System for detecting cardiopulmonary condition of patient with chronic obstructive pulmonary disease based on virtual reality

Abstract

The invention discloses a heart and lung condition detection system for a patient with chronic obstructive pulmonary disease based on virtual reality, and belongs to the technical field of intelligent medical treatment. The problem that the cardiopulmonary state of a patient in daily activities is difficult to comprehensively reflect in the prior art is solved, data packet loss is found in time by detecting the time interval of the data stream, and data transmission errors are reduced in combination with monitoring of the data transmission rate; data are analyzed in combination with physiological rules, and untrusted data are removed; therefore, a reliable data basis is provided for cardiopulmonary function evaluation, and the rehabilitation effect can be improved. A multi-dimensional respiratory cycle portrait of a chronic obstructive pulmonary disease patient is analyzed in real time by constructing a cardiopulmonary function prediction model, and an abnormal mode deviating from a baseline is accurately recognized; a user behavior analysis mechanism is introduced to evaluate the psychological state and the rehabilitation effect, personalized early warning information and rehabilitation suggestions are generated, it is ensured that a patient can be intervened in time when abnormity occurs, and meanwhile the monitoring burden of medical staff is relieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent medical treatment, in particular to a chronic obstructive pulmonary disease patient cardiopulmonary condition detection system based on virtual reality. BACKGROUND

[0002] Chronic obstructive pulmonary disease is a chronic respiratory disease characterized by persistent airflow obstruction, characterized by persistent airflow limitation and corresponding respiratory symptoms, often accompanied by cardiopulmonary dysfunction, which seriously affects the quality of life and exercise tolerance of patients.

[0003] Traditional detection methods are mostly static detection, such as lung function test, electrocardiogram, etc., but it is difficult to fully reflect the cardiopulmonary state of patients in daily activities, and cannot effectively improve the rehabilitation effect of patients.

[0004] Therefore, the present application does not meet the existing needs, and a chronic obstructive pulmonary disease patient cardiopulmonary condition detection system based on virtual reality is proposed. SUMMARY

[0005] The present application aims to provide a chronic obstructive pulmonary disease patient cardiopulmonary condition detection system based on virtual reality, which can detect data packet loss in time by detecting the time interval of data stream, combine real-time monitoring data transmission rate to reduce data transmission error, and analyze data combined with physiological rules to eliminate untrusted data, thereby providing reliable data basis for cardiopulmonary function assessment, which helps to improve the rehabilitation effect; by constructing a cardiopulmonary function prediction model, the multidimensional respiratory cycle image of chronic obstructive pulmonary disease patients is analyzed in real time, and the abnormal mode deviating from the baseline is accurately identified; and the psychological state and rehabilitation effect are evaluated by introducing a user behavior analysis mechanism, personalized warning information and rehabilitation suggestions are generated, ensuring that patients can be intervened in time when abnormalities occur, while reducing the monitoring burden of medical staff, solving the problems raised in the above background technology.

[0006] To achieve the above-mentioned purpose, the present application provides the following technical scheme: The chronic obstructive pulmonary disease patient cardiopulmonary condition detection system based on virtual reality comprises: A multidimensional data acquisition unit configured to acquire cardiopulmonary physiological data of a user in real time based on an intelligent respiratory sensor, an oximeter, an electromyographic sensor, and a thoracic impedance sensor, and transmit the data to a virtual reality interaction unit for prediction; A data verification unit configured to perform integrity verification on the acquired cardiopulmonary physiological data by using time sequence continuity check and transmission rate monitoring to determine whether the data has packet loss; and align and fuse the data to generate a multidimensional respiratory cycle image; The virtual reality interaction unit is configured to construct a three-dimensional digital lung model based on the cardiopulmonary physiological data, construct an immersive virtual environment to which the three-dimensional digital lung model belongs based on a VR device, and convert the breathing behavior of a user into an interaction instruction in the virtual environment to drive changes in visual and auditory elements of the virtual environment, thereby forming a biological feedback loop. The cardiopulmonary warning unit is configured to construct a cardiopulmonary function prediction model based on the multi-dimensional respiratory cycle image, identify abnormal patterns deviating from a baseline from the cardiopulmonary function prediction model, determine a standard feature vector in a normal breathing state by monitoring a breathing feature vector in an abnormal breathing state of the user and combining a cardiopulmonary function curve, calculate a confidence degree of cardiopulmonary function degradation caused by breathing abnormalities, determine a cardiopulmonary function abnormality misjudgment condition and give a warning, and prompt the user in a preset visual metaphor manner through the virtual reality interaction unit.

[0007] Further, the data verification unit comprises: The packet loss verification module is configured to extract a timestamp sequence of each data stream in the cardiopulmonary physiological data, and calculate a time interval between consecutive data packets. If the interval is greater than 2 times the sampling period, it is determined that there is data packet loss in the time period, and a warning prompt and error correction processing are performed. An adaptive error correction algorithm is introduced to automatically repair the lost data packets. The rate verification module is configured to obtain the acquisition frequency of the historical data and the corresponding normal transmission rate, and preset a threshold based on the historical normal transmission rate. A dynamic threshold adjustment mechanism is introduced to automatically adjust the transmission rate threshold according to the real-time network condition. The real-time monitoring multi-dimensional data acquisition unit is configured to periodically compare the real-time transmission rate of each channel with the threshold to determine whether the real-time transmission rate of each channel is abnormal. If the rate of any channel continuously falls below the preset threshold, a data transmission abnormality warning is triggered.

[0008] Further, the data verification unit further comprises: The alignment and fusion module is configured to use a unified clock source as a reference to align the timestamp data of all channels in the time domain, use the inspiration start event of the intelligent respiratory sensor as a reference point to slice and fuse all data in the respiratory cycle, correlate the multi-dimensional representation of complete physiological events, and generate a multi-dimensional respiratory cycle image. The logic verification module is configured to preset an analysis period based on the data acquisition frequency, perform sliding window analysis on the fused data according to the preset period, set a logic judgment rule based on physiological rules, determine whether the cardiopulmonary physiological data is abnormal, further identify untrusted data caused by motion artifacts or device failure, and obtain a multi-dimensional respiratory cycle image that passes the verification.

