Respiratory disease oxygen precise conveying system based on sensor

Abstract

The invention discloses a respiratory disease oxygen precise conveying system based on a sensor, and relates to the technical field of medical equipment, and the system comprises a sensing module which integrates multiple sensors to collect and process airway and physiological data of a patient; the disease matching module identifies disease types through database comparison; the dynamic adaptation module calculates exclusive oxygen concentration based on a multi-mode fusion initial algorithm; the conveying module executes adjustment; the closed-loop feedback module continuously fine-tunes the oxygen concentration through a closed-loop stepped fine-tuning algorithm to form a closed-loop link, so that accurate dynamic control of oxygen delivery is realized; according to the invention, multi-sensor collaborative acquisition and disease precise matching are combined with a dynamic algorithm and closed-loop feedback, so that individuation and dynamic oxygen delivery are realized, the system avoids misjudgment by relying on a standardized database, and the oxygen concentration, flow and pressure are ensured to be adjusted in real time along with physiological indexes of a patient through multi-parameter collaborative adjustment and stepped fine adjustment; the treatment comfort, the effectiveness and the long-term adaptability are improved.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, specifically to a sensor-based precise oxygen delivery system for respiratory diseases. Background Technology

[0002] In the treatment of respiratory diseases, oxygen delivery is a key auxiliary means to improve patients' respiratory function and maintain stable vital signs. Its accuracy is directly related to the treatment effect and patient safety. With the development of medical technology towards precision and intelligence, the integration of sensor technology, data processing technology and clinical diagnosis and treatment is becoming increasingly close, providing technical support for the upgrading of oxygen delivery systems. In clinical practice, the airway status and physiological indicators of respiratory patients show dynamic changes. There are significant differences in airway resistance, gas metabolism, blood oxygen level and respiratory rhythm corresponding to different diseases. It is necessary to achieve accurate identification of diseases and personalized adaptation of oxygen supply plans through multi-dimensional indicator monitoring. Standardized pathological feature data based on respiratory clinical diagnosis and treatment guidelines provide a scientific basis for the correlation between diseases and oxygen supply parameters. The maturity of technologies such as multi-sensor synchronous acquisition, data noise reduction processing and signal standardization makes it possible to obtain high-quality patient data in real time, promoting the transformation of oxygen delivery systems from traditional experience-based to data-driven, meeting the urgent clinical demand for precise oxygen supply.

[0003] Traditional oxygen delivery technologies have many limitations and are difficult to adapt to the precision requirements of respiratory disease treatment. In the indicator acquisition stage, traditional systems often suffer from problems such as single sensor types, asynchronous data acquisition, and a lack of effective noise reduction mechanisms. External interference and equipment fluctuations can easily lead to data distortion. At the same time, the inconsistent output signal formats of different sensors increase the difficulty of subsequent data processing. In terms of symptom identification, traditional methods rely heavily on single indicator judgment or manual experience assessment, lacking a scientific judgment logic based on comprehensive comparison of multiple indicators. This can easily lead to misjudgment due to abnormalities in a single indicator and fail to accurately match the actual type of patient's condition. In terms of oxygen supply parameter adjustment, traditional systems lack systematic algorithm support that combines multiple indicators for initial oxygen concentration calculation, resulting in insufficiently targeted parameter settings. Furthermore, the adjustment of oxygen concentration, flow rate, and pressure during delivery lacks coordination and stability. More importantly, traditional technologies generally lack closed-loop feedback mechanisms and cannot respond to the dynamic changes in the patient's physiological indicators in real time. This makes it difficult for oxygen supply parameters to continuously adapt to fluctuations in the patient's condition, which not only affects the treatment effect but may also pose potential safety risks due to insufficient or excessive oxygen supply. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a sensor-based precise oxygen delivery system for respiratory diseases. By integrating sensors for airway resistance, exhaled air, blood oxygen saturation, and respiratory rate, it synchronously collects the patient's physiological data. Combined with a symptom matching module, a dynamic adaptation module, a delivery execution module, and a closed-loop feedback module, it achieves precise adjustment of oxygen concentration. Through multi-index comprehensive comparison for symptom identification, multi-mode fusion initial algorithm, and closed-loop stepwise fine-tuning algorithm, it improves the accuracy of oxygen delivery, meets the individualized needs of patients, enhances treatment efficacy and safety, and has clinical application value.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a sensor-based precise oxygen delivery system for respiratory diseases, the system comprising: Sensor acquisition module: integrates airway resistance sensor, exhaled air... Sensors, including a blood oxygen saturation sensor and a respiratory rate sensor, are used to synchronously collect real-time data on the patient's airway and physiological indicators. The raw data is then denoised and converted before being transmitted synchronously. Symptom matching module: It has a built-in database of respiratory disease pathological features, receives the processed raw data, completes the identification of disease type through comprehensive comparison of multiple indicators, and synchronously outputs the disease type signal and corresponding pre-stored parameters; Dynamic adaptation module: Utilizes a multi-mode fusion initial algorithm to calculate the disease-specific initial oxygen concentration by combining the received disease type signal and parameters, and generates and outputs adjustment instructions; Delivery execution module: It has a built-in oxygen concentration regulating valve, flow controller, and pressure regulator. It receives regulation commands and completes the regulation of oxygen concentration, oxygen flow rate, delivery pressure, and oxygen delivery. Closed-loop feedback module: Using a closed-loop step fine-tuning algorithm, the oxygen concentration after fine-tuning is calculated based on real-time data and fed back to the dynamic adaptation module to update the adjustment command, forming a closed-loop link.

[0006] Furthermore, in the sensing and acquisition module, the airway resistance sensor is embedded inside the ventilator patient interface, fitting snugly against the breathing path, to collect real-time data on changes in airway resistance during the patient's breathing, reflecting information related to the degree of airway patency; the exhaled air... The sensor is connected in series to the ventilator's expiratory circuit tubing to collect real-time data on the carbon dioxide concentration in the patient's exhaled air, capturing the dynamic changes in carbon dioxide concentration during exhalation. The blood oxygen saturation sensor is a finger clip-on non-invasive wearable structure that collects real-time data on the blood oxygen saturation at the patient's fingertip, reflecting information related to the degree of oxygen binding in the blood. The respiratory rate sensor is hardware-linked with the airway resistance sensor, indirectly collecting real-time data on the patient's respiratory rate and corresponding respiratory count information by capturing periodic fluctuations in airway resistance.

