Oxygen generation system capable of automatically adjusting oxyhemoglobin saturation and inhalation volume of oxygen inhaler

Abstract

The oxygen generation system comprises an oxygen generator, a nasal oxygen tube and an oxygen saturation tester, key parameters are collected in real time through the oxygen generator, an ROX index is calculated, the change trend of the ROX index is predicted, an intelligent controller can actively adjust the oxygen flow or give out an early warning, and the oxygen saturation and the inhalation volume of an oxygen uptake person can be automatically adjusted through the oxygen generator, the nasal oxygen tube and the oxygen saturation tester. The dynamic management of the oxygen uptake state is realized; a GRU neural network model is embedded in the intelligent control processor, high-frequency prediction is carried out on the ROX trend, timely intervention can be carried out before deterioration, the slow reaction risk is remarkably reduced, and the treatment safety of high-risk users is improved; a lung function individualized correction mechanism is introduced, and the calculation result of the ROX index is dynamically adjusted by combining the FVC and FEV1 indexes of the user, so that the adaptability of the system in the crowd with limited lung functions is improved, and the real individualized oxygen therapy is realized; the pressure sensor is used for collecting breathing waveforms, filtering and recognition algorithms are combined, abnormal conditions such as apnea and tidal breathing are quickly recognized, and oxygen therapy parameter adjustment is automatically linked.

Description

Technical Field

[0001] The present invention relates to the technical field of ventilators, and more specifically, to an oxygen production system capable of automatically adjusting blood oxygen saturation and the inhalation volume of an oxygen inhaler. Background Art

[0002] An oxygen concentrator, also known as an oxygen generator or oxygen machine, is a medical device used to extract oxygen from the surrounding air, concentrate it, and deliver it to patients who need additional oxygen. Such devices are commonly used in hospitals, clinics, home healthcare, and nursing care settings to help those with respiratory problems or other health issues maintain an adequate oxygen supply.

[0003] However, the existing oxygen concentrator system only controls the intake and output of oxygen, regardless of whether the user needs it or whether it is used normally.

[0004] In the existing technology, ultrasonic gas sensors for oxygen concentrators on the market generally only realize real-time status monitoring of oxygen and concentration in the oxygen concentrator. In the existing technology, ultrasonic gas sensors only collect machine data of the oxygen concentrator. Such sensors cannot collect user data and do not have comprehensive data analysis capabilities. They cannot combine the oxygen concentrator, user usage and user absorption to obtain accurate oxygen production of the oxygen concentrator. When the user inhales oxygen, he or she does not know the relationship between his or her current blood oxygen status and the amount of oxygen he or she needs to use and the status of the oxygen concentrator. Abnormal blood oxygen status of the user may occur due to insufficient air flow from the oxygen concentrator.

[0005] Conventional oxygen concentrators are equipped with a flow adjustment knob for users to independently adjust the flow rate of the oxygen concentrator. Some users are equipped with a blood oxygen probe to monitor their blood oxygen and other physiological information. For many users, this method that requires human intervention is actually very inconvenient. For example, when the user is sleeping and breathing oxygen, he cannot participate in the working status of the oxygen concentrator and the acquisition of his own physiological signals, as well as abnormal analysis, which poses a great safety hazard.

[0006] Therefore, it is necessary to provide an oxygen production system that automatically adjusts the blood oxygen saturation and the inhalation volume of the oxygen inhaler to solve the shortcomings of the existing technology. Summary of the Invention

[0007] In view of this, and in order to solve the above problems, the present invention provides an oxygen production system with automatic adjustment of blood oxygen saturation and the inhalation volume of the oxygen inhaler, including an oxygen concentrator, a nasal oxygen cannula, and a blood oxygen saturation tester. The above equipment collects key parameters such as blood oxygen saturation, heart rate, respiratory rate and inhaled oxygen concentration in real time, calculates the ROX index and predicts its changing trend. The intelligent controller can actively adjust the oxygen flow or issue an early warning to achieve dynamic management of the oxygen inhalation status; the intelligent control processor is embedded with a GRU neural network model to perform high-frequency prediction of the ROX trend, which can intervene in time before deterioration, significantly reduce the risk of delayed reaction, and improve the treatment safety of high-risk users; an individualized lung function correction mechanism is introduced, combined with the user's FVC and FEV1 indicators, to dynamically adjust the calculation result of the ROX index, improve the adaptability of the system in people with limited lung function, and realize truly personalized oxygen therapy; the pressure sensor collects respiratory waveforms and combines filtering and recognition algorithms to quickly identify abnormal conditions such as apnea and tidal respiration, and automatically link oxygen therapy parameter adjustment.