[0009] Further, the cardiopulmonary warning unit comprises: a model construction module configured to collect historical data and known acute exacerbation cases, extract features related to cardiopulmonary function therefrom, train the features based on a neural network model, and construct a prediction model capable of identifying abnormal patterns; a baseline construction module configured to obtain multi-dimensional respiratory cycle portraits over a period of time, and establish a user's cardiopulmonary function baseline based on the multi-dimensional respiratory cycle portraits in a normal state, the baseline data including characteristic parameters of normal respiratory cycles; classify abnormal patterns based on multi-dimensional respiratory cycle portraits in an abnormal state, including mild abnormality, moderate abnormality, and severe abnormality; real-time receive multi-dimensional respiratory cycle portraits of a user, compare real-time cardiopulmonary physiological data with baseline data, and identify whether there is an abnormal pattern deviating from the baseline; analyze the identified abnormal pattern, determine its possible cause and risk level, and generate an acute exacerbation risk warning signal according to the analysis result of the abnormal pattern, the warning signal including a warning level and corresponding recommended measures.

[0010] Further, classifying abnormal patterns based on multi-dimensional respiratory cycle portraits in an abnormal state further includes: introducing a multi-modal data fusion technology, combining cardiopulmonary physiological data and user behavior data, and analyzing whether the user's behavior data in a virtual environment is compatible with the cardiopulmonary physiological data; introducing a user behavior analysis mechanism to analyze the user's behavior patterns and reactions in a virtual environment; generate personalized warning information and rehabilitation recommendations based on the user's behavior analysis results, and real-time feedback to medical staff.

[0011] Further, the virtual reality interaction unit includes: a three-dimensional construction module configured to initialize a three-dimensional digital lung model based on pre-processed data, the model including alveoli, airways, and lung lobe structures, and set the initial state of each structure according to the user's physiological data; real-time receive cardiopulmonary physiological data, dynamically update the three-dimensional digital lung model, adjust the inflation and contraction of alveoli according to respiratory rate and tidal volume, adjust the density and speed of virtual particle flow in the airway according to peak expiratory flow rate, and adjust the color of alveoli according to blood oxygen saturation; if abnormal data is detected, a warning is issued through the transparency or color of the corresponding area of the model; ​The scene building module is configured to create an immersive virtual environment by using a VR device, preset a plurality of virtual application scenes, including a rehabilitation game scene, a mirror observation scene and a meditation guide scene; according to initial physiological data and preference settings of a user, the virtual environment is initialized, and visual, auditory and tactile elements of the virtual environment are set to ensure that the user can naturally interact with the environment; The breathing behavior data of the user is received in real time, and the breathing behavior is converted into an interaction instruction in the virtual environment; according to the breathing rhythm and depth of the user, the visual and auditory elements in the virtual environment are dynamically adjusted.

[0012] Further, the virtual reality interaction unit further comprises: The instruction generation module is configured to convert the breathing behavior data of the user into an interaction instruction in the virtual environment, and input the interaction instruction through a handle or gesture recognition of the VR device; The data recording module is configured to record long-term physiological data of the user and performance in the virtual environment, perform data analysis and trend prediction, and provide a personalized rehabilitation training scheme for the user according to the long-term data.

[0013] Further, the virtual reality interaction unit further comprises: The environment feedback module is configured to drive changes in visual and auditory elements in the virtual environment according to the interaction instruction of the user; and dynamically adjust the feedback intensity and content according to the real-time physiological data of the user and the performance in the virtual environment; The physiological feedback module is configured to adjust the visual, auditory and tactile feedback in the virtual environment in real time by monitoring the cardiopulmonary physiological data of the user and the interaction behavior in the virtual environment in real time; and guide the user to adjust the breathing behavior through a biological feedback mechanism, and dynamically update the state of the three-dimensional digital lung model and the virtual environment according to the adjustment of the user.

[0014] Further, after the three-dimensional digital lung model is built, further comprising: Real-time acquisition of cardiopulmonary physiological data of the user, the cardiopulmonary physiological data including airway pressure, respiratory flow, tidal volume and blood oxygen saturation; A physics engine is integrated, the physics engine being pre-defined with a biomechanical model of a virtual lung tissue, the biomechanical model including an elastic resistance parameter for simulating the elasticity of alveoli and airways, a viscous resistance parameter for simulating airflow friction resistance, and an inertial mass parameter for simulating the overall inertia of the lung; The real-time acquired cardiopulmonary physiological data is input as a boundary condition to the physics engine, and by solving the dynamic equation of the biomechanical model, the three-dimensional digital lung model is driven in real time to produce deformation and dynamic effects consistent with the real physiological process of the user; Based on real-time airway pressure and respiratory flow data, the fluid dynamics effects of virtual particle flow in the virtual airway are calculated by a physical engine; According to the calculated flow rate and pressure difference, the flow state of the virtual particle flow is adjusted in real time, including laminar flow, turbulent flow or vortex flow state, and is visually presented in the airway structure of the three-dimensional digital lung model; When the calculated flow state value exceeds the preset turbulent flow threshold, a turbulent flow or vortex flow visual effect is generated and rendered in the corresponding airway region of the three-dimensional digital lung model; Based on the tidal volume and airway pressure data, the volume change rate and stress distribution of the alveolar unit in the three-dimensional digital lung model are calculated in real time by the physical engine; According to the volume change rate and stress distribution, the three-dimensional geometry of each alveolar unit is dynamically updated to simulate the inflation and deflation process of the alveoli; When real-time cardiopulmonary physiological data indicates an abnormal state, according to the type and severity of the abnormality, the physical parameters of the local region in the biomechanical model are modified by the physical engine and the three-dimensional digital lung is updated in stages.