[0007] Furthermore, in the sensor acquisition module, when denoising the raw data, the raw data of airway and physiological indicators collected by each sensor are first uniformly collected, and invalid data caused by mechanical vibration of the equipment, patient limb movement, and external environmental interference are screened out. Irregular abnormal fluctuation values ​​in the data are removed, and valid and continuous raw data are retained. Signal conversion processing is carried out on the valid raw data after denoising, converting the analog signals output by each sensor into standard digital signals, completing the unified conversion of different sensor output signal formats, and regularizing the converted digital signals.

[0008] Furthermore, in the symptom matching module, the respiratory symptom pathological feature database is constructed based on clinical diagnosis and treatment guidelines for respiratory diseases. It is a standardized, structured database that specifically pre-stores relevant characteristic parameters for three clinically prevalent respiratory diseases: COPD, asthma, and acute respiratory failure. The database also reserves a parameter update and iteration interface, allowing for the adjustment and optimization of built-in parameters according to clinical treatment guidelines. For each type of symptom, the database stores corresponding airway resistance and expiratory flow rate parameters. The system includes clinical baseline values, upper and lower safety thresholds for four indicators: blood oxygen concentration, blood oxygen saturation, and respiratory rate. The upper safety threshold is the highest permissible value of the corresponding indicator within the normal physiological and disease-adapted range. The lower safety threshold is the lowest permissible value of the corresponding indicator within the normal physiological and disease-adapted range. Simultaneously, it stores the baseline oxygen concentration, disease characteristic correction coefficient, dynamic weights of the four indicators, and fine-tuning sensitivity coefficients for each disease type. All parameters are classified and archived according to disease type to facilitate quick retrieval of corresponding data for comparison during disease matching.

[0009] Furthermore, in the symptom matching module, the specific steps for identifying the symptom type through comprehensive comparison of multiple indicators are as follows: first, receiving the normalized and standardized digital signal, and then extracting airway resistance and exhaled air... Real-time values ​​of four indicators—concentration, blood oxygen saturation, and respiratory rate—are compared using a multi-indicator comprehensive comparison. This comparison is based on the clinical baseline values, upper and lower safety thresholds, and feature matching ranges for each of the four indicators across three disease categories stored in the respiratory disease pathology feature database. Each indicator's real-time value is compared with the corresponding parameter for each disease category to determine if it falls within the feature matching range for that disease. After comparison, the number of indicators falling within the matching range for each disease is counted, prioritizing the disease type with the most falling-range indicators. A second verification is then conducted by assessing the deviation of non-falling-range indicators from the corresponding parameter for that disease to rule out misjudgments caused by a single abnormal indicator. If two disease categories have the same number of falling-range indicators, airway resistance and expiratory flow are used as the determining factor. The degree of matching between concentration indicators and corresponding symptom parameters serves as the final criterion for symptom type identification.

[0010] Furthermore, in the dynamic adaptation module, the mathematical expression for the initial algorithm of multi-mode fusion is:

[0011] in, Initial oxygen concentration specific to the condition; This is the correction coefficient for symptom characteristics; i represents the ordinal number of the four indicators, i=1 corresponds to the airway resistance indicator, i=2 corresponds to the exhaled airway resistance indicator, and so on. Concentration index, i=3 corresponds to blood oxygen saturation index, i=4 corresponds to respiratory rate index; The four indicators are dynamically weighted, with the core indicators being airway resistance and exhaled air. The sum of the weights for concentration is greater than the sum of the weights for blood oxygen saturation and respiratory rate; For the first Real-time values ​​of the indicators; For the first Clinical baseline values ​​for each indicator; For the first The upper limit of the safety threshold for this indicator; For the first The lower limit of the safety threshold for this indicator; To determine the baseline oxygen concentration corresponding to the disease, the algorithm first calls the disease type signal and corresponding pre-stored parameters output by the disease matching module during execution. , , , , and The specific values ​​are then substituted into the real-time values ​​of the four indicators for calculation. After the calculation is completed, the results are normalized to form a definite initial oxygen concentration value. Based on this value, an adjustment command containing oxygen concentration, flow rate, and pressure parameters is generated and output.

[0012] Furthermore, in the aforementioned dynamic adaptation module, the adjustment command is constructed based on the calculation result of the multi-mode fusion initial algorithm, combined with the received symptom type signal and pre-stored parameters in the database. The command content includes a target oxygen concentration value, an adapted symptom identifier, an execution time limit, and a safe adaptation range. The target oxygen concentration value is the symptom-specific initial oxygen concentration calculated by the multi-mode fusion initial algorithm, which determines the oxygen concentration benchmark that the subsequent execution mechanism needs to achieve. The adapted symptom identifier corresponds to the symptom type signal and is used to determine the symptom scenario corresponding to the command. The execution time limit is set according to the symptom type and the patient's real-time index fluctuations to ensure the timeliness and relevance of the command execution. The safe adaptation range is associated with the upper and lower limits of the oxygen concentration safety threshold for the corresponding symptom in the respiratory symptom pathological feature database, serving as boundary constraints during the execution process.

[0013] Furthermore, the delivery execution module incorporates an oxygen concentration regulating valve, a flow controller, and a pressure regulator, which are sequentially deployed along the oxygen delivery path. The oxygen concentration regulating valve is positioned between the oxygen source output and the gas mixing chamber, and is used to adjust the mixing ratio of pure oxygen and diluent gas according to the initial oxygen concentration value specific to the disease in the adjustment command, thereby achieving on-demand adaptation of oxygen concentration. The flow controller is positioned between the gas mixing chamber output and the delivery pipeline, and is used to control the oxygen delivery volume per unit time according to the parameters associated with the adjustment command, ensuring that the flow rate is stable and meets the adaptation requirements of the corresponding disease. The pressure regulator is positioned between the flow controller output and the oxygen delivery interface, and is used to sense the gas pressure in the delivery pipeline in real time, automatically compensate for pressure fluctuations, maintain stable delivery pressure, and ensure pressure consistency during the oxygen delivery process.