[0008] An oxygen production system with automatic adjustment of blood oxygen saturation and inhaled amount of an oxygen inhaler comprises an oxygen concentrator, a nasal oxygen cannula, and a blood oxygen saturation tester. The oxygen concentrator is connected to the oxygen inhaler via the nasal oxygen cannula, the nasal oxygen cannula being used to directly connect to the oxygen inhaler to supply oxygen to the oxygen inhaler and to collect the respiratory rate and inhaled oxygen concentration of the oxygen inhaler in real time. The blood oxygen saturation tester is connected to the oxygen inhaler and the oxygen concentrator respectively. The blood oxygen saturation tester monitors the blood oxygen saturation and heart rate of the oxygen inhaler. The oxygen concentrator determines whether the current oxygen inhalation amount of the oxygen concentrator needs to be adjusted based on the respiratory rate, inhaled oxygen concentration and blood oxygen saturation of the oxygen inhaler, and can adjust the oxygen output flow rate of the oxygen concentrator in real time through an electronic flowmeter control module. At the same time, the oxygen output concentration, oxygen output flow rate equipment data and the user's blood oxygen saturation of the oxygen concentrator can be exchanged with an external terminal through a wireless communication module. The system also includes an intelligent control processor, which has an embedded lightweight GRU recurrent neural network model. The system also includes an intelligent control processor, which outputs data through an oxygen concentrator, a nasal oxygen cannula, and a blood oxygen saturation tester to predict the ROX trend in the next 3 minutes. If the predicted ROX value is <3.85 or the decline rate is >0.2 / min, the oxygen flow rate is automatically increased or an alarm is triggered.

[0009] Among them, the intelligent control processor is embedded with a lightweight GRU recurrent neural network model. The output data of the oxygen concentrator, nasal oxygen cannula, and blood oxygen saturation tester include blood oxygen saturation, heart rate, respiratory rate, and inhaled oxygen concentration. Specifically, blood oxygen saturation and heart rate are collected in real time at a 15-second cycle by the blood oxygen saturation tester in contact with the user's fingertips; the respiratory waveform is collected in real time by the first pressure sensor installed at the air outlet of the nasal oxygen cannula, and then calculated in real time by the FFT fast Fourier transform method; the inhaled oxygen concentration is measured in real time by the oxygen concentration sensor between the air outlet of the oxygen concentrator and the nasal oxygen cannula, and is updated every 15 seconds.

[0010] Furthermore, a ROX index is calculated based on the blood oxygen saturation and the respiratory rate, and an oxygenation index is calculated based on the blood oxygen saturation, so as to adjust the oxygen flow of the current oxygen concentrator. The calculated value of the ROX index = (blood oxygen saturation * 100) / (inspired oxygen concentration fraction * respiratory rate * heart rate value). The ROX index is used to evaluate the ventilation treatment effect within the corresponding ventilation treatment time. The higher the ROX index value, the better the patient's lung oxygenation condition.

[0011] Furthermore, the ROX index is compared with the ROX index threshold to obtain the ROX index comparison result, and the blood oxygen saturation, inspired oxygen concentration fraction, respiratory rate, and heart rate values ​​are compared with the set threshold to obtain the comparison result. According to the comparison result, corresponding ventilation treatment suggestions can be provided to the user, thereby helping medical staff to judge or predict the ventilation effect and prompting medical staff to switch the ventilation treatment mode in time, thereby shortening the patient's machine time and improving the prognosis.