[0015] Further, the abnormal breathing state of the user is monitored, the current breathing feature vector under the abnormal breathing state is obtained, and the standard breathing feature vector of the user under the normal breathing state is determined according to the cardiopulmonary function curve; The variable frequency of the current breathing feature vector is determined, and the confidence of the user's respiratory abnormality leading to cardiopulmonary function degradation is calculated according to the variable frequency:

[0016] Where P represents the confidence of the user's respiratory abnormality leading to cardiopulmonary function degradation, ln represents the natural logarithm, represents the variable frequency of the current breathing feature vector, represents the reference variable frequency of the user's standard breathing feature vector, represents the respiratory frequency change factor caused by environmental influence, the value range is [0.3-0.5], and f() represents the influence coefficient evaluation function of the user's cardiopulmonary function fluctuation with the change of the motion state, represents the motion state intensity factor of the user when the abnormal breathing state occurs, represents the flat recovery time vector of the user after performing the relevant motion intensity, represents the motion after breathing recovery time vector evaluated based on the user's physical fitness, represents the dynamic adaptation factor of the user's physical fitness to different motion intensities; According to the confidence of the user's respiratory abnormality leading to cardiopulmonary function degradation, the user's cardiopulmonary function abnormality misjudgment situation is determined and a response warning is given.

[0017] Compared with the prior art, the present application has the following advantages: 1. In the present application, by detecting the time interval of the data stream, data packet loss can be discovered in time; and by using an adaptive error correction algorithm to repair the lost data, the data integrity is ensured; by combining a dynamic threshold adjustment mechanism to monitor the data transmission rate in real time, the data transmission error is reduced; and based on a unified clock source, the multi-channel data is time-domain aligned and fused to generate a multi-dimensional respiratory cycle image, and the data is analyzed in combination with physiological rules to eliminate untrusted data; thereby providing a reliable data basis for the assessment of cardiopulmonary function, which helps to improve the rehabilitation effect.

[0018] 2. In the present application, by constructing a cardiopulmonary function prediction model, the multi-dimensional respiratory cycle image of a chronic obstructive pulmonary disease patient is monitored and analyzed in real time, the abnormal pattern deviating from the baseline is accurately identified, and an acute exacerbation risk warning is generated; and a user behavior analysis mechanism is introduced to evaluate the psychological state and rehabilitation effect, and personalized warning information and rehabilitation suggestion are fed back to medical staff, so that the patient can be intervened in time when an abnormality occurs, and the monitoring burden of medical staff is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The flowchart of the present application is based on virtual reality for detecting the cardiopulmonary condition of a chronic obstructive pulmonary disease patient. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0021] In order to solve the technical problem that the existing detection methods are mostly static detection, such as lung function test, electrocardiogram, etc., but it is difficult to fully reflect the cardiopulmonary state of the patient in daily activities, and it is difficult to effectively improve the rehabilitation effect of the patient, please refer to Figure 1 The technical solutions provided in the present embodiment are as follows: The cardiopulmonary condition detection system for chronic obstructive pulmonary disease patients based on virtual reality comprises: The multi-dimensional data acquisition unit is configured to acquire the cardiopulmonary physiological data of the user in real time based on the intelligent respiratory sensor, the oximeter, the electromyography sensor and the thoracic impedance sensor, and transmit the data to the virtual reality interaction unit for prediction. The multi-dimensional data acquisition unit includes the wireless intelligent respiratory sensor for measuring the tidal volume, the respiratory rate, the inhalation-exhalation ratio and the peak expiratory flow rate of the user, the finger clip or wrist-worn oximeter for monitoring the blood oxygen saturation and the heart rate, the facial electromyography sensor of the VR head-mounted device and the wearable thoracic impedance sensor of the chest for monitoring the activity state of the respiratory auxiliary muscle group and the respiratory depth based on the virtual reality technology.

[0022] The data verification unit is configured to perform integrity verification on the acquired cardiopulmonary physiological data by using the time sequence continuity check and the transmission rate monitoring, to determine whether there is a packet loss in the cardiopulmonary physiological data, and to align and fuse the data to generate a multi-dimensional respiratory cycle image. For example, when recording the peak expiratory flow rate, the format is recorded as the blood oxygen level is X and the trapezius electromyography activity is Y when the PEF is reached, so as to accurately evaluate the cardiopulmonary state. The data verification unit includes: The packet loss detection module is configured to extract the time stamp sequence of each data stream in the cardiopulmonary physiological data, and calculate the time interval between the continuous data packets. If the interval is greater than twice the sampling period, it is determined that there is a data packet loss in the time period, and a warning prompt and error correction processing are performed. An adaptive error correction algorithm is introduced to automatically repair the lost data and ensure the integrity of the data.

[0023] The rate detection module is configured to obtain the acquisition frequency of the historical data and the corresponding normal transmission rate, and preset a threshold value based on the historical normal transmission rate. A dynamic threshold adjustment mechanism is introduced to automatically adjust the transmission rate threshold value according to the real-time network condition, thereby improving the robustness of the system. The real-time transmission rate of each channel in the multi-dimensional data acquisition unit is monitored in real time, and the real-time transmission rate of each channel is compared with the threshold value at regular intervals to determine whether there is an abnormality in the real-time transmission rate of each channel. If the rate of any channel continuously decreases below the preset threshold value, a data transmission abnormality warning is triggered. For example, if all sensor rates decrease synchronously during normal use of the user, it may indicate a network environment problem. If only a single channel rate is abnormal, it is highly suggested that the sensor itself has a fault, thereby improving the accuracy of fault positioning.

[0024] The beneficial effects achieved by the above content: by detecting the time interval of the data stream, the data packet loss phenomenon can be found and warned in time, and the lost data can be automatically repaired by means of the adaptive error correction algorithm, ensuring the integrity of the data; in combination with the dynamic threshold adjustment mechanism, the data transmission rate is monitored in real time, effectively avoiding data errors caused by abnormal transmission, and further enhancing the stability of the system; through a unified clock source, the multi-channel data is time-domain aligned, and the multi-dimensional data in the respiratory cycle is associated and fused to generate a multi-dimensional respiratory cycle image, providing a solid data foundation for accurate assessment of cardiopulmonary function.