[0014] Furthermore, in the delivery execution module, the adjustment of oxygen concentration, oxygen flow rate, and delivery pressure is achieved by first retrieving the initial oxygen concentration value specific to the condition from the adjustment command via the oxygen concentration regulating valve, and adjusting the medium ratio between the main oxygen delivery path and the dilution gas branch to match the oxygen concentration of the output mixed gas with the initial oxygen concentration value specific to the condition; then, the flow controller retrieves the appropriate oxygen flow rate parameter from the adjustment command, and adjusts the delivery flow rate of the mixed gas per unit time based on the oxygen flow rate parameter to ensure that the actual delivery flow rate matches the oxygen flow rate parameter; finally, the pressure regulator retrieves the preset delivery pressure parameter from the adjustment command. The preset delivery pressure parameter of the pressure regulator is the human breathing adaptation pressure range of 0.02MPa-0.06MPa, and performs pressure calibration and stabilization processing on the mixed gas after flow rate adjustment to ensure that the actual delivery pressure at the output end matches the preset delivery pressure parameter. After the oxygen concentration, oxygen flow rate, and delivery pressure are all adjusted to the values ​​corresponding to the adjustment command, the oxygen delivery operation is executed.

[0015] Furthermore, in the aforementioned closed-loop feedback module, the mathematical expression of the closed-loop step fine-tuning algorithm is:

[0016] in, The oxygen concentration after fine-tuning at time t; This represents the time point in the continuous sampling process of the sensor; The initial oxygen concentration is calculated using a multi-mode fusion initial algorithm. To fine-tune the sensitivity coefficient, the value range is 0.01-0.02, and it is set differently according to the type of disease; for Time and -1 moment The change in the four sensing indicators is used as the basis for calculation, with the sensing indicator having the highest weight being selected; i represents the index of the four indicators. The time interval for sensor sampling; This is a safety constraint factor, with a value range of 0.8-1.0.

[0017] Compared with existing technologies, this sensor-based precise oxygen delivery system for respiratory diseases has the following advantages: I. This invention integrates four types of sensors to simultaneously collect real-time data on airway and physiological indicators. After noise reduction to remove invalid interference data and standardization of signal format, it provides high-quality data support for subsequent analysis. Relying on a standardized structured respiratory disease pathological feature database built based on clinical treatment guidelines, it completes disease type identification through multi-indicator comprehensive comparison and secondary verification mechanism, effectively avoiding misjudgment caused by abnormal single indicator. This deep integration of multi-sensor collaborative acquisition and refined disease matching enables accurate identification of common respiratory diseases, allowing oxygen delivery solutions to closely match the patient's actual disease type. It breaks through the limitations of vague disease judgment and insufficient targeted solutions in traditional oxygen delivery. At the same time, the database reserves parameter update and iteration interfaces, which can dynamically optimize built-in parameters according to clinical treatment guidelines, ensuring that the system adapts to clinical treatment needs in the long term, and improving the accuracy and adaptability of oxygen delivery from the source.

[0018] II. This invention calculates the initial oxygen concentration specific to the patient's condition using a multi-mode fusion initial algorithm. It then constructs a comprehensive adjustment command by combining the condition type, execution time, and safety thresholds. The oxygen concentration, oxygen flow rate, and delivery pressure are sequentially adjusted by each component in the delivery execution module according to a reasonable path, achieving multi-parameter collaborative adaptation. A closed-loop stepwise fine-tuning algorithm dynamically calculates the fine-tuned oxygen concentration based on real-time sensor data and feeds it back to the dynamic adaptation module to update the adjustment command, forming a complete closed-loop link. This ensures that oxygen delivery parameters can be adjusted in real-time according to changes in the patient's physiological indicators, avoiding the supply-demand imbalance problem caused by traditional fixed-parameter delivery. It guarantees the safety and stability of oxygen delivery through safety constraints while accurately adapting to individual differences and dynamic changes in the condition of different patients, improving the comfort and effectiveness of oxygen therapy for respiratory patients and achieving dynamic, personalized, and precise oxygen delivery.

[0019] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from an examination of the following, or may be learned from the practice of the invention. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0021] Figure 1 A flowchart of a sensor-based precise oxygen delivery system for respiratory diseases; Figure 2 This is a schematic diagram of data transmission in a sensor-based precise oxygen delivery system for respiratory diseases. Figure 3 This is a schematic diagram of data transmission for the closed-loop feedback module of the present invention. Detailed Implementation

[0022] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0023] Example 1: Daily oxygen supply scenarios for patients with chronic obstructive pulmonary disease Patients with chronic obstructive pulmonary disease (COPD) require long-term home-based rehabilitation due to persistent incomplete airway obstruction. Stable and precise oxygen delivery is crucial for controlling the disease and improving quality of life. When this sensor-based precise oxygen delivery system for respiratory diseases is activated at home, the system quickly enters a fully operational state, with each module working collaboratively to meet the oxygen supply needs of long-term rehabilitation. Figure 1 As shown.

[0024] As the core component of data acquisition, the sensor acquisition module integrates four types of sensors precisely deployed in preset positions to ensure the timeliness and accuracy of simultaneous multi-dimensional data acquisition. The airway resistance sensor is embedded inside the ventilator's patient interface at a key location close to the breathing path. This installation method allows the sensor to directly contact the patient's airflow, capturing real-time data on changes in airway resistance during respiration. This directly reflects the degree of airway patency, a core indicator for assessing the stability of patients with chronic obstructive pulmonary disease (COPD). Continuous monitoring can promptly detect dynamic changes in airway obstruction, providing firsthand information for subsequent oxygen supply adjustments. Exhaled air... The sensor is connected in series to the ventilator's expiratory circuit tubing, seamlessly integrating with the respiratory airflow. It continuously captures dynamic fluctuations in the patient's exhaled carbon dioxide concentration, clearly demonstrating the body's gas exchange efficiency. This prevents the condition from worsening due to undetected carbon dioxide retention, providing crucial information for determining whether the oxygenation regimen is suitable for gas metabolism needs. The blood oxygen saturation sensor uses a finger clip-on, non-invasive wearable structure. Patients do not experience the pain of invasive procedures and it does not affect daily activities such as getting up and eating. It accurately collects blood oxygen saturation data from the fingertip, directly reflecting the actual oxygen binding in the blood. This data is irreplaceable for assessing the patient's hypoxia and adjusting oxygenation intensity. The respiratory rate sensor and airway resistance sensor are designed with hardware linkage. By capturing the regular fluctuations in airway resistance with the respiratory cycle, respiratory rate data is indirectly collected. This linkage ensures the synchronization of respiratory rate and airway resistance data, avoiding time differences or data gaps that may occur with separate collections. It accurately reflects changes in the patient's respiratory rate, providing complete data support for a comprehensive assessment of the patient's condition.