[0012] Furthermore, the normal dead space volume of the oropharynx is 50 ml, the tidal volume is 500 ml, the respiratory rate is 20 breaths / min, and the amount of inhaled pure oxygen is: 0.5*V (the amount of oxygen in the dead space 0.5 seconds before inspiration) + 1*V (the amount of pure oxygen inhaled during inspiration) + (500-0.5*V-1*V)*20%=1.2*V+100. Since the inspired oxygen fraction varies at different oxygen flow rates, when the oxygen flow rate is between 1-6 L / min, the inspired oxygen fraction increases by 0.04 for every 1 L / min increase in oxygen flow rate. Therefore, ROX=blood oxygen saturation / (21+4*V)%, where V is the inspired oxygen flow rate (L / min) and 21 represents the oxygen fraction in sea level air. The ROX index is adjusted by adjusting the inspired oxygen flow rate to meet the required ROX index value.

[0013] Furthermore, the wireless communication module is one or more of a WIFI module, a Bluetooth module, a NB-IOT module, a LORA module, and an infrared module.

[0014] Furthermore, when the blood oxygen saturation tester detects that the user's blood oxygen saturation is abnormal, and the user's blood oxygen saturation is still abnormal after the oxygen concentrator adjusts the oxygen production, the blood oxygen saturation tester uploads the information to the external terminal (i.e., the cloud server) through the wireless communication module. At this time, the user or doctor can monitor the user's oxygen inhalation status in real time from the cloud server and remind the user of abnormal content. Among them, the cloud server receives and stores the user's real-time ROX trend curve, blood oxygen saturation, respiratory rate, heart rate data, and oxygen flow setting records. After the server AI model identifies the abnormal data features, it automatically pushes specific ventilation mode switching suggestions to the user terminal and the bound medical terminal, including but not limited to increasing the oxygen flow, changing to a high-flow nasal cannula oxygen inhalation mode, or preparing for non-invasive ventilation; the ventilation mode suggestions are pushed to the user and medical staff terminals in the form of text and voice messages at the same time.

[0015] Furthermore, when the oximeter detects an abnormal blood oxygen saturation in the user, and if the abnormal blood oxygen saturation remains after the oxygen concentrator adjusts the oxygen production, the oximeter triggers an alarm module to initiate an alarm process. The alarm module includes a local buzzer alarm device, an LED flashing prompt device, and a remote push alarm system. The remote push alarm system communicates with the user's mobile terminal and a preset medical staff terminal via the wireless communication module. When the ROX value for two consecutive prediction cycles falls below the set risk threshold, the alarm module issues an alarm via voice broadcast, SMS, or APP notification, alerting medical staff to promptly initiate emergency medical intervention.

[0016] Furthermore, the oxygen concentrator also includes a humidifier. When the oxygen at the nasal oxygen cannula is detected to be relatively dry, the humidifier is turned on to provide humidified oxygen to the oxygen user. The humidifier specifically includes a humidity sensor, an atomizer generator, and a PID controller. The humidity sensor is located at the outlet of the nasal oxygen cannula and measures the oxygen humidity every 10 seconds and feeds real-time humidity data back to the intelligent control processor. The control processor has a built-in PID algorithm that compares the humidity value with a preset target humidity range (40%-60%) in real time, outputting a precise PWM signal to drive the atomizer generator to stabilize the oxygen humidity within the user's comfortable range.

[0017] Furthermore, a first pressure sensor is provided at the air outlet of the nasal oxygen cannula, and is used to collect the respiratory rate of the oxygen user. Specifically, the first pressure sensor is a MEMS micro-pressure sensor, which collects the oxygen user's respiratory waveform data every 10 seconds and transmits it to the intelligent control processor. The control processor has a built-in third-order filter and abnormal waveform recognition algorithm, which automatically identifies abnormal waveform characteristics such as Cheyne-Stokes respiration, apnea, and forced breathing. When an abnormal respiratory waveform is detected and the synchronously collected blood oxygen saturation value drops significantly, the intelligent control processor is triggered to perform high oxygen concentration adjustment or ventilation mode change in advance, and the abnormal event and waveform characteristic data are automatically uploaded to the cloud server for storage and filing for subsequent medical analysis.