[0025] The alignment and fusion module is configured to use a unified clock source as a reference to time-domain align the timestamp data of all channels; use the inspiration start event of the intelligent respiratory sensor as a reference point to slice and fuse all data in the respiratory cycle, associate the multi-dimensional representation of complete physiological events, and generate a multi-dimensional respiratory cycle image, providing a unique data basis for accurately assessing respiratory efficiency, muscle compensation degree, and oxygenation effect.

[0026] The logic verification module is configured to preset an analysis period based on the data acquisition frequency, such as every 30 seconds, perform sliding window analysis on the fused data according to the preset period, set logical judgment rules based on physiological rules, judge whether the cardiopulmonary physiological data is abnormal, further identify untrustworthy data caused by motion artifacts or device failure, and obtain the multi-dimensional respiratory cycle image that passes verification; for example, rule 1 is set as: if the blood oxygen saturation is > 70%, but the peak expiratory flow rate = 0, it is determined that there is a conflict in the data, and the data in this period is untrustworthy; rule 2 is set as: if the tidal volume surges while the thoracic impedance changes weakly, it is suggested that the respiratory sensor may be disturbed by non-respiratory airflow, such as the user speaking.

[0027] The beneficial effects achieved by the above content: based on physiological rules, the untrustworthy data caused by motion artifacts or device failure is effectively identified and removed, further improving the credibility of the data and the accuracy of the analysis results; thereby not only improving the overall performance of the system, but also providing a more scientific and reliable basis for the rehabilitation training of patients with chronic obstructive pulmonary disease, which helps to improve the rehabilitation effect and quality of life of patients.

[0028] The virtual reality interaction unit is configured to construct a three-dimensional digital lung model based on cardiopulmonary physiological data, construct an immersive virtual environment to which the three-dimensional digital lung model belongs based on a VR device, and convert the user's breathing behavior into an interaction instruction in the virtual environment to drive changes in the visual and auditory elements of the virtual environment, forming a biological feedback loop. Different visual attributes of the three-dimensional digital lung model can map different physiological states of the user in real time, for example: the color and fullness of the alveoli map the ventilation efficiency and blood oxygen level, the density and speed of the virtual particle flow in the airway map the real-time airflow speed, and the transparency or color warning of a specific area of the model maps the estimated airway resistance or inflammation level of the area; the virtual reality interaction unit comprises: The three-dimensional construction module is configured to initialize the three-dimensional digital lung model based on the preprocessed data, the model including alveoli, airways, and lung lobe structures, and set the initial state of each structure according to the user's physiological data; for example, the initial fullness of the alveoli is set according to the blood oxygen saturation and tidal volume, and the initial resistance of the airway is set according to the peak expiratory flow rate. Real-time reception of cardiopulmonary physiological data enables dynamic updating of the three-dimensional digital lung model; the inflation and contraction of the alveoli are adjusted according to the breathing rate and tidal volume, the density and speed of the virtual particle flow in the airway are adjusted according to the peak expiratory flow rate, and the color of the alveoli is adjusted according to the blood oxygen saturation; if abnormal data is detected, such as a sudden increase in airway resistance or a decrease in blood oxygen saturation, a warning is issued through the transparency or color of the corresponding area of the model.

[0029] The scene construction module is configured to create an immersive virtual environment using a VR device, such as a VR headset and an interaction device, and to preset multiple virtual application scenarios, including a rehabilitation game scenario, a mirror observation scenario, and a meditation guide scenario; for example, the rehabilitation game scenario includes virtual plant cultivation, virtual ship navigation, and other tasks, and the user cultivates virtual plants, pushes virtual ships to sail, or completes specific challenges by controlling the breathing rhythm and depth; in the mirror observation scenario, the user can enter the inside or around of the three-dimensional digital lung model, and intuitively observe the dynamic changes of the lungs under different breathing modes; in the meditation guide scenario, a relaxing environment and breathing guidance are provided, and when the system detects that the user is breathing rapidly or the blood oxygen level is decreasing, the user is automatically triggered and guided to enter the relaxation scenario, supplemented by breathing guidance. According to the initial physiological data and preference settings of the user, the virtual environment is initialized, and the visual, auditory, and tactile elements of the virtual environment are set to ensure that the user can naturally interact with the environment; real-time reception of the user's breathing behavior data converts the breathing behavior into an interaction instruction in the virtual environment; the visual and auditory elements in the virtual environment are dynamically adjusted according to the user's breathing rhythm and depth; for example, when the user's breathing is stable, the virtual plants grow vigorously, and the background music is soothing; when the user's breathing is rapid, the virtual plants grow slowly, and the background music is tense.

[0030] An instruction generation module is configured to convert the user's breathing behavior data, such as breathing rate, tidal volume, and inhalation-exhalation ratio, into interactive instructions in the virtual environment. For example, the user's inhalation action can trigger the irrigation of virtual plants, and the exhalation action can propel the virtual ship to sail. The interactive instructions are input through the handle of the VR device or gesture recognition, ensuring that the user can naturally interact with the virtual environment.

[0031] A data recording module is configured to record the user's long-term physiological data and performance in the virtual environment, conduct data analysis and trend prediction, and provide personalized rehabilitation training programs for the user based on long-term data. For example, if the system finds that the user's breathing is rapid more frequently in a certain period of time, it will automatically adjust the rehabilitation training program and increase the frequency of breathing relaxation training. At the same time, according to the user's progress, the difficulty and complexity of the rehabilitation training are gradually increased.

[0032] An environmental feedback module is configured to change the visual and auditory elements in the virtual environment according to the user's interactive instructions. For example, when the user successfully completes a deep breath, the virtual plants will bloom, and the background music will play a cheerful melody. At the same time, the changes in the virtual environment will be fed back to the user through visual and auditory feedback, forming a biofeedback loop. For example, the user can see the impact of their breathing behavior on the growth of virtual plants and hear background music that matches their breathing rhythm, thereby intuitively understanding their breathing status. The feedback intensity and content are dynamically adjusted according to the user's real-time physiological data and performance in the virtual environment. For example, if the user performs well in the rehabilitation game scene, the task difficulty is increased to provide more challenging virtual tasks. If the user shows anxiety in the mirror observation scene, a meditation guidance scene is automatically triggered to help the user relax.

