or RR > 15 times / minute) are used to determine the patient's respiratory status, automatically select the ventilation mode, and dynamically adjust the inspiration to expiration time ratio (I:E).

[0020] In step S3, the air flow output is controlled by the solenoid valve, and the duty cycle is adjusted using a pulse width modulation (PWM) signal to achieve the adjustment of the I:E ratio.

[0021] In step S4, a simulation model is constructed using MATLAB and Simulink to simulate the performance of the ventilator under different respiratory states. The accuracy of the model is verified by comparing it with the hardware test data to ensure that the system can adjust the ventilation parameters according to the patient's real-time needs.

[0022] The technical effects and advantages of the performance evaluation system and method for a ventilator of the present invention are as follows: the present invention collects the patient's blood oxygen saturation (SpO2) and respiratory rate (RR) in real time, utilizes the biofeedback mechanism of the control module to intelligently select the ventilation mode and dynamically adjust the inspiration and expiration time ratio (I:E), and combines the solenoid valve of the execution module to accurately control the airflow output to ensure that the ventilation parameters are highly matched with the patient's needs. At the same time, a simulation model is constructed through MATLAB and Simulink to simulate the respiratory dynamics under real physiological conditions, and compared with the hardware test data for verification. The mode selection accuracy rate reaches 81.25%, and the RR extraction error is less than 10%, thereby improving the reliability and accuracy of the system. The present invention can not only adaptively adjust the ventilation mode according to the patient's status (such as non-critical, critical, and mildly critical), but also improve the patient's comfort through non-invasive monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The present invention is a structural schematic diagram of a performance evaluation system for a ventilator.

[0024] Figure 2 A technical roadmap for the use of existing MoCA scores in predicting weaning outcomes in ICU patients receiving mechanical ventilation.

[0025] Figure 3 This is a circuit diagram of a conventional turbine blower.

[0026] Figure 4 It is a structural schematic diagram of the ventilator of the present invention.

[0027] Figure 5 Schematic diagram of the configuration required for different ventilation modes of the present invention.

[0028] Figure 6 This is a schematic diagram of a simulated pressure signal when the breathing ratio is 1:2 according to the present invention.

[0029] Figure 7 Schematic diagram comparing the waveforms of the simulated respiratory signal, reference respiratory signal, and PPG signal of the present invention.

[0030] Figure 8 Schematic diagram of the results of the case screening test of the present invention.

[0031] In the figure: (a) represents mode 1, (b) represents mode 2, and (c) represents mode 3. PEEP represents positive end-expiratory pressure, that is, the positive pressure maintained in the airway at the end of exhalation. α represents the time required for the airway pressure to reach its peak during inspiration. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0033] Example 1

[0034] See Figure 1 As shown in the schematic diagram, the present invention provides a performance evaluation system for a ventilator, which includes a sensor module, a control module, an execution module and a performance evaluation module; the sensor module is used to collect the patient's blood oxygen saturation and respiratory rate; the control module uses a biofeedback mechanism to intelligently select the ventilation mode based on the blood oxygen saturation and respiratory rate data, and dynamically adjusts the ratio of inspiration and expiration time; the execution module adjusts the ratio of inspiration and expiration by controlling the opening and closing time of the airflow; the performance evaluation module constructs a simulation model through MATLAB and Simulink, and evaluates its accuracy by comparing it with hardware test data. Figure 4 Shown is a ventilator for use with the present invention.

[0035] Furthermore, the sensor module is used to collect the patient's blood oxygen saturation and respiratory rate, including: the sensor module includes a pulse oximeter and a bioelectric signal acquisition unit, the pulse oximeter is used to non-invasively monitor the patient's blood oxygen saturation (SpO2), and the bioelectric signal acquisition unit is used to collect respiratory waveform data to extract the respiratory rate (RR). The pulse oximeter is a high-precision device based on optical sensing technology. It calculates the SpO2 level through photoplethysmography (PPG). This non-invasive method does not require puncture or direct contact with blood, which improves the comfort of patients. The bioelectric signal acquisition unit uses impedance spirometry to record the breathing pattern by measuring the impedance changes of the chest caused by respiratory activity; the bioelectric signal acquisition unit uses an analog front-end (AFE) chip to capture the tiny electrical signals caused by changes in chest volume, converts them into digital signals, and extracts the respiratory rate (RR) through a signal processing algorithm.