[0018] Furthermore, the system also includes a lung function individualization correction module for performing individualized correction of the ROX index based on the lung function data input by the user; the lung function individualization correction module includes:

[0019] The data access unit is configured to receive lung function indicators of the user, including vital capacity (FVC) and forced expiratory volume (FEV1), from which the sources include:

[0020] Users manually input the FVC and FEV1 values ​​from the hospital test report through the accompanying mobile terminal application;

[0021] Or connect to an external spirometer and import data via USB, Bluetooth or WIFI;

[0022] The feature analysis unit is used to combine the FVC and FEV1 values ​​input by the user and their corresponding physiological parameters such as age, gender, and height, call the preset prediction model to calculate the standard FVC and FEV1, and calculate the correction factor μ using the following formula: Where k is the empirical weighting coefficient, ranging from 0.6 to 1.2, and the default value is 1.0;

[0023] The dynamic adjustment unit receives the correction factor μ and combines it with the real-time acquired blood oxygen saturation, respiratory rate, heart rate, and inspired oxygen concentration data to output the corrected individualized ROX index:

[0024] The ROX' index is used as one of the input parameters for subsequent oxygen flow adjustment judgment, ventilation mode suggestion and alarm logic.

[0025] Beneficial effects of the present invention: The present invention provides an oxygen production system that automatically adjusts blood oxygen saturation and the inhalation volume of oxygen inhalers, including an oxygen concentrator, a nasal oxygen cannula, and a blood oxygen saturation tester. The above-mentioned equipment collects key parameters such as blood oxygen saturation, heart rate, respiratory rate and inhaled oxygen concentration in real time, calculates the ROX index and predicts its changing trend. The intelligent controller can actively adjust the oxygen flow or issue an early warning to achieve dynamic management of the oxygen inhalation status; the intelligent control processor is embedded with a GRU neural network model to perform high-frequency prediction of the ROX trend, which can intervene in time before deterioration, significantly reduce the risk of delayed reaction, and improve the treatment safety of high-risk users; introduces an individualized lung function correction mechanism, combines the user's FVC and FEV1 indicators, and dynamically adjusts the calculation result of the ROX index, thereby improving the adaptability of the system in people with limited lung function and realizing truly personalized oxygen therapy; uses a pressure sensor to collect respiratory waveforms and combines filtering and recognition algorithms to quickly identify abnormal conditions such as apnea and tidal breathing, and automatically links oxygen therapy parameter adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is the overall flow chart of the present invention.

[0027] Figure 2 It is the overall flow chart of the present invention.

[0028] Figure 3 It is a partially enlarged schematic diagram of the process of the present invention.

[0029] Figure 4 It is a partially enlarged schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0030] Example:

[0031] like Figure 1-2As shown, an oxygen production system with automatic adjustment of blood oxygen saturation and inhalation volume of an oxygen inhaler includes an oxygen concentrator, a nasal oxygen cannula, and a blood oxygen saturation tester. The oxygen concentrator is connected to the oxygen inhaler through the nasal oxygen cannula. The nasal oxygen cannula is used to directly connect to the oxygen inhaler to supply oxygen to the oxygen inhaler and collect the respiratory rate and inhaled oxygen concentration of the oxygen inhaler in real time. The blood oxygen saturation tester is connected to the oxygen inhaler and the oxygen concentrator respectively. The blood oxygen saturation tester monitors the blood oxygen saturation and heart rate of the oxygen inhaler. The oxygen concentrator determines whether the current oxygen inhalation volume of the oxygen concentrator needs to be adjusted based on the respiratory rate, inhaled oxygen concentration and blood oxygen saturation of the oxygen inhaler, and can adjust the oxygen output flow rate of the oxygen concentrator in real time through the electronic flowmeter control module. At the same time, the oxygen output concentration, oxygen output flow device data and the user's blood oxygen saturation of the oxygen concentrator can be exchanged with an external terminal through a wireless communication module. The system also includes an intelligent control processor, which has a built-in lightweight GRU recurrent neural network model. The system also includes an intelligent control processor, which outputs data through an oxygen concentrator, a nasal oxygen tube, and a blood oxygen saturation tester to predict the ROX trend in the next 3 minutes. If the predicted ROX value is <3.85 or the rate of decline is

