A wearable system for assessing risk of postoperative pulmonary complications

Abstract

The application discloses a wearable system for evaluating postoperative pulmonary complication risk, comprising: a data acquisition module, including a vest, a pulse oximeter and a signal recording box, for collecting physiological signals of an evaluated person during submaximal exercise test; a data processing module connected with the data acquisition module, for processing the physiological signals to obtain characteristic data based on the peak and slope of the physiological signals; a model calculation module connected with the data processing module, for inputting the characteristic data into a model trained by using a stochastic gradient algorithm and including preoperative clinical physiological characteristic data, to calculate a postoperative pulmonary complication incidence rate; and a model evaluation module connected with the model calculation module, for determining whether the evaluated person has postoperative pulmonary complications according to a Melbourne Group Score Scale. The system can accurately predict the probability of pulmonary complications by the cooperation of the modules, and provide precise medical intervention guidance for clinical treatment.

Description

Technical Field

[0001] This invention relates to the field of predicting pulmonary complications. Specifically, it relates to a wearable system for assessing the risk of postoperative pulmonary complications. Background Technology

[0002] Valvular heart disease is a significant component of cardiovascular disease in my country. According to statistics from the Extracorporeal Circulation Group of the Chinese Society for Biomedical Engineering, approximately 80,000 individuals undergo heart valve surgery assessment annually in my country, accounting for one-third of all cardiac and vascular surgeries. The respiratory and cardiovascular systems are spatially adjacent and their functions are highly coordinated; therefore, all patients assessed after heart surgery experience varying degrees of respiratory dysfunction. Approximately 10% to 25% of these patients develop postoperative pulmonary complications, such as pulmonary infection, pleural effusion, and ventilatory failure. Postoperative pulmonary complications can significantly increase postoperative mortality and hospitalization costs for those assessed for heart valve surgery, leading to prolonged stays in intensive care units and extended hospital stays.

[0003] Preoperative risk assessment of postoperative pulmonary complications in patients undergoing heart valve surgery plays a crucial role in clinical treatment planning, prognosis, complication prediction, treatment efficacy evaluation, and estimation of medical resource needs. Pulmonary function tests and cardiopulmonary exercise tests are standard tests for predicting postoperative pulmonary complication risk. However, these tests have limitations, such as expensive equipment, contraindications and adverse safety events, and the potential for inaccurate results due to a lack of professional knowledge and standardized procedures.

[0004] The American Thoracic Society guidelines recommend continuous monitoring of physiological parameters, such as oxygen saturation, to determine the occurrence of hypoxemia. However, in most clinical settings, oxygen saturation is measured intermittently, which may miss early signs of deterioration and lead to adverse events. In recent years, an increasing number of studies have focused on acquiring continuous respiratory physiological data during trials. Various monitoring devices have been developed that can continuously, objectively, and conveniently monitor the respiratory physiological data of subjects during the perioperative period. In addition to focusing on basic indicators such as respiratory rate, tidal volume, resting oxygen saturation, and maximum oxygen saturation, researchers are also interested in other indicators derived from these respiratory physiological parameters. Continuous physiological data monitoring has been used to predict mortality in subjects with chronic obstructive pulmonary disease (COPD) due to exercise-induced hypoxemia events; and to predict asthma severity by monitoring respiratory rate variability.

[0005] Currently, there is no device for continuous preoperative respiratory physiological monitoring in patients undergoing heart valve surgery to predict postoperative pulmonary complications. Therefore, this device is innovatively designed based on data-driven and dynamic monitoring methods to predict postoperative pulmonary complications in patients undergoing heart valve surgery through continuous respiratory physiological monitoring via wearable devices, filling a gap in this field. Summary of the Invention

[0006] The present invention is proposed based on the above-mentioned needs of the prior art. The technical problem to be solved by the present invention is that there is no device in the prior art for preoperative continuous respiratory physiological monitoring of patients undergoing heart valve surgery to predict postoperative pulmonary complications.

[0007] To solve the above problems, the present invention is implemented using the following technical solution:

[0008] A wearable system for assessing the risk of postoperative pulmonary complications, comprising:

[0009] The data acquisition module includes a vest, a pulse oximeter, and a signal recording box, used to collect physiological signals of the subject during a submaximal exercise experiment. The vest is connected to the signal recording box to collect respiratory signals, including thoracic and abdominal respiratory signals. The pulse oximeter is connected to the signal recording box to collect blood oxygenation signals. The signal recording box is equipped with an electrocardiogram (ECG) sensor to collect ECG signals. The submaximal exercise experiment includes a first stage of performing the submaximal exercise experiment and a second stage after performing the submaximal exercise experiment.