[0033] A physiological feedback module is configured to adjust the visual, auditory, and tactile feedback in the virtual environment in real time by monitoring the user's cardiopulmonary physiological data and interactive behavior in the virtual environment. For example, when the user's breathing is rapid, the visual elements such as the color of the lung model in the virtual environment will issue a warning, and the auditory elements such as the background music will become more intense. The user is guided to adjust their breathing behavior through the biofeedback mechanism, and the state of the three-dimensional digital lung model and the virtual environment is dynamically updated according to the user's adjustment. For example, when the user sees that the virtual plants grow slowly due to rapid breathing, they will naturally adjust their breathing rhythm to restore the vitality of the plants. When the user's breathing gradually stabilizes, the visual and auditory elements in the virtual environment will gradually return to normal.

[0034] A cardiopulmonary warning unit is configured to construct a cardiopulmonary function prediction model based on a multi-dimensional breathing cycle portrait, identify abnormal patterns that deviate from the baseline, generate acute exacerbation risk warnings, and prompt the user through the virtual reality interaction unit in a preset visual metaphor manner. The cardiopulmonary warning unit includes: The model construction module is configured to collect historical data and known acute exacerbation cases, extract features related to cardiopulmonary function from them, such as respiratory cycle stability, blood oxygen saturation fluctuation range, respiratory frequency change rate, and respiratory auxiliary muscle group activity intensity, train the features based on a neural network model, and construct a prediction model that can identify abnormal patterns.

[0035] The baseline construction module is configured to obtain multi-dimensional respiratory cycle portraits over a period of time, establish a user's cardiopulmonary function baseline based on multi-dimensional respiratory cycle portraits in a normal state, and baseline data includes normal respiratory cycle characteristic parameters such as average respiratory frequency, average blood oxygen saturation, and normal respiratory auxiliary muscle group activity range. Abnormal patterns are classified based on multi-dimensional respiratory cycle portraits in an abnormal state, including mild, moderate, and severe abnormalities. Multi-modal data fusion technology is introduced to combine cardiopulmonary physiological data and user behavior data to analyze whether the user's behavior data in the virtual environment is compatible with the cardiopulmonary physiological data. For example, by analyzing the synchronization of action frequency and breathing behavior of the user in the virtual environment, potential breathing difficulties are identified. A user behavior analysis mechanism is introduced to analyze the user's behavior patterns and reactions in the virtual environment. For example, by analyzing the user's reaction speed and task completion in the rehabilitation game scene, the user's psychological state and rehabilitation effect are evaluated. According to the user's behavior analysis results, personalized early warning information and rehabilitation recommendations are generated and fed back to medical staff in real time. For example, for users who show anxiety in the rehabilitation game, the frequency of meditation guidance scenes is increased to help users relax. Automatic alerts are sent to the attending physician, providing real-time physiological data and warning information for the user. Real-time multi-dimensional respiratory cycle portraits are received, and real-time cardiopulmonary physiological data is compared with baseline data to identify whether there is an abnormal pattern that deviates from the baseline. For example, if the real-time respiratory frequency is more than 1.5 times the baseline value and the blood oxygen saturation is less than 90% of the baseline value, it is identified as an abnormal pattern. The identified abnormal pattern is analyzed to determine its possible causes and risk level. For example, if the respiratory frequency increases significantly and the blood oxygen saturation decreases, it may indicate an acute exacerbation risk. According to the analysis results of the abnormal pattern, an acute exacerbation risk warning signal is generated, including warning levels such as low risk, medium risk, and high risk, as well as corresponding recommended measures. For example, for high-risk warnings, the user is advised to immediately stop the current activity, perform respiratory relaxation training, and contact medical personnel.

[0036] The above-mentioned beneficial effects are achieved: by constructing a heart-lung function prediction model, real-time monitoring and analyzing the multi-dimensional respiratory cycle image of patients with chronic obstructive pulmonary disease, accurately identifying abnormal patterns deviating from the baseline, generating acute exacerbation risk warnings; and introducing a user behavior analysis mechanism to evaluate the psychological state and rehabilitation effect, generating personalized warning information and rehabilitation suggestions to feed back to medical staff, ensuring that patients can receive timely intervention when abnormalities occur, while reducing the monitoring burden of medical staff.

[0037] Working principle: by collecting the user's cardiopulmonary physiological data, verifying data packet loss and transmission anomalies, and using adaptive error correction algorithms and dynamic threshold adjustment mechanisms to repair and adjust the data; time-domain alignment and fusion of multi-channel data to generate a multi-dimensional respiratory cycle image, combined with logical verification to further analyze the accuracy of the data; based on the multi-dimensional respiratory cycle image, a prediction model is constructed to identify abnormal patterns and generate acute exacerbation risk warnings, and through virtual reality interaction technology to build an immersive virtual environment, convert respiratory behavior into interactive instructions to drive changes in visual and auditory elements of the virtual environment, forming a biological feedback loop; and dynamically adjusting the feedback content to guide users to adjust their breathing behavior, improving rehabilitation effectiveness and quality of life.