[0032] If the rate is >0.2 / min, the oxygen flow rate will be automatically increased or an alarm will be triggered. The intelligent control processor embeds a lightweight GRU recurrent neural network model. The oxygen concentrator, nasal oxygen cannula, and blood oxygen saturation meter output data including blood oxygen saturation, heart rate, respiratory rate, and inspired oxygen concentration. Specifically, blood oxygen saturation and heart rate are collected in real time at a 15-second cycle by the blood oxygen saturation meter in contact with the user's fingertips. The respiratory rate is collected in real time by the first pressure sensor installed at the nasal oxygen cannula outlet, and the respiratory waveform is calculated in real time using the FFT fast Fourier transform method. The inspired oxygen concentration is measured in real time by the oxygen concentration sensor between the oxygen concentrator outlet and the nasal oxygen cannula, and is updated every 15 seconds. The ROX index is calculated based on the blood oxygen saturation and the respiratory rate, and the oxygenation index is calculated based on the blood oxygen saturation to adjust the oxygen flow of the current oxygen concentrator. The calculated ROX index value = (blood oxygen saturation * 100) / (inspired oxygen concentration fraction * respiratory rate * heart rate). The ROX index is used to evaluate the ventilation treatment effect within the corresponding ventilation treatment time. The higher the ROX index value, the better the patient's lung oxygenation status. The ROX index is compared with the ROX index threshold to obtain a ROX index comparison result. The blood oxygen saturation, inspired oxygen concentration fraction, respiratory rate, and heart rate values ​​are all compared with the set threshold to obtain a comparison result. Based on the comparison result, corresponding ventilation treatment recommendations can be provided to the user, thereby helping medical staff to judge or predict ventilation effects and prompting medical staff to switch ventilation treatment modes in a timely manner, thereby shortening the patient's ventilation time and improving prognosis. The normal dead space volume of the oropharynx is 50 ml, the tidal volume is 500 ml, the respiratory rate is 20 breaths / min, and the amount of inhaled pure oxygen is: 0.5*V (the amount of oxygen in the dead space 0.5 seconds before inspiration) + 1*V (the amount of pure oxygen inhaled during inspiration) + (500-0.5*V-1*V)*20% = 1.2*V+100. Since the inspired oxygen fraction varies at different oxygen flow rates, when the oxygen flow rate is between 1-6 L / min, the inspired oxygen fraction increases by 0.04 for every 1 L / min increase in oxygen flow rate. Therefore, ROX = blood oxygen saturation / (21+4*V)%, where V is the inspired oxygen flow rate (L / min) and 21 represents the oxygen fraction in sea level air. The ROX index is adjusted by adjusting the inspired oxygen flow rate to meet the required ROX index value.

[0033] like Figure 4 As shown, the wireless communication module is one or more of a WIFI module, a Bluetooth module, a NB-IOT module, a LORA module, and an infrared module.

[0034] like Figure 4As shown, when the blood oxygen saturation tester detects that the user's blood oxygen saturation is abnormal, and the user's blood oxygen saturation is still abnormal after the oxygen concentrator adjusts the oxygen production, the blood oxygen saturation tester uploads the information to the external terminal (i.e., the cloud server) through the wireless communication module. At this time, the user or doctor can monitor the user's oxygen inhalation status in real time from the cloud server and remind the user of abnormal content. Among them, the cloud server receives and stores the user's real-time ROX trend curve, blood oxygen saturation, respiratory rate, heart rate data, and oxygen flow setting records. After the server AI model identifies the abnormal data features, it automatically pushes specific ventilation mode switching suggestions to the user terminal and the bound medical terminal, including but not limited to increasing the oxygen flow, changing to a high-flow nasal cannula oxygen inhalation mode, or preparing for non-invasive ventilation; the ventilation mode suggestions are pushed to the user and medical staff terminals in the form of text and voice messages at the same time.