[0025] After the four types of sensors simultaneously collect raw data, the sensor acquisition module immediately initiates the data preprocessing process. First, the raw airway and physiological data collected by each sensor are uniformly aggregated, integrating the scattered multi-source data into a centralized dataset. Then, invalid data caused by mechanical vibrations during equipment operation, limb movements during patient activities at home, and irrelevant interference factors in the external environment are specifically screened out. Simultaneously, irregular abnormal fluctuations in the data are strictly eliminated, retaining only valid and continuous raw data. This processing minimizes interference from factors unrelated to the patient's condition, ensuring the authenticity and reliability of the data used in subsequent analysis and preventing misadjustments to the oxygen supply plan due to false data. For the valid raw data after noise reduction, the system further performs signal conversion processing, uniformly converting the different analog signals output by each sensor into standard digital signals. This completely resolves the incompatibility issue of different sensor signal formats, achieving signal standardization. Subsequently, the converted digital signals are normalized, making the data presentation more standardized and readable. This lays a solid foundation for the rapid retrieval and efficient analysis by the symptom matching module, improving the overall data processing efficiency of the system.

[0026] After data preprocessing, the standardized digital signals are transmitted in real time to the symptom matching module, which immediately initiates the symptom identification process. The module's built-in respiratory symptom pathological feature database is a standardized structured database strictly based on clinical treatment guidelines for respiratory diseases. It specifically stores relevant characteristic parameters for three clinically prevalent respiratory diseases: chronic obstructive pulmonary disease (COPD), asthma, and acute respiratory failure. This ensures that database resources are focused on serving core clinical needs. Simultaneously, the database reserves parameter update and iteration interfaces, allowing for dynamic adjustment and optimization of built-in parameters based on updates to clinical treatment guidelines, ensuring the system remains consistent with the latest clinical standards and possesses long-term applicability. For each symptom, the database stores corresponding airway resistance and exhaled airway pressure. The system stores clinical baseline values, upper and lower safety thresholds for four indicators: blood oxygen concentration, blood oxygen saturation, and respiratory rate. Simultaneously, it stores the baseline oxygen concentration, symptom characteristic correction coefficient, dynamic weights of the four indicators, and fine-tuning sensitivity coefficients for each symptom type. All parameters are categorized and archived by symptom type. This structured storage method allows the module to quickly retrieve corresponding symptom parameters, improving comparison efficiency. After receiving the standardized data, the symptom matching module first extracts the real-time values ​​of the four indicators. Then, based on the clinical baseline values, upper and lower safety thresholds, and feature matching ranges of the four indicators corresponding to chronic obstructive pulmonary disease (COPD) in the database, it initiates a multi-indicator comprehensive comparison process. It precisely compares the real-time values ​​of each of the four indicators with the corresponding indicator parameters for COPD to determine whether the real-time values ​​of each indicator fall within the feature matching range of the symptom. After the comparison is completed, the system automatically counts the number of indicators that fall within the feature matching range for the disease, and prioritizes the chronic obstructive pulmonary disease (COPD) type with the most indicators falling within the range. Then, it conducts a secondary verification by comparing the deviation of the indicators that do not fall within the range with the corresponding indicator parameters for the disease. The deviation range and clinical significance are analyzed in detail to effectively eliminate misjudgments caused by abnormalities in a single indicator. This is because COPD patients may experience temporary deviations in individual indicators due to factors such as short-term activity or emotional fluctuations. The secondary verification can significantly improve the accuracy of disease identification and ensure that the judgment result is highly consistent with the patient's actual condition.

[0027] After the symptom type is confirmed, the symptom matching module synchronously outputs the chronic obstructive pulmonary disease (COPD) type signal and corresponding pre-stored parameters. Upon receiving this information, the dynamic adaptation module immediately initiates the multi-modal fusion initial algorithm, the mathematical expression of which is:

[0028] in, Initial oxygen concentration specific to the condition; This is the correction coefficient for symptom characteristics; i represents the ordinal number of the four indicators. The dynamic weights of the four indicators; For the first Real-time values ​​of the indicators; For the first Clinical baseline values ​​for each indicator; For the first The upper limit of the safety threshold for this indicator; For the first The lower limit of the safety threshold for this indicator; To determine the baseline oxygen concentration corresponding to the specific symptom, the system combines the received symptom type signal with pre-stored parameters in the database, including baseline oxygen concentration, symptom characteristic correction coefficients, and dynamic weights of four indicators, to precisely calculate the patient's unique initial oxygen concentration. This calculation method, based on individual symptoms and real-time indicators, completely eliminates the need for a generic initial oxygen concentration setting, achieving a high degree of adaptation to the patient's specific condition and avoiding the problems of insufficient or excessive oxygen supply that may occur with traditional fixed-concentration oxygen supply. Subsequently, the dynamic adaptation module uses this calculation result as its core, combining the received symptom type signal and pre-stored parameters in the database to construct a complete adjustment command. The command content includes the target oxygen concentration value, the adapted symptom identifier, the execution time, and the safe adaptation range. Among them, the target oxygen concentration value is the disease-specific initial oxygen concentration calculated by the multi-mode fusion initial algorithm, which clarifies the oxygen concentration benchmark that the subsequent execution agency needs to achieve; the adapted disease identifier and disease type signal directly correspond to ensure that the execution agency accurately identifies the disease scenario corresponding to the instruction; the execution time is set according to the characteristics of chronic obstructive pulmonary disease and the patient's real-time index fluctuations, taking into account both stability and flexibility; the safety adaptation range is strictly linked to the upper and lower limits of the safe oxygen concentration threshold for chronic obstructive pulmonary disease in the respiratory disease pathological feature database, setting strict boundary constraints for subsequent oxygen delivery and avoiding the risk of oxygen supply parameters exceeding the safe range from the source.