[0038] In one embodiment, after constructing the three-dimensional digital lung model, it further includes: Real-time acquisition of user's cardiopulmonary physiological data, including airway pressure, respiratory flow, tidal volume and blood oxygen saturation; Integrating a physics engine, the physics engine is pre-defined with a biomechanical model of virtual lung tissue, the biomechanical model includes elastic resistance parameters for simulating alveolar and airway elasticity, viscous resistance parameters for simulating airflow friction resistance, and inertial mass parameters for simulating overall lung inertia; Inputting the real-time acquired cardiopulmonary physiological data as boundary conditions into the physics engine, by solving the dynamic equation of the biomechanical model, real-time driving the three-dimensional digital lung model to produce deformation and dynamic effects consistent with the user's real physiological process; Based on real-time airway pressure and respiratory flow data, the physics engine calculates the fluid dynamics effect of virtual particle flow in the virtual airway; According to the calculated flow rate and pressure difference, the flow state of the virtual particle flow is adjusted in real time, including laminar flow, turbulent flow or vortex flow state, and is visually presented in the airway structure of the three-dimensional digital lung model; When the calculated flow state value exceeds the preset turbulent flow threshold, turbulent flow or vortex flow visual effects are generated and rendered in the corresponding airway area of the three-dimensional digital lung model; Based on tidal volume and airway pressure data, the physics engine calculates the volume change rate and stress distribution of the alveolar unit in the three-dimensional digital lung model in real time; According to the volume change rate and stress distribution, the three-dimensional geometry of each alveolar unit is dynamically updated to simulate the inflation and deflation process of the alveoli; When the real-time cardiopulmonary physiological data indicates the presence of an abnormal state, according to the type and severity of the abnormality, the physical parameters of the local region in the biomechanical model are modified by the physics engine and the three-dimensional digital lung is updated in stages.

[0039] In this embodiment, real-time refers to the time taken by the system to process and output a frame of data, which is less than or equal to twice the physiological data sampling period; In this embodiment, the physics engine refers to a computational module dedicated to biomechanical simulation, the core of which is the integration of a biomechanical model tailored to the characteristics of lung tissue, capable of solving dynamic equations involving elastic resistance, viscous resistance, and inertial mass. It can be implemented based on the finite element method, particle spring system, or other physical modeling methods specific to soft tissue simulation; In this embodiment, the elastic resistance parameter is used to represent the elastic recoil force of the lung parenchyma and alveolar wall, which is represented in the model as the relationship between stress and strain, and can be defined by a nonlinear elastic modulus. The viscous resistance parameter is used to represent the frictional resistance encountered by airflow in the airway, which is represented in the model as the relationship between pressure and flow, and can be defined by the resistance value of the airway or the viscosity of the fluid. The inertial mass parameter is used to represent the force required to accelerate or decelerate the gas and lung tissue, which is represented in the model as the relationship between pressure and flow rate, and can be defined by the inertial value; In this embodiment, virtual particle flow refers to a fluid visualization method based on the Lagrangian perspective. The 'virtual particle' is a macroscopic tracer point, whose trajectory is driven by the physics engine according to the calculated flow field (such as the velocity vector field), which is used to intuitively represent the flow path and state of the airflow; In this embodiment, the flow state value is represented as the local Reynolds number. The preset turbulence threshold is represented as the critical Reynolds number; In this embodiment, the alveolar unit refers to the smallest structural unit in the three-dimensional digital lung model for mechanical calculation and morphological update. It can represent a single alveolus or a pulmonary lobule containing multiple alveoli according to the model accuracy. Each alveolar unit is modeled as a physical entity with volume, elasticity, and mass in the physics engine; In this embodiment, an abnormal state refers to a state determined by comparing real-time cardiopulmonary physiological data with pre-stored normal value ranges (e.g., adult tidal volume 6-8 mL / kg, blood oxygen saturation >95%) or by identifying specific pathological patterns (e.g., airway obstruction, decreased lung compliance) through pattern recognition algorithms; In this embodiment, the phased update refers to the physics engine adjusting in two phases after detecting an abnormal state: In the first phase, the physical parameters of the corresponding region in the biomechanical model are modified instantaneously according to the type of abnormality [for example, the viscous resistance parameter R is increased by 300% in the airway obstruction region]; In the second phase, the geometry of the three-dimensional digital lung model is gradually adjusted over several respiratory cycles after the parameter modification to reflect the steady state under the pathological condition [for example, simulating lung overinflation].

[0040] The beneficial effects of the above technical solution are as follows: by updating the user's three-dimensional digital lung model in real time, the user's breathing status can be combined to simulate the real-time changes in the lung status, thereby intuitively determining the user's cardiopulmonary function impairment and changes, making the cardiopulmonary assessment of the user more accurate and reliable.

[0041] Furthermore, monitor the user's abnormal breathing state, obtain the current respiratory feature vector under the abnormal breathing state, and determine the user's standard respiratory feature vector under the normal breathing state based on the cardiopulmonary function curve; Determine the frequency of variables in the current respiratory feature vector, and calculate the confidence level that the user's respiratory abnormalities lead to cardiopulmonary function deterioration based on the variable frequencies: Where P represents the confidence level that the user's abnormal breathing leads to cardiopulmonary function deterioration, and ln represents the natural logarithm. The variable frequency is represented as the current respiratory feature vector. The frequency of the reference variable is represented as the standard respiratory feature vector of the user. The factor representing the change in respiratory rate due to environmental influences is denoted as [0.3-0.5], and f() represents the evaluation function of the influence coefficient of the user's cardiopulmonary function fluctuations with changes in exercise state. This represents the intensity factor of the user's motion state at the time of the abnormal breathing state. This represents the recovery time vector after the user performs the relevant intensity of exercise. Represented as a vector of post-exercise respiratory recovery time based on user physical condition assessment. This represents the user's physical condition as a dynamic adaptation factor to different exercise intensities; Based on the confidence level that a user's abnormal breathing leads to a decline in cardiopulmonary function, we can determine the misjudgment of a user's cardiopulmonary function abnormality and issue a response warning.

[0042] In this embodiment, the respiratory feature vector is an n-dimensional vector whose elements consist of physiological parameters extracted during a complete respiratory cycle, including but not limited to: respiratory rate, tidal volume, inspiratory time, expiratory time, peak inspiratory flow rate, and peak expiratory flow rate. In the embodiment, the variable frequency is represented as the rate at which the respiratory feature vector changes significantly over a plurality of consecutive breath cycles; In the embodiment, the recovery time vector is represented as a multi-dimensional average time metric, including the time required for the respiratory rate to recover to pre-exercise levels, the time required for the tidal volume to recover to pre-exercise levels, and the time required for the blood oxygen saturation to recover to baseline levels; In the embodiment, the post-exercise respiratory recovery time vector is a theoretical recovery time vector calculated based on the user's age, BMI index, and baseline lung function (e.g. FEV1 as a percentage of predicted value) using a pre-defined regression model; In the embodiment, the exercise state intensity factor is determined by mapping the user's motion amplitude and frequency in the exercise environment, and referring to the user's real-time heart rate data, to a pre-set intensity level (e.g. low = 0.8, medium = 1.0, high = 1.5). The dynamic adaptation factor is a parameter that is self-adaptively adjusted based on the user's recent rehabilitation training data. If the user has recently performed well and improved recovery ability, the value of the factor will be increased accordingly, for example from 1.0 to 1.1.