[0035] like Figure 3 As shown, when the blood oxygen saturation tester detects that the user's blood oxygen saturation is abnormal, and the user's blood oxygen saturation is still abnormal after the oxygen concentrator adjusts the oxygen production, the blood oxygen saturation tester triggers the alarm module to perform alarm processing. Among them, the alarm module includes a local buzzer alarm device, an LED flashing prompt device and a remote push alarm system; the remote push alarm system is connected to the user's mobile terminal and the preset medical staff terminal through the wireless communication module. When the ROX value is lower than the set risk threshold for two consecutive prediction cycles, the alarm module issues an alarm in the form of voice broadcast, SMS or APP notification to remind medical staff to perform emergency medical intervention in time. The oxygen concentrator also includes a humidifier. When it is detected that the oxygen at the nasal oxygen tube is relatively dry, the humidifier is turned on to provide humidified oxygen to the oxygen inhaler. The humidifier specifically includes a humidity sensor, an atomizer, and a PID controller. The humidity sensor, located at the outlet of the nasal oxygen cannula, measures the oxygen humidity every 10 seconds and feeds real-time humidity data to an intelligent control processor. The control processor incorporates a built-in PID algorithm that compares the humidity value with a preset target humidity range (40%-60%) in real time, outputting a precise PWM signal to drive the atomizer to stabilize the oxygen humidity within the user's comfortable range. A first pressure sensor is located at the outlet of the nasal oxygen cannula, which is used to collect the oxygen user's respiratory rate. Specifically, the first pressure sensor, a MEMS micro-pressure sensor, collects respiratory waveform data every 10 seconds and transmits it to the intelligent control processor. The control processor incorporates a built-in third-order filter and an abnormal waveform recognition algorithm to automatically identify abnormal waveform features such as Cheyne-Stokes respiration, apnea, and forced breathing. When an abnormal respiratory waveform is detected and the simultaneously collected blood oxygen saturation value drops significantly, the intelligent control processor is triggered to preemptively adjust the oxygen concentration or switch ventilation modes. The abnormal event and waveform feature data are automatically uploaded to a cloud server for storage and subsequent medical analysis.

[0036] like Figure 3 As shown, the system also includes a lung function individualization correction module for performing individualized correction of the ROX index based on the lung function data input by the user; the lung function individualization correction module includes:

[0037] The data access unit is configured to receive lung function indicators of the user, including vital capacity (FVC) and forced expiratory volume (FEV1), from which the sources include:

[0038] Users manually input the FVC and FEV1 values ​​from the hospital test report through the accompanying mobile terminal application;

[0039] Or connect to an external spirometer and import data via USB, Bluetooth or WIFI;

[0040] The feature analysis unit is used to combine the FVC and FEV1 values ​​input by the user and their corresponding physiological parameters such as age, gender, and height, call the preset prediction model to calculate the standard FVC and FEV1, and calculate the correction factor μ using the following formula: Where k is the empirical weighting coefficient, ranging from 0.6 to 1.2, and the default value is 1.0;

[0041] The dynamic adjustment unit receives the correction factor μ and combines it with the real-time acquired blood oxygen saturation, respiratory rate, heart rate, and inspired oxygen concentration data to output the corrected individualized ROX index:

[0042] The ROX' index is used as one of the input parameters for subsequent oxygen flow adjustment judgment, ventilation mode suggestion and alarm logic.

[0043] Beneficial effects of the present invention: The present invention provides an oxygen production system that automatically adjusts blood oxygen saturation and the inhalation volume of oxygen inhalers, including an oxygen concentrator, a nasal oxygen cannula, and a blood oxygen saturation tester. The above-mentioned equipment collects key parameters such as blood oxygen saturation, heart rate, respiratory rate and inhaled oxygen concentration in real time, calculates the ROX index and predicts its changing trend. The intelligent controller can actively adjust the oxygen flow or issue an early warning to achieve dynamic management of the oxygen inhalation status; the intelligent control processor is embedded with a GRU neural network model to perform high-frequency prediction of the ROX trend, which can intervene in time before deterioration, significantly reduce the risk of delayed reaction, and improve the treatment safety of high-risk users; introduces an individualized lung function correction mechanism, combines the user's FVC and FEV1 indicators, and dynamically adjusts the calculation result of the ROX index, thereby improving the adaptability of the system in people with limited lung function and realizing truly personalized oxygen therapy; uses a pressure sensor to collect respiratory waveforms and combines filtering and recognition algorithms to quickly identify abnormal conditions such as apnea and tidal breathing, and automatically links oxygen therapy parameter adjustment.