[0029] Once the adjustment command is generated, it is immediately transmitted to the delivery execution module. This module's built-in oxygen concentration regulating valve, flow controller, and pressure regulator are deployed sequentially along the oxygen delivery path, forming an orderly execution chain. The oxygen concentration regulating valve is positioned between the oxygen source output and the gas mixing chamber. Upon receiving the adjustment command, it quickly retrieves the initial oxygen concentration value specific to the patient's condition and precisely adjusts the mixing ratio of pure oxygen and diluent gas. This ensures that the oxygen concentration of the output mixed gas perfectly matches the initial oxygen concentration value specific to the condition, providing the patient with an appropriate concentration of oxygen, directly improving hypoxia, and avoiding irritation or other discomfort to the respiratory mucosa caused by improper concentration. The flow controller is positioned between the gas mixing chamber output and the delivery pipeline. Based on the oxygen flow parameters associated with the adjustment command, it precisely controls the oxygen delivery volume per unit time, ensuring that the actual delivery flow rate matches the oxygen flow parameters. This adapts to the stable breathing rhythm of patients with chronic obstructive pulmonary disease, ensuring that the patient receives sufficient oxygen with each breath, while avoiding problems such as difficulty breathing due to excessive flow or insufficient flow meeting the patient's needs. The pressure regulator is positioned between the flow controller output and the oxygen delivery interface. It senses changes in gas pressure within the delivery pipeline in real time. Upon detecting pressure fluctuations, it automatically activates a compensation mechanism to quickly adjust the pressure and maintain a stable delivery pressure. This continuous pressure stabilization design allows for smoother breathing for patients, reduces airway discomfort caused by pressure fluctuations, and improves tolerance to long-term oxygen supply. The delivery execution module precisely adjusts each parameter in sequence: first, it regulates the oxygen concentration; then, it controls the flow rate; and finally, it stabilizes the pressure. Once the oxygen concentration, flow rate, and delivery pressure all reach the values ​​corresponding to the adjustment commands, the oxygen delivery operation is officially initiated, providing the patient with stable and appropriate oxygen support.

[0030] Throughout the oxygen delivery process, the closed-loop feedback module remains continuously operational, acquiring real-time airway and physiological data from the patient through continuous sampling by sensors. A closed-loop step-wise fine-tuning algorithm is then applied, the mathematical expression of which is:

[0031] in, The oxygen concentration after fine-tuning at time t; This represents the time point in the continuous sampling process of the sensor; The initial oxygen concentration is calculated using a multi-mode fusion initial algorithm. To fine-tune the sensitivity coefficient; for Time and -1 moment The change in the four sensing indicators; i represents the index of the four indicators; The time interval for sensor sampling; As a safety constraint factor, the changes in four sensing indicators between time t and time t-1 are calculated by combining the fine-tuning sensitivity coefficient, the safety constraint factor, and the sensor sampling time interval, thus obtaining the fine-tuned oxygen concentration at time t. This fine-tuning result is then fed back to the dynamic adaptation module in real time. The dynamic adaptation module updates the adjustment instructions promptly based on the feedback information, and the new adjustment instructions are then transmitted to the delivery execution module to adjust the oxygen delivery parameters, forming a complete closed-loop chain. This continuously looping closed-loop adjustment mechanism can respond in real time to the dynamic changes in the patient's physiological indicators. Whether it's an increase in respiratory rate due to increased patient activity or slight fluctuations in airway resistance, the system can quickly capture and specifically adjust the oxygen delivery parameters, ensuring that oxygen delivery remains highly adapted to the patient's real-time physiological state. This avoids lag in oxygen delivery parameters due to fluctuations in the patient's condition, continuously optimizes the oxygen delivery effect, provides stable and reliable oxygen support for the long-term rehabilitation of patients with chronic obstructive pulmonary disease, effectively reduces the risk of disease exacerbation, and improves the patient's quality of rehabilitation and comfort. Figure 3 As shown.

[0032] In this embodiment, the system addresses the stable oxygen supply needs of patients with chronic obstructive pulmonary disease (COPD) undergoing home rehabilitation. It simultaneously collects airway and physiological indicators from multiple sensors, ensuring data quality through noise reduction, signal conversion, and normalization. Utilizing a structured database and comprehensive comparison of multiple indicators, it accurately identifies symptoms. A multi-modal fusion initial algorithm calculates a specific initial oxygen concentration, and orderly deployed execution components precisely adjust oxygen concentration, flow rate, and pressure. Finally, a closed-loop step-wise fine-tuning algorithm forms a continuously optimized closed-loop system. The entire process achieves full automation and precision in data acquisition, symptom identification, oxygen supply adaptation, and dynamic adjustment. This effectively adapts to the long-term rehabilitation needs of patients, avoids the risks of indiscriminate oxygen supply, improves the safety and comfort of home rehabilitation, and provides reliable oxygen support for the long-term treatment of COPD patients.

[0033] Example 2: Emergency oxygen supply scenario for patients with acute asthma attacks An asthma patient experiencing a sudden acute attack during emergency room visits suffers from rapid airway spasm and constriction, resulting in urgent symptoms such as difficulty breathing and hypoxia. In such cases, rapid activation of precise oxygen delivery support is crucial to buy time and lay the foundation for resuscitation. Medical staff immediately activate this sensor-based precise oxygen delivery system for respiratory diseases. The system quickly enters emergency mode, with all modules working efficiently and collaboratively to meet the timeliness and precision requirements of emergency resuscitation. Figure 2 As shown.

[0034] The sensor acquisition module responds rapidly in emergency scenarios, with four integrated sensors simultaneously initiating data acquisition to ensure the acquisition of key physiological indicators of the patient in the shortest possible time. The airway resistance sensor is embedded inside the ventilator's patient interface, fitting snugly against the breathing path. Once the patient wears the interface, the sensor immediately comes into direct contact with the airflow, acquiring real-time data on sudden changes in airway resistance caused by airway spasm. This rapid change in airway resistance is one of the core characteristics of an acute asthma attack; quickly capturing this data allows medical staff to immediately assess the severity of airway spasm, providing crucial information for determining the urgency of the condition. Exhaled air... The sensor is connected in series to the ventilator's expiratory circuit tubing, creating a real-time linkage with the patient's exhalation process. It rapidly captures drastic fluctuations in the carbon dioxide concentration in exhaled air. Changes in carbon dioxide concentration directly reflect whether gas exchange in the body is obstructed. Timely monitoring can effectively determine the risk of carbon dioxide retention, preventing it from worsening the patient's condition. This provides crucial information for adjusting oxygen supply and developing treatment plans during emergency care. The blood oxygen saturation sensor uses a finger clip-on, non-invasive wearable structure, allowing medical staff to quickly put it on the patient without complicated operations, saving time in emergency rescue. Simultaneously, the sensor can monitor changes in blood oxygen saturation at the patient's fingertip in real time. During acute asthma attacks, patients often experience a rapid drop in blood oxygen levels; real-time monitoring of blood oxygen levels allows for accurate assessment of the degree of hypoxia, providing a direct basis for setting emergency oxygen supply intensity. The respiratory rate sensor and airway resistance sensor are linked in hardware. By capturing the rapid periodic fluctuations in airway resistance, the system indirectly collects real-time data on the patient's respiratory rate. This linked acquisition method ensures the synchronization of respiratory rate data and airway resistance data, avoids the delay that may occur with separate acquisition, accurately reflects the patient's shortness of breath caused by airway spasm, and provides complete data support for a comprehensive assessment of the severity of the patient's condition.