[0036] Furthermore, the control module intelligently selects the ventilation mode based on the blood oxygen saturation and respiratory rate data and the biofeedback mechanism, and dynamically adjusts the ratio of inspiration and expiration time, including: the control module first filters the blood oxygen saturation (SpO2) and respiratory rate (RR) data collected by the sensor module to remove noise interference. The blood oxygen saturation is collected by the pulse oximeter, which may cause baseline drift or high-frequency noise due to ambient light, patient movement or poor skin contact. Therefore, the control module uses a moving average filter to smooth the photoplethysmography (PPG) signal with a sampling frequency of 125Hz; the respiratory rate data is collected by the bioelectric signal acquisition unit through impedance spirography, which records the impedance changes of the chest caused by breathing. Since the respiratory waveform may be interfered with by muscle activity or other electrical signals, the control module uses a low-pass filter (LPF) to remove high-frequency noise and combines it with a high-pass filter (HPF) to eliminate baseline drift, thereby extracting a clear respiratory signal.

[0037] During the data analysis stage, the control module uses the biofeedback mechanism to determine the patient's respiratory status and automatically adjust the ventilation mode. The blood oxygen saturation (SpO2) level is an indicator for evaluating the patient's oxygenation status. Its normal range is above 94%. Below 94% indicates the risk of hypoxemia. The respiratory rate (RR) is a parameter for evaluating the degree of shortness of breath. The respiratory rate of a healthy adult is 12-20 times / minute (bpm). Above 15bpm may indicate dyspnea or compensatory breathing. The biofeedback mechanism sets SpO2 < 94% or RR > 15bpm as the judgment criteria for starting enhanced ventilation support. Specifically, the control module performs logical judgment by comparing the real-time SpO2 and RR values ​​with the threshold: when SpO2>94%, the system recognizes that the patient is in a non-critical state (mode 1) and adopts continuous positive airway pressure (CPAP) mode, I:E=1:2, that is, the ratio of inspiration and expiration is 1:2 to maintain a stable airway pressure; when SpO2≤94% and RR>15bpm, the system recognizes that the patient is in a critical state (mode 2), continues to use I:E=1:2, and reduces the total respiratory cycle time TCT by controlling the pulse width modulation signal frequency, that is, TCT is less than 4 seconds, and increases the respiratory rate to cope with the rapid breathing demand. The calculation formula of the total respiratory cycle time is TCT=T i +T e , where T i is the inspiratory time, T eWhen SpO2 ≤ 94% and RR ≤ 15 bpm, the system identifies the patient as mildly critical (Mode 3) and switches to I:E = 1:4, extending the expiratory time. Simultaneously, by controlling the pulse-width modulation signal frequency, it increases the total respiratory cycle time (TCT) to ≥ 4 seconds, thereby reducing airway pressure and avoiding complications caused by hyperventilation. The following table shows ventilation mode selection based on blood oxygen saturation (SpO2) and respiratory rate (RR).

[0038] Table 1 Ventilation mode selection based on SpO2 and RR

[0039]

[0040]

[0041] In Table 1, blood oxygen saturation (SpO2) is used to assess the patient's oxygenation status, with a normal value above 94%; respiratory rate (RR) indicates the number of breaths per minute and is used to determine the degree of shortness of breath, with the unit being "breaths per minute" (bpm); total respiratory cycle time (TCT) indicates the time of a complete respiratory cycle, calculated as TCT = T i +T e The I:E ratio indicates the ratio of inspiratory time to expiratory time. Mode 1 (non-critical) is suitable for patients with blood oxygen saturation greater than 94% and provides an I:E ratio of 1:2. Mode 2 (critical) is suitable for patients with blood oxygen saturation ≤ 94% and respiratory rate greater than 15 breaths / minute, using an I:E ratio of 1:2. Mode 3 (mildly critical) is suitable for patients with blood oxygen saturation ≤ 94% and respiratory rate ≤ 15 breaths / minute, using an I:E ratio of 1:4. Figure 5 , shows the hardware configuration required for different ventilation modes according to the provisions of Table 1, (a) represents mode 1: SPO2>94%, so the I:E ratio is selected as 1:2; (b) represents mode 2: SPO2≤94% and RR>15bpm, so the I:E ratio is selected as 1:2; (c) represents mode 3: Spo2≤94% and RR≤15BPM, so the I:E ratio is selected as 1:4.