[0044] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. An oxygen production system that automatically adjusts blood oxygen saturation and the amount of oxygen inhaled by the oxygen inhaler, characterized by: The device comprises an oxygen concentrator, a nasal oxygen cannula, and a blood oxygen saturation tester. The oxygen concentrator is connected to the oxygen inhaler through the nasal oxygen cannula. The nasal oxygen cannula is used to directly connect to the oxygen inhaler to supply oxygen to the oxygen inhaler and collect the respiratory rate and inhaled oxygen concentration of the oxygen inhaler in real time. The blood oxygen saturation tester is connected to the oxygen inhaler and the oxygen concentrator respectively. The blood oxygen saturation tester monitors the blood oxygen saturation and heart rate of the oxygen inhaler. The oxygen concentrator determines whether the current oxygen consumption of the oxygen concentrator needs to be adjusted based on the respiratory rate, inhaled oxygen concentration and blood oxygen saturation of the oxygen inhaler, and can adjust the oxygen output flow of the oxygen concentrator in real time through the electronic flowmeter control module. At the same time, the oxygen output concentration, oxygen output flow device data and the user's blood oxygen saturation of the oxygen concentrator can be exchanged with an external terminal through a wireless communication module. The system also includes an intelligent control processor, which has an embedded lightweight GRU recurrent neural network model. The system also includes an intelligent control processor, which outputs data through an oxygen concentrator, a nasal oxygen cannula, and a blood oxygen saturation tester to predict the ROX trend in the next 3 minutes. If the predicted ROX value is <3.85 or the decline rate is >0.2 / min, the oxygen flow rate is automatically increased or an alarm is triggered.

2. The oxygen production system with automatic adjustment of blood oxygen saturation and oxygen inhalation volume of the oxygen inhaler according to claim 1, characterized in that: According to the blood oxygen saturation and the respiratory rate, the ROX index is calculated, and the oxygenation index is calculated according to the blood oxygen saturation, so as to adjust the oxygen flow of the current oxygen concentrator. The calculated value of the ROX index = (blood oxygen saturation *100) / (inspired oxygen concentration fraction*respiratory rate*heart rate). The ROX index is used to evaluate the ventilation treatment effect within the corresponding ventilation treatment time. The higher the ROX index value, the better the patient's lung oxygenation condition.

3. The oxygen production system with automatic adjustment of blood oxygen saturation and oxygen inhalation volume of the oxygen inhaler as claimed in claim 2, characterized in that: The ROX index is compared with the ROX index threshold to obtain the ROX index comparison result, and the blood oxygen saturation, inspired oxygen concentration fraction, respiratory rate, and heart rate values ​​are compared with the set threshold to obtain the comparison result. According to the comparison result, corresponding ventilation treatment suggestions can be provided to the user, thereby helping medical staff to judge or predict the ventilation effect and prompting medical staff to switch the ventilation treatment mode in time, thereby shortening the patient's ventilation time and improving the prognosis.

4. The oxygen production system with automatic adjustment of blood oxygen saturation and oxygen inhalation volume of the oxygen inhaler as claimed in claim 3, characterized in that: The dead space volume of the normal oropharynx is 50ml, the tidal volume is 500ml, the respiratory rate is 20 times / min, and the amount of pure oxygen inhaled is: 0.5*V (the amount of oxygen in the dead space 0.5s before inspiration) + 1*V (the amount of pure oxygen inhaled during inspiration) + (500-0.5*V-1*V)*20%=1.2*V+100. Since the inspired oxygen fraction is different at different oxygen flow rates, when the oxygen flow rate is between 1-6L / min, the inspired oxygen fraction increases by 0.04 for every 1L / min increase in oxygen flow rate. Therefore, ROX= Blood oxygen saturation / (21+4*V)%, V is the inhaled oxygen flow rate (L / min), 21 represents the oxygen fraction in sea level air, and the ROX index is adjusted by adjusting the inhaled oxygen flow rate to meet the required ROX index value.

5. The oxygen production system with automatic adjustment of blood oxygen saturation and oxygen inhalation volume of the oxygen inhaler according to claim 1, characterized in that: The wireless communication module is one or more of a WIFI module, a Bluetooth module, a NB-IOT module, a LORA module, and an infrared module.