[0043] The beneficial effects of the above technical solutions are that determining the confidence level by investigating the causes of the user's respiratory abnormalities can avoid misidentification and accurately determine the user's respiratory abnormality, thereby improving the accuracy and stability.

[0044] It should be noted that, in this document, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0045] Although embodiments of the present application have been shown and described, it would be appreciated by those skilled in the art that changes can be made in these embodiments without departing from the principles and spirit of the application, the scope of which is defined by the following claims and their equivalents.

Claims

1. A virtual reality based system for detecting cardiopulmonary status of a patient with chronic obstructive pulmonary disease, characterized in that, The application comprises: a multi-dimensional data acquisition unit configured to acquire real-time cardiopulmonary physiological data of a user based on an intelligent respiratory sensor, an oximeter, an electromyographic sensor, and a thoracic impedance sensor, and transmit the data to a virtual reality interaction unit for prediction; a data verification unit configured to verify the integrity of the acquired cardiopulmonary physiological data using time sequence continuity inspection and transmission rate monitoring to determine whether the data has packet loss; align and fuse the data to generate a multi-dimensional respiratory cycle image; a virtual reality interaction unit configured to construct a three-dimensional digital lung model based on the cardiopulmonary physiological data, construct an immersive virtual environment to which the three-dimensional digital lung model belongs based on a VR device, and convert the user's respiratory behavior into an interaction instruction in the virtual environment to drive changes in the visual and auditory elements of the virtual environment, forming a biological feedback loop; a cardiopulmonary warning unit configured to construct a cardiopulmonary function prediction model based on the multi-dimensional respiratory cycle image, identify abnormal patterns that deviate from the baseline, monitor the respiratory feature vector in the abnormal respiratory state of the user, determine the standard feature vector in the normal respiratory state in combination with the cardiopulmonary function curve, calculate the confidence level of cardiopulmonary function degradation caused by respiratory abnormalities, determine cardiopulmonary function abnormality misjudgment, and provide a warning, and prompt the user through the virtual reality interaction unit in a preset visual metaphor.

2. The virtual reality based chronic obstructive pulmonary disease patient cardiopulmonary condition detection system as claimed in claim 1, wherein, The data verification unit comprises: a packet loss inspection module configured to extract the timestamp sequence of each data stream in the cardiopulmonary physiological data, calculate the time interval between consecutive data packets, determine that there is data loss in the time period if the interval is greater than twice the sampling period, and provide a warning prompt and error correction processing, and introduce an adaptive error correction algorithm to automatically repair the lost data packets; a rate inspection module configured to obtain the acquisition frequency of historical data and the corresponding normal transmission rate, preset a threshold based on the historical normal transmission rate, and introduce a dynamic threshold adjustment mechanism to automatically adjust the transmission rate threshold based on real-time network conditions; real-time monitor the real-time transmission rate of each channel in the multi-dimensional data acquisition unit, periodically compare the real-time transmission rate of each channel with the threshold, and determine whether the real-time transmission rate of each channel is abnormal; if the rate of any channel is continuously lower than the preset threshold, a data transmission anomaly warning is triggered.

3. The virtual reality based chronic obstructive pulmonary disease patient cardiopulmonary condition detection system as claimed in claim 2, wherein, The data verification unit further comprises: an alignment and fusion module configured to use a unified clock source as a reference to align the timestamp data of all channels in the time domain, use the inspiration start event of the intelligent respiratory sensor as a reference point to slice and fuse all data within the respiratory cycle, correlate the multi-dimensional representation of complete physiological events, and generate a multi-dimensional respiratory cycle image; a logic verification module configured to preset an analysis period based on the data acquisition frequency, perform sliding window analysis on the fused data in the preset period, set logic judgment rules based on physiological rules, determine whether the cardiopulmonary physiological data is abnormal, further identify untrustworthy data caused by motion artifacts or device failure, and obtain a multi-dimensional respiratory cycle image that passes verification.

4. The virtual reality based chronic obstructive pulmonary disease patient cardiopulmonary condition detection system as claimed in claim 1, wherein, The cardiopulmonary warning unit comprises: The model construction module is configured to collect historical data and known acute exacerbation cases, extract features related to cardiopulmonary function therefrom, train the features based on a neural network model, and construct a prediction model capable of identifying abnormal patterns; The baseline construction module is configured to obtain multi-dimensional respiratory cycle portraits within a period of time, establish a cardiopulmonary function baseline of the user based on the multi-dimensional respiratory cycle portraits in a normal state, and the baseline data includes characteristic parameters of normal respiratory cycles; The abnormal patterns are classified based on the multi-dimensional respiratory cycle portraits in an abnormal state, including mild abnormality, moderate abnormality and severe abnormality; Real-time multi-dimensional respiratory cycle portraits of the user are received, and real-time cardiopulmonary physiological data are compared with baseline data to identify whether there is an abnormal pattern deviating from the baseline; The identified abnormal pattern is analyzed to determine its possible cause and risk level; and an acute exacerbation risk warning signal is generated based on the analysis result of the abnormal pattern, the warning signal including a warning level and corresponding recommended measures.

5. The virtual reality based chronic obstructive pulmonary disease patient cardiopulmonary condition detection system as claimed in claim 4, wherein, The abnormal patterns are classified based on the multi-dimensional respiratory cycle portraits in an abnormal state, and further include: A multi-modal data fusion technology is introduced to combine cardiopulmonary physiological data and user behavior data to analyze whether the behavior data of the user in the virtual environment is adapted to the cardiopulmonary physiological data; A user behavior analysis mechanism is introduced to analyze the behavior pattern and reaction of the user in the virtual environment; Based on the analysis result of the user behavior, personalized warning information and rehabilitation recommendations are generated and fed back to medical staff in real time.