[0035] Because the patient was in an acute state, their limbs were struggling and moving to some extent. Furthermore, equipment in the emergency environment might experience slight vibrations due to handling and movement. Therefore, after acquiring the raw data, the sensor acquisition module immediately initiated a targeted data preprocessing process. First, the raw airway and physiological data collected by each sensor were uniformly aggregated, integrating the multi-source, scattered data into a centralized dataset. Then, invalid interfering data caused by the patient's limb movements was carefully filtered out, along with abnormal fluctuations caused by mechanical vibrations during equipment handling. Only valid and continuous raw data were strictly retained. This processing effectively filters out various interfering factors in the emergency scenario, ensuring that the data used for subsequent analysis accurately reflects the patient's condition and avoids misjudgments of oxygenation plans due to interfering data, thus providing data assurance for the accuracy of emergency resuscitation. After denoising, the system quickly performs signal conversion processing on the valid raw data, converting the analog signals output by each sensor into standard digital signals. This completely solves the problem of incompatibility between different sensor signal formats and achieves signal standardization. Subsequently, the converted digital signals are standardized to make the data presentation more standardized and readable, ensuring that the symptom matching module can quickly call and efficiently analyze the data, meeting the timeliness requirements of emergency rescue and gaining valuable time for rapid symptom identification and oxygen supply activation.

[0036] After data preprocessing, the standardized digital signals are immediately transmitted to the symptom matching module, which quickly initiates the emergency symptom identification process. The module's built-in respiratory symptom pathological feature database is built based on clinical treatment guidelines for respiratory diseases. It is a standardized, structured database that specifically stores relevant characteristic parameters for three clinically prevalent respiratory diseases: chronic obstructive pulmonary disease (COPD), asthma, and acute respiratory failure. This ensures rapid retrieval of core symptom data in emergency scenarios, improving identification efficiency. Simultaneously, the database reserves a parameter update and iteration interface, allowing for adjustments and optimization of built-in parameters according to clinical treatment guidelines, ensuring the system always conforms to the latest clinical standards. For each symptom category, the database stores corresponding airway resistance and exhaled air... The system stores the clinical baseline values, upper and lower safety thresholds for four indicators: blood oxygen concentration, blood oxygen saturation, and respiratory rate. Simultaneously, it stores the baseline oxygen concentration, symptom characteristic correction coefficient, dynamic weights of each of the four indicators, and fine-tuning sensitivity coefficients for each symptom type. All parameters are categorized and archived by symptom type for easy retrieval and comparison by the module. After receiving the normalized data, the symptom matching module quickly extracts the real-time values ​​of the four indicators. Based on the clinical baseline values, upper and lower safety thresholds, and feature matching ranges for the four indicators corresponding to asthma in the database, it initiates a multi-indicator comprehensive comparison process. It precisely compares the real-time values ​​of each of the four indicators with the corresponding asthma indicator parameters to determine whether the real-time values ​​of each indicator fall within the feature matching range for asthma. Because the patient is in an acute attack state, some indicators may rapidly deviate from the normal range due to the rapid changes in the condition. After comparison, the system statistics show that two symptom types have the same number of indicators falling within the feature matching range. At this time, the module immediately compares airway resistance and expiratory oxygen... The degree of matching between the concentration index and the corresponding parameters for asthma is used as the final criterion, because during an acute asthma attack, airway resistance rises sharply and exhaled air... Drastic fluctuations in concentration are the most specific indicator. Based on this, interference from other diseases can be quickly eliminated, and acute asthma attacks can be accurately identified. Subsequently, the corresponding disease type signal and pre-stored parameters are output simultaneously, providing key support for the rapid activation of the adaptive oxygen supply plan.

[0037] Once the symptom type is confirmed, the asthma acute attack type signal and corresponding pre-stored parameters output by the symptom matching module are transmitted in real time to the dynamic adaptation module. This module immediately initiates the emergency adaptation process, employing a multi-mode fusion initial algorithm. Combining the received symptom type signal with pre-stored parameters such as baseline oxygen concentration, symptom feature correction coefficients, and dynamic weights of four indicators, it quickly calculates a symptom-specific initial oxygen concentration suitable for this acute attack scenario. This rapid calculation method based on acute attack symptoms and real-time indicators allows for precise matching of the initial oxygen concentration setting to the patient's emergency condition, avoiding the problem of general concentrations failing to quickly alleviate hypoxia and buying time for emergency rescue. Subsequently, the dynamic adaptation module uses this calculation result as its core, combining the received symptom type signal and pre-stored parameters to quickly construct a complete adjustment command. The command content includes the target oxygen concentration value, the adapted symptom identifier, the execution time, and the safe adaptation range. Among them, the target oxygen concentration value is the initial oxygen concentration specific to the disease calculated by the multi-mode fusion initial algorithm, which clarifies the oxygen concentration benchmark that the subsequent implementing agency needs to quickly reach; the adapted disease identifier directly corresponds to the acute asthma attack type signal, ensuring that the implementing agency accurately identifies the emergency scenario corresponding to the instruction; the execution time is set to short-term immediate effect according to the urgency of the acute asthma attack, meeting the timeliness requirements of emergency rescue; the safety adaptation range is strictly linked to the upper and lower limits of the safe oxygen concentration threshold for asthma in the respiratory disease pathological feature database, setting strict boundary constraints for emergency oxygen supply and avoiding secondary risks caused by improper oxygen supply during emergency rescue.