[0042] Furthermore, the execution module adjusts the ratio of inspiration and exhalation by controlling the opening and closing time of the airflow, including: the execution module has a built-in solenoid valve, the solenoid valve is a device that uses electromagnetic force to control the flow of fluid, and in the ventilator, it adjusts the delivery of oxygen or air by rapid switching, thereby controlling the duration of inspiration and exhalation. The control module generates a pulse width modulation (PWM) signal, and determines the time ratio of the solenoid valve opening and closing in each respiratory cycle by adjusting the duty cycle of the PWM signal. The duty cycle directly determines the time for the airflow to pass, thereby affecting the I:E ratio. Taking an I:E ratio of 1:2 as an example, assuming that the total respiratory cycle time (TCT) is 3 seconds, the inhalation time T i is 1 second, exhalation time T e For example, if the I:E ratio is 1:4, the total cycle time is 5 seconds, with an inhale time of 1 second and an exhale time of 4 seconds. The duty cycle is adjusted to 20%, and the solenoid valve opens for 1 second and closes for 4 seconds every 5 seconds.

[0043] Furthermore, the performance evaluation module constructs a simulation model through MATLAB and Simulink, and evaluates its accuracy by comparing with hardware test data, including: designing a simulation model based on MATLAB and Simulink, and generating simulated pulse oximeter and bioelectric signal acquisition unit output signals through MATLAB functions, which represent the patient's SpO2 and RR data respectively. Then, a control module was built in Simulink to implement feedback logic in the form of an algorithm block. The control module decides the ventilation mode and the inspiration and expiration time ratio (I:E ratio) based on the input SpO2 and RR values. The execution module adjusts the airflow output by simulating the switching behavior of the solenoid valve, which is specifically manifested as changes in the pressure signal to simulate the actual ventilation process. A multi-compartment lung equivalent circuit model is also introduced to simulate the dynamic changes in pressure, flow and volume of the lungs, so that the simulation environment is closer to actual physiological conditions, such as Figure 6 As shown in the figure, when the inspiration-expiration ratio (I:E) is 1:2, the pressure signal waveform generated by the simulation model reflects the dynamic characteristics of the ventilator, in which the pressure rises rapidly to the peak value during the inspiration phase and gradually drops to the baseline level during the expiration phase.

[0044] The input parameters of the simulation experiment include SpO2 and RR, where the SpO2 range is set to 88% to 99% and the RR range is 12 to 30 times / minute, corresponding to three different conditions: non-critical state, critical state and mild critical state. The setting of the experimental scenario is based on the respiratory state that actual patients may encounter. For example, when SpO2>94%, it simulates mild respiratory discomfort and the system should select mode 1 (I:E=1:2); when SpO2≤94% and RR>15 times / minute, it simulates acute respiratory distress and the system should switch to mode 2 (I:E=1:2); and when SpO2≤94% and RR≤15 times / minute, mode 3 (I:E=1:4) is selected. In Simulink, the corresponding pressure signal is generated by the MATLAB function. For example, taking the RR of 20 times / minute as an example, 8 complete respiratory cycles are designed, including the inspiratory time (T i ) 1 second, exhalation time (T e ) waveform for 2 seconds, PEEP (positive end expiratory pressure) was set to 5 cmH2O, and peak inspiratory flow rate (PIFR) was set to 100 L / min to achieve a tidal volume of 500 mL. In the Mode 1 scenario (SpO2>94%), the simulation model shows that the system can stably output a breathing pattern of I:E=1:2, and the pressure signal presents a regular 1-second inhalation and 2-second exhalation cycle. The flow and volume waveforms are consistent with expectations, indicating that the system can maintain comfortable ventilation support in mild dyspnea. In the Mode 2 scenario (SpO2≤94%, RR>15 times / minute), the simulation results show that the system quickly switches to a high-frequency breathing mode, providing more breaths per minute (for example, 20 times / minute) to cope with acute hypoxia. The pressure waveform shows that the inspiratory peak reaches 25cmH2O, which meets the requirements of lung-protective ventilation. In the Mode 3 scenario (SpO2≤94%, RR≤15 times / minute), the system is adjusted to I:E=1:4, and the exhalation time is extended to 4 seconds. The simulation waveform shows a lower respiratory rate (12 times / minute), which is suitable for patients with more stable hypoxic conditions and avoids complications caused by hyperventilation.