6. The oxygen production system with automatic adjustment of blood oxygen saturation and oxygen inhalation volume of the oxygen inhaler according to claim 1, characterized in that: When the oximeter detects that the user's blood oxygen saturation is abnormal, and the user's blood oxygen saturation is still abnormal after the oxygen concentrator adjusts the oxygen production, the oximeter uploads the information to the external terminal (i.e., the cloud server) through the wireless communication module. At this time, the user or doctor can monitor the user's oxygen inhalation status in real time from the cloud server and remind the user of abnormal content. The cloud server receives and stores the user's real-time ROX trend curve, blood oxygen saturation, respiratory rate, heart rate data, and oxygen flow setting records. After the server AI model identifies the abnormal data features, it automatically pushes specific ventilation mode switching suggestions to the user terminal and the bound medical terminal, including but not limited to increasing the oxygen flow, changing to a high-flow nasal cannula oxygen inhalation mode, or preparing for non-invasive ventilation; the ventilation mode suggestions are pushed to the user and medical staff terminals in the form of text and voice messages at the same time.

7. The oxygen production system with automatic adjustment of blood oxygen saturation and oxygen inhalation volume of the oxygen inhaler according to claim 1, characterized in that: When the oximeter detects an abnormal blood oxygen saturation in the user, and if the abnormality persists even after the oxygen concentrator adjusts the oxygen production, the oximeter triggers an alarm module. The alarm module includes a local buzzer alarm, a flashing LED prompt, and a remote push alarm system. The remote push alarm system communicates with the user's mobile terminal and a pre-set medical staff terminal via the wireless communication module. If the ROX value falls below the set risk threshold for two consecutive prediction cycles, the alarm module issues an alarm via voice broadcast, text message, or app notification, alerting medical staff to promptly initiate emergency medical intervention.

8. The oxygen production system with automatic adjustment of blood oxygen saturation and oxygen inhalation volume of the oxygen inhaler according to claim 1, characterized in that: The oxygen concentrator further comprises a humidifying device. When it is detected that the oxygen at the nasal oxygen tube is relatively dry, the humidifying device is turned on to provide humidified oxygen to the oxygen inhaler.

9. The oxygen production system with automatic adjustment of blood oxygen saturation and oxygen inhalation volume of the oxygen inhaler according to claim 1, characterized in that: A first pressure sensor is installed at the outlet of the nasal oxygen cannula. The first pressure sensor is used to collect the respiratory rate of the oxygen user. The first pressure sensor is specifically a MEMS micro-pressure sensor, which collects the oxygen user's respiratory waveform data every 10 seconds and transmits it to the intelligent control processor. The control processor has a built-in third-order filter and abnormal waveform recognition algorithm to automatically identify abnormal waveform characteristics of Cheyne-Stokes respiration, apnea, and forced breathing. When an abnormal respiratory waveform is detected and the synchronously collected blood oxygen saturation value drops significantly, the intelligent control processor is triggered to perform high oxygen concentration adjustment or ventilation mode change in advance, and the abnormal event and waveform characteristic data are automatically uploaded to the cloud server for storage and filing for subsequent medical analysis.

10. The oxygen production system with automatic adjustment of blood oxygen saturation and oxygen inhalation volume of the oxygen inhaler according to claim 1, characterized in that: The system also includes a lung function individualization correction module for performing individualized correction of the ROX index based on the lung function data input by the user; the lung function individualization correction module includes: The data access unit is configured to receive lung function indicators of the user, including forced vital capacity (FVC) and forced expiratory volume in one second (FEV1), from sources including: Users manually input the FVC and FEV1 values ​​from the hospital test report through the accompanying mobile terminal application; Or connect to an external spirometer and import data via USB, Bluetooth or WIFI; The feature analysis unit is used to combine the FVC and FEV1 values ​​input by the user and their corresponding physiological parameters such as age, gender, and height, call the preset prediction model to calculate the standard FVC and FEV1, and calculate the correction factor μ using the following formula: Where k is the empirical weighting coefficient, ranging from 0.6 to 1.2, and the default value is 1.0; The dynamic adjustment unit receives the correction factor μ and combines it with the real-time acquired blood oxygen saturation, respiratory rate, heart rate, and inspired oxygen concentration data to output the corrected individualized ROX index: The ROX' index is used as one of the input parameters for subsequent oxygen flow adjustment judgment, ventilation mode suggestion and alarm logic.

Images