[0038] Once the adjustment command is generated, it is immediately transmitted to the delivery execution module. This module's built-in oxygen concentration regulating valve, flow controller, and pressure regulator are deployed sequentially along the oxygen delivery path, rapidly initiating the emergency execution process. The oxygen concentration regulating valve, located between the oxygen source output and the gas mixing chamber, quickly retrieves the initial oxygen concentration value specific to the patient's condition upon receiving the adjustment command. It then rapidly adjusts the mixing ratio of pure oxygen and diluent gas to ensure that the output mixed gas oxygen concentration precisely matches the initial oxygen concentration value specific to the condition. This quickly alleviates the severe hypoxia caused by the patient's acute attack, providing support for stabilizing vital signs. The flow controller, located between the gas mixing chamber output and the delivery pipeline, rapidly regulates the mixed gas delivery flow rate per unit time based on the oxygen flow parameters associated with the adjustment command. This ensures that the actual delivery flow rate matches the oxygen flow parameters, adapting to the rapid breathing rhythm of patients during acute asthma attacks. This ensures that the patient receives sufficient oxygen with each breath, avoiding the problem of insufficient flow to meet the rapid breathing needs. The pressure regulator is positioned between the flow controller output and the oxygen delivery interface. Based on the human breathing pressure range, it rapidly calibrates and stabilizes the pressure of the mixed gas after flow regulation. It senses real-time pressure changes in the delivery pipeline and automatically compensates for any pressure fluctuations, maintaining stable delivery pressure. This rapid pressure stabilization design prevents excessive pressure from irritating already constricted airways and worsening patient discomfort, while ensuring smooth breathing and improving patient tolerance during emergency care. The delivery module adjusts oxygen concentration first, then flow rate, and finally pressure, rapidly and precisely regulating each parameter. Once oxygen concentration, flow rate, and delivery pressure all reach the values ​​corresponding to the adjustment commands, it immediately delivers oxygen to the patient, providing crucial support for the rescue of patients experiencing acute asthma attacks.

[0039] During emergency oxygen supply, the closed-loop feedback module maintains a high-frequency operating state. Utilizing a closed-loop step-adjustment algorithm, it continuously calculates the changes in four sensor indicators between time t and time t-1 based on real-time patient data obtained through continuous sensor sampling. Combining the fine-tuning sensitivity coefficient, safety constraint factor, and sensor sampling time interval, it quickly derives the fine-tuned oxygen concentration at time t. This fine-tuning result is then fed back to the dynamic adaptation module in real time. The dynamic adaptation module immediately updates the adjustment instructions based on the feedback information, and the new adjustment instructions are quickly transmitted to the delivery execution module to promptly adjust oxygen delivery parameters, forming an efficient closed-loop link. This high-frequency closed-loop adjustment mechanism can respond in real time to the rapid dynamic changes in the physiological indicators of patients with acute asthma attacks. Whether it's the decrease in resistance due to the relief of airway spasm or the metabolic changes brought about by the increase in blood oxygen levels, the system can quickly capture and specifically adjust the oxygen delivery parameters, ensuring that oxygen delivery always maintains a high degree of adaptation to the patient's real-time condition. This avoids lag in oxygen delivery parameters due to rapid changes in the patient's condition, continuously optimizes the emergency oxygen supply effect, creates stable vital sign conditions for subsequent treatment by medical staff, effectively improves the success rate of rescuing patients with acute asthma attacks, and reduces the risk of disease deterioration.

[0040] This embodiment focuses on the emergency rescue scenario of patients with acute asthma attacks. The system responds quickly to emergency needs by collecting key indicators at high frequency through multiple sensors, selectively filtering interfering data in the emergency environment, accurately identifying acute attack symptoms based on specific indicator matching, rapidly calculating the appropriate oxygen concentration using a multi-mode fusion initial algorithm, quickly adjusting oxygen supply parameters through execution components, and then adapting to fluctuations in the patient's condition in real time using a high-frequency closed-loop fine-tuning algorithm. The entire process balances the timeliness and accuracy of emergency rescue, quickly alleviating the patient's hypoxia symptoms, avoiding the aggravation of the condition due to improper oxygen supply, buying valuable time for rescue work, improving the success rate of rescuing patients with acute asthma attacks, and fully demonstrating the practical value and reliable performance of the system in emergency medical scenarios.

[0041] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A sensor-based precision oxygen delivery system for respiratory diseases, characterized in that, The system includes: Sensing and collecting module: integrated airway resistance sensor, exhaled gas sensor, blood oxygen saturation sensor, respiratory rate sensor, for synchronous collection of patient airway index and physiological index real-time data, denoising and signal conversion of original data, and synchronous transmission; Symptom matching module: It has a built-in respiratory disease pathological feature database, receives the processed raw data, completes the disease type identification through comprehensive comparison of multiple indicators, and synchronously outputs the disease type signal and corresponding pre-stored parameters; Dynamic adaptation module: Utilizes a multi-mode fusion initial algorithm to calculate the disease-specific initial oxygen concentration by combining the received disease type signal and parameters, and generates and outputs adjustment instructions; Delivery execution module: It has a built-in oxygen concentration regulating valve, flow controller, and pressure regulator. It receives regulation commands and completes the regulation of oxygen concentration, oxygen flow rate, delivery pressure, and oxygen delivery. Closed-loop feedback module: Using a closed-loop step fine-tuning algorithm, the oxygen concentration after fine-tuning is calculated based on real-time data and fed back to the dynamic adaptation module to update the adjustment command, forming a closed-loop link.

2. The sensor-based precise oxygen delivery system for respiratory diseases according to claim 1, characterized in that, In the sensing and collecting module, the airway resistance sensor is embedded in the inner side of the ventilator patient interface and adheres to the breathing airway to collect real-time data of the resistance change in the airway during the patient's breathing process and reflect information related to the airway patency; the exhaled gas The sensor is connected in series to the ventilator exhalation circuit pipeline to collect real-time data of the carbon dioxide concentration in the patient's exhaled gas, capture the dynamic change of the carbon dioxide concentration during exhalation, and the blood oxygen saturation sensor is a non-invasive wearable structure of a finger clip type to collect real-time data of the blood oxygen saturation of the patient's fingertip and reflect information related to the oxygen binding degree in the blood. The respiratory rate sensor and the airway resistance sensor are linked in hardware. By capturing the periodic fluctuation data of airway resistance, the sensor indirectly collects the patient's real-time respiratory rate data and related information on the number of breaths.

3. The sensor-based precise oxygen delivery system for respiratory diseases according to claim 1, characterized in that, In the sensor acquisition module, when denoising the raw data, the raw data of airway and physiological indicators collected by each sensor are first uniformly collected, and invalid data caused by mechanical vibration of equipment, patient limb movement, and external environmental interference are screened out. Irregular abnormal fluctuation values ​​in the data are removed, and valid and continuous raw data are retained. Signal conversion processing is carried out on the valid raw data after denoising, converting the analog signals output by each sensor into standard digital signals, completing the unified conversion of different sensor output signal formats, and regularizing the converted digital signals.