[0045] In order to evaluate the accuracy of the simulation model, the output results of the simulation model were compared with the test data of the hardware prototype under the same conditions. The hardware test was based on 12 healthy volunteers with an average age of 33.9 years and 32 subjects with an age range of 19 to 92 years, and real SpO2 and RR data were collected. In the simulation model, RR was extracted from the PPG signal using the discrete wavelet transform (DWT) algorithm (using sym5 wavelet, 9-level decomposition). The results showed that it was highly consistent with the RR measured by the hardware through the bioelectric signal acquisition unit. Figure 7As shown in Figure 2, under the RR condition of 20 times / min, the number of peaks of the simulation signal and the hardware signal is the same, and the mean absolute percentage error (MAPE) of the two is less than 10%. The comparison of ventilation mode selection further shows that, as Figure 8 As shown in the figure, among the 32 subjects with SpO2 < 94%, the hardware and simulation models had consistent selections in 26 cases, while there were only 6 cases with disagreements, and the mode selection accuracy reached 81.25%.

[0046] Through the above simulation evaluation, the system can automatically select the appropriate ventilation mode and adjust the I:E ratio based on real-time changes in SpO2 and RR to meet the needs of different respiratory states. For example, in acute hypoxia (mode 2), the system increases the respiratory rate to improve oxygenation; in stable hypoxia (mode 3), it prolongs the exhalation time to reduce the risk of lung damage.

[0047] A performance evaluation method for a ventilator, comprising the following specific steps:

[0048] In step S1 , the patient's blood oxygen saturation (SpO2) is monitored by a pulse oximeter using photoplethysmography, and a bioelectric signal acquisition unit uses impedance spirometry to acquire a respiratory waveform and respiratory rate (RR).

[0049] Step S2: Filter the collected blood oxygen saturation and respiratory rate data to remove noise. Based on the biofeedback mechanism, preset thresholds (SpO2 < 94% or RR > 15 times / minute) are used to determine the patient's respiratory status, automatically select the ventilation mode, and dynamically adjust the inspiration to expiration time ratio (I:E).

[0050] In step S3, the air flow output is controlled by the solenoid valve, and the duty cycle is adjusted using a pulse width modulation (PWM) signal to achieve the adjustment of the I:E ratio.

[0051] In step S4, a simulation model is constructed using MATLAB and Simulink to simulate the performance of the ventilator under different respiratory states. The accuracy of the model is verified by comparing it with the hardware test data to ensure that the system can adjust the ventilation parameters according to the patient's real-time needs.

[0052] The present invention collects the patient's blood oxygen saturation (SpO2) and respiratory rate (RR) in real time, uses the biofeedback mechanism of the control module to intelligently select the ventilation mode and dynamically adjust the inspiration and expiration time ratio (I:E), and combines the solenoid valve of the execution module to accurately control the airflow output to ensure that the ventilation parameters are highly matched with the patient's needs. At the same time, a simulation model is constructed through MATLAB and Simulink to simulate the respiratory dynamics under real physiological conditions, and compared with the hardware test data for verification. The mode selection accuracy rate reaches 81.25%, and the RR extraction error is less than 10%, which improves the reliability and accuracy of the system. The present invention can not only adaptively adjust the ventilation mode according to the patient's status (such as non-critical, critical, and mildly critical), but also improve the patient's comfort through non-invasive monitoring.

[0053] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0054] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A performance evaluation system for a ventilator, comprising a sensor module for collecting blood oxygen saturation and respiratory rate of a patient, characterized in that: It also includes a control module, an execution module and a performance evaluation module; the control module intelligently selects the ventilation mode based on blood oxygen saturation and respiratory rate data using a biofeedback mechanism and dynamically adjusts the ratio of inhalation and exhalation time; the execution module adjusts the ratio of inhalation and exhalation by controlling the opening and closing time of the airflow; the performance evaluation module constructs a simulation model using MATLAB and Simulink, and evaluates its accuracy by comparing it with hardware test data The blood oxygen saturation level is an indicator for evaluating the patient's oxygenation status. Its normal range is above 94%. A level below 94% indicates the risk of hypoxemia. The respiratory rate is a parameter for evaluating the degree of shortness of breath. The respiratory rate of a healthy adult is 12-20 times / minute. A level above 15 times / minute indicates the risk of compensatory breathing. The biofeedback mechanism sets blood oxygen saturation <94% or respiratory rate >15bpm as the judgment criteria for initiating enhanced ventilation support: when blood oxygen saturation >94% The system recognizes that the patient is in a non-critical condition and adopts continuous positive pressure ventilation mode, that is, the ratio of inspiration and expiration is 1:2; when the blood oxygen saturation is ≤94% and the respiratory rate is >15 times / minute, the system recognizes that the patient is in a critical condition and continues to use I:E=1:2, reducing the total respiratory cycle time TCT by controlling the pulse width modulation signal frequency, that is, TCT is less than 4 seconds; when the blood oxygen saturation is ≤94% and the respiratory rate is 15 times / minute, the system recognizes that the patient is in a mild critical condition, and the ratio of inspiration and expiration is 1:

4.

2. A performance evaluation system for a ventilator according to claim 1, characterized in that The execution module has a built-in solenoid valve, which generates a pulse width modulation signal and determines the time ratio of the solenoid valve opening and closing in each breathing cycle by adjusting the duty cycle of the pulse width modulation signal. The duty cycle determines the time of airflow passing through and affects the ratio of inspiration and exhalation.

3. A performance evaluation system for a ventilator according to claim 1, characterized in that The sensor module includes a pulse oximeter and a bioelectric signal acquisition unit. The pulse oximeter non-invasively monitors blood oxygen saturation through photoplethysmography, and the bioelectric signal acquisition unit uses impedance spirometry to collect respiratory waveforms to extract respiratory rate.

4. A performance evaluation system for a ventilator according to claim 1, characterized in that: The control module filters the collected blood oxygen saturation and respiratory rate data, uses a moving average filter to process the photoplethysmography signal with a sampling frequency of 125 Hz, and uses a low-pass filter and a high-pass filter to process the respiratory waveform signal to remove noise and baseline drift.

5. A ventilator performance evaluation system according to claim 1, wherein the performance evaluation module constructs a simulation model based on MATLAB and Simulink to generate simulated blood oxygen saturation and respiratory rate signals, builds a control module algorithm block to implement feedback logic, introduces a multi-compartment lung equivalent circuit model to simulate lung dynamics, and outputs pressure, flow, and volume waveforms.

6. A performance evaluation system for a ventilator according to claim 5, wherein the simulation experiment input parameters of the performance evaluation module are blood oxygen saturation and respiratory rate, wherein the blood oxygen saturation range is set to 88% to 99%, and the respiratory rate range is set to 12 to 30 times / minute.

7. A performance evaluation system for a ventilator according to claim 5, wherein the performance evaluation module sets three scenarios: mode 1, i.e., blood oxygen saturation > 94%; mode 2, i.e., blood oxygen saturation ≤ 94% and respiratory rate > 15 times / minute; mode 3, i.e., blood oxygen saturation ≤ 94% and respiratory rate ≤ 15 times / minute.

8. A ventilator performance evaluation system according to claim 5, wherein the simulation model is compared with hardware test data for verification. The hardware test is based on 12 healthy volunteers and 32 subjects. The respiratory rate is extracted using discrete wavelet transform. The mean absolute percentage error is less than 10%, and the mode selection accuracy reaches 81.25%.

9. A performance evaluation method for a ventilator, characterized in that: A performance evaluation system for a ventilator as described in any one of claims 1 to 8 comprises the following specific steps: Step S1, monitoring the patient's blood oxygen saturation using photoplethysmography with a pulse oximeter, and collecting respiratory waveforms and respiratory rate using impedance spirometry with a bioelectrical signal acquisition unit; Step S2: Filter the collected blood oxygen saturation and respiratory rate data to remove noise. Based on the biofeedback mechanism, a preset threshold is set: blood oxygen saturation <94% or respiratory rate >15 breaths / minute. The patient's respiratory status is determined, the ventilation mode is automatically selected, and the ratio of inspiration to expiration time is dynamically adjusted. Step S3, controlling the airflow output through the solenoid valve and adjusting the duty cycle using the pulse width modulation signal to achieve the adjustment of the inhalation and exhalation ratio; In step S4, a simulation model is constructed using MATLAB and Simulink to simulate the performance of the ventilator under different respiratory states. The accuracy of the model is verified by comparing it with the hardware test data to ensure that the system can adjust the ventilation parameters according to the patient's real-time needs.

Images