6. The virtual reality based chronic obstructive pulmonary disease patient cardiopulmonary condition detection system as claimed in claim 1, wherein, The virtual reality interaction unit includes: A three-dimensional construction module configured to initialize a three-dimensional digital lung model based on preprocessed data, the model including alveoli, airways and lung lobe structures, and to set the initial state of each structure according to the physiological data of the user; Real-time cardiopulmonary physiological data are received to dynamically update the three-dimensional digital lung model; the inflation and contraction of alveoli are adjusted according to the respiratory rate and tidal volume, the density and speed of virtual particle flow in the airway are adjusted according to the peak expiratory flow rate, and the color of alveoli is adjusted according to the blood oxygen saturation; if abnormal data are detected, a warning is issued through the transparency or color of the corresponding area of the model; A scene construction module configured to create an immersive virtual environment using a VR device, preset a plurality of virtual application scenarios including a rehabilitation game scenario, a mirror observation scenario and a meditation guidance scenario; initialize the virtual environment according to the initial physiological data and preference settings of the user, and set the visual, auditory and tactile elements of the virtual environment to ensure that the user can naturally interact with the environment; Real-time respiratory behavior data of the user are received to convert the respiratory behavior into interaction instructions in the virtual environment; the visual and auditory elements in the virtual environment are dynamically adjusted according to the respiratory rhythm and depth of the user.

7. The virtual reality based chronic obstructive pulmonary disease patient cardiopulmonary condition detection system as claimed in claim 6, wherein, After the three-dimensional digital lung model is constructed, further including: Real-time cardiopulmonary physiological data of the user are obtained, including airway pressure, respiratory flow, tidal volume and blood oxygen saturation; integrate a physical engine, which is predefined with a biomechanical model of virtual lung tissue, the biomechanical model including an elastic resistance parameter for simulating alveolar and airway elasticity, a viscous resistance parameter for simulating airflow friction resistance, and an inertial mass parameter for simulating lung overall inertia; input real-time acquired cardiopulmonary physiological data as boundary conditions into the physical engine, and drive the three-dimensional digital lung model to produce deformation and dynamic effects consistent with the user's real physiological process by solving the dynamic equations of the biomechanical model; based on real-time airway pressure and respiratory flow data, calculate the fluid dynamics effect of virtual particle flow in the virtual airway through the physical engine; according to the calculated flow rate and pressure difference, real-time adjust the flow state of the virtual particle flow, including laminar flow, turbulent flow or vortex flow state, and visually present in the airway structure of the three-dimensional digital lung model; when the calculated flow state value exceeds the preset turbulent flow threshold, generate and render turbulent flow or vortex flow visual effects in the corresponding airway region of the three-dimensional digital lung model; based on the tidal volume and airway pressure data, calculate the volume change rate and stress distribution of the alveolar unit in the three-dimensional digital lung model through the physical engine in real time; according to the volume change rate and stress distribution, dynamically update the three-dimensional geometry of each alveolar unit to simulate the inflation and deflation process of the alveoli; when the real-time cardiopulmonary physiological data indicates the presence of abnormal state, modify the physical parameters of the local region in the biomechanical model and perform stage update on the three-dimensional digital lung according to the type and severity of the abnormality through the physical engine.

8. The virtual reality based chronic obstructive pulmonary disease patient cardiopulmonary condition detection system as claimed in claim 6, wherein, The virtual reality interaction unit further comprises: an instruction generation module configured to convert the user's breathing behavior data into interaction instructions in the virtual environment, and input the interaction instructions through the handle or gesture recognition of the VR device; a data recording module configured to record the user's long-term physiological data and performance in the virtual environment, perform data analysis and trend prediction, and provide personalized rehabilitation training programs for the user according to the long-term data.

9. The virtual reality based chronic obstructive pulmonary disease patient cardiopulmonary condition detection system as claimed in claim 8, wherein, The virtual reality interaction unit further comprises: an environmental feedback module configured to drive changes in visual and auditory elements in the virtual environment according to the user's interaction instructions, and dynamically adjust the feedback intensity and content according to the user's real-time physiological data and performance in the virtual environment; a physiological feedback module configured to adjust the visual, auditory and tactile feedback in the virtual environment in real time by monitoring the user's cardiopulmonary physiological data and interaction behavior in the virtual environment, and guide the user to adjust the breathing behavior through the biological feedback mechanism, and dynamically update the state of the three-dimensional digital lung model and the virtual environment according to the user's adjustment.

10. The virtual reality based chronic obstructive pulmonary disease patient cardiopulmonary condition detection system as claimed in claim 4, wherein, monitor the user's abnormal breathing state, acquire the current breathing feature vector under the abnormal breathing state, and determine the standard breathing feature vector of the user under the normal breathing state according to the cardiopulmonary function curve; determine the variable frequency of the current breathing feature vector, and calculate the confidence degree of the user's respiratory abnormality leading to cardiopulmonary function degradation according to the variable frequency: wherein P represents the confidence degree of the abnormal breathing of the user leading to the degradation of the cardiopulmonary function, ln represents the natural logarithm, a variable frequency representing the current breathing feature vector, a reference variable frequency representing the standard breathing feature vector of the user, a breathing frequency change factor caused by the environmental influence, the value range of which is [0.3-0.5], and f() represents an influence coefficient evaluation function of the fluctuation of the cardiopulmonary function of the user with the change of the motion state, a motion state intensity factor representing the motion state of the user when the abnormal breathing state occurs, a recovery time vector representing the relevant motion intensity after the user performs the relevant motion, a recovery time vector representing the recovery time of the user after the motion based on the physical constitution of the user, a dynamic adaptation factor representing the physical constitution of the user to different motion intensities; determine the user's cardiopulmonary function abnormality misjudgment situation and respond to the warning according to the confidence degree of the user's respiratory abnormality leading to cardiopulmonary function degradation.

Images