4. The sensor-based precise oxygen delivery system for respiratory diseases according to claim 1, characterized in that, In the disease matching module, the respiratory disease pathological feature database is constructed based on clinical diagnosis and treatment guidelines for respiratory diseases. It is a standardized, structured database that specifically pre-stores relevant feature parameters for three clinically prevalent respiratory diseases: COPD, asthma, and acute respiratory failure. The database also reserves a parameter update and iteration interface, allowing for the adjustment and optimization of built-in parameters according to clinical treatment guidelines. For each type of disease, the database stores corresponding airway resistance and expiratory flow rate parameters. The system stores the clinical baseline values, upper and lower safety thresholds for four indicators: blood oxygen concentration, blood oxygen saturation, and respiratory rate. It also stores the baseline oxygen concentration, disease characteristic correction coefficient, dynamic weights of the four indicators, and fine-tuning sensitivity coefficients for each disease type. All parameters are archived according to disease type.

5. A sensor-based precise oxygen delivery system for respiratory diseases according to claim 1, characterized in that, In the symptom matching module, the specific steps for identifying symptom types through comprehensive comparison of multiple indicators are as follows: first, receive the normalized and standardized digital signal, and then extract airway resistance and exhaled air... Real-time values ​​of four indicators—concentration, blood oxygen saturation, and respiratory rate—are compared using a multi-indicator comprehensive comparison. This comparison is based on the clinical baseline values, upper and lower safety thresholds, and feature matching ranges for each of the four indicators across three disease categories stored in the respiratory disease pathology feature database. Each indicator's real-time value is compared with the corresponding parameter for each disease category to determine if it falls within the feature matching range for that disease. After comparison, the number of indicators falling within the matching range for each disease is counted, prioritizing the disease type with the most falling-range indicators. A second verification is then conducted by comparing the deviation of non-falling-range indicators with the corresponding parameter for the disease with the most falling-range indicators, eliminating misjudgments caused by a single abnormal indicator. If the number of falling-range indicators is the same for two disease categories, airway resistance and expiratory flow rate are used as the determining factors. The degree of matching between concentration indicators and corresponding symptom parameters serves as the final criterion for symptom type identification.

6. The sensor-based precise oxygen delivery system for respiratory diseases according to claim 1, characterized in that, In the dynamic adaptation module, the mathematical expression for the initial algorithm of multi-mode fusion is: in, Initial oxygen concentration specific to the condition; This is the correction coefficient for symptom characteristics; i represents the ordinal number of the four indicators. The dynamic weights of the four indicators; For the first Real-time values ​​of the indicators; For the first Clinical baseline values ​​for each indicator; For the first The upper limit of the safety threshold for each indicator; For the first The lower limit of the safety threshold for this indicator; This represents the baseline oxygen concentration corresponding to the specific symptom.

7. A sensor-based precise oxygen delivery system for respiratory diseases according to claim 1, characterized in that, In the dynamic adaptation module, the adjustment command is constructed based on the calculation result of the multi-mode fusion initial algorithm, combined with the received symptom type signal and pre-stored parameters in the database. The command content includes the target oxygen concentration value, the adapted symptom identifier, the execution time, and the safe adaptation range. The target oxygen concentration value is the symptom-specific initial oxygen concentration calculated by the multi-mode fusion initial algorithm, which determines the oxygen concentration benchmark that the subsequent execution mechanism needs to achieve. The adapted symptom identifier corresponds to the symptom type signal and is used to determine the symptom scenario corresponding to the command. The execution time is set according to the symptom type and the patient's real-time index fluctuations. The safe adaptation range is associated with the upper and lower limits of the oxygen concentration safety threshold for the corresponding symptom in the respiratory symptom pathological feature database, serving as boundary constraints during the execution process.

8. A sensor-based precise oxygen delivery system for respiratory diseases according to claim 1, characterized in that, In the delivery execution module, an oxygen concentration regulating valve, a flow controller, and a pressure regulator are deployed sequentially along the oxygen delivery path. The oxygen concentration regulating valve is deployed between the oxygen source output and the gas mixing chamber, and is used to adjust the mixing ratio of pure oxygen and diluent gas according to the initial oxygen concentration value specific to the disease in the adjustment command. The flow controller is deployed between the gas mixing chamber output and the delivery pipeline, and is used to control the oxygen delivery volume per unit time according to the parameters associated with the adjustment command. The pressure regulator is deployed between the flow controller output and the oxygen delivery interface, and is used to sense the gas pressure in the delivery pipeline in real time, automatically compensate for pressure fluctuations, and maintain stable delivery pressure.

9. A sensor-based precise oxygen delivery system for respiratory diseases according to claim 1, characterized in that, In the delivery execution module, the adjustment of oxygen concentration, oxygen flow rate and delivery pressure is completed by first retrieving the disease-specific initial oxygen concentration value in the adjustment command through the oxygen concentration adjustment valve, and adjusting the medium ratio between the main oxygen delivery line and the dilution gas branch line so that the oxygen concentration of the output mixed gas matches the disease-specific initial oxygen concentration value. Next, the flow controller retrieves the appropriate oxygen flow parameters from the adjustment command, and adjusts the delivery flow rate of the mixed gas per unit time based on the oxygen flow parameters to make the actual delivery flow rate consistent with the oxygen flow parameters. Finally, the pressure regulator retrieves the preset delivery pressure parameters from the adjustment command. The preset delivery pressure parameters of the pressure regulator are within the human breathing pressure range of 0.02MPa-0.06MPa. The mixed gas after flow regulation is pressure calibrated and stabilized to make the actual delivery pressure at the output end match the preset delivery pressure parameters. After the oxygen concentration, oxygen flow rate, and delivery pressure are all adjusted to the values ​​corresponding to the adjustment command, the oxygen delivery operation is executed.

10. A sensor-based precise oxygen delivery system for respiratory diseases according to claim 1, characterized in that, In the closed-loop feedback module, the mathematical expression of the closed-loop step fine-tuning algorithm is: in, The oxygen concentration after fine-tuning at time t; This represents the time point in the continuous sampling process of the sensor; The initial oxygen concentration is calculated using a multi-mode fusion initial algorithm. To fine-tune the sensitivity coefficient; for Time and -1 moment The change in the four sensing indicators; i represents the index of the four indicators; The time interval for sensor sampling; This is a safety constraint factor.

Images