[0010] A data processing module, connected to the data acquisition module, is used to process the physiological signal based on the peak value and slope of the physiological signal to obtain feature data, wherein the feature data includes the maximum respiratory rate in the first stage, the average abdominal breathing contribution ratio in the first stage, and the average abdominal breathing contribution ratio in the second stage.

[0011] The model calculation module, connected to the data processing module, is used to input the feature data into the model obtained by training the feature data using the stochastic gradient algorithm, and calculate the incidence of postoperative pulmonary complications.

[0012] The model evaluation module, connected to the model calculation module, determines whether the evaluated subject has experienced postoperative pulmonary complications based on the Melbourne Group Rating Scale, assesses the model accuracy, and optimizes the model parameters.

[0013] Optionally, the system further includes:

[0014] The experimental assessment module acquires the subject's Borg dyspnea score, PRE score for exertion, and New York heart function classification to determine whether the target subject can perform submaximal exercise experiments.

[0015] Optionally, the step of processing the physiological signal based on its peak value and slope to obtain feature data includes:

[0016] The chest breathing signal and the abdominal breathing signal are superimposed, and the peak value and slope of the superimposed signal are used to determine the respiratory cycle and corresponding respiratory time period of the subject in each stage of the submaximal exercise experiment.

[0017] Determine the maximum respiratory rate based on the respiratory cycle of the first stage;

[0018] Based on the respiratory time period corresponding to the respiratory cycle in the first stage, the abdominal respiratory amplitude is integrated with the sum of the chest respiratory amplitude and the abdominal respiratory amplitude at the corresponding time to obtain the average abdominal respiratory contribution ratio in the first stage and the average abdominal respiratory contribution ratio in the second stage. The respiratory amplitude is the impedance amplitude obtained by processing the respiratory signal through electrical impedance tomography.

[0019] Optionally, after superimposing the chest breathing signal and the abdominal breathing signal, the process includes: using a bandpass filter to filter the signal to be processed to remove DC and heart-related impedance change data.

[0020] Optionally, after superimposing the chest breathing signal and the abdominal breathing signal, the process includes: preprocessing the signal to be processed using a digital high-pass filter to remove data below a preset frequency threshold.

[0021] Optionally, determining the respiratory cycle and corresponding respiratory time period of the subject at each stage of the submaximal exercise experiment using the peak value and slope of the superimposed signal includes:

[0022] The superimposed signal is divided into a waiting rising crossover state and a waiting falling crossover state;

[0023] Using the following formula, while waiting for the rising crossover state, the current minimum value of the superimposed signal is continuously updated to obtain the exhalation endpoint; while waiting for the falling crossover state, the current maximum value of the superimposed signal is continuously updated using the following formula to obtain the inhalation endpoint.

[0024] MICS = MICSfact-60 / maxBR

[0025] MDCS = MDCSfact - 60 / maxBR

[0026] Where MICSfact and MDCSfact are normalized parameters, maxBR is the assumed maximum respiratory rate, MICS represents the minimum identical cross interval, and MDCS represents the minimum different cross interval.

[0027] The respiratory cycle and the corresponding respiratory time period are determined based on adjacent respiratory endpoints and inspiratory endpoints.

[0028] Optionally, the system further includes:

[0029] The data display module is connected to the data processing module and is used to display the data received and processed by the data processing module in real time.

[0030] Optionally, the submaximal exercise experiment also includes a preparation phase before conducting the submaximal exercise experiment.

[0031] Optionally, the system further includes:

[0032] The user management module, connected to the model evaluation module, is used to store user information, identify the person being evaluated, perform the submaximal exercise experiment, and store the corresponding experimental data.

[0033] Optionally, the system further includes:

[0034] An alarm device, connected to the data acquisition module, is used to determine whether the electrocardiogram signal exceeds a preset threshold. If it exceeds the preset threshold, the alarm device issues a prompt signal, wherein the prompt signal includes at least one of the following or a combination thereof: sound, text, image, and light.

[0035] Optionally, training a model using preoperative clinical and physiological characteristic data including the said feature data using a stochastic gradient descent algorithm includes:

[0036] Using preoperative clinical physiological characteristics as variables, and the probability of postoperative pulmonary complications as a function, we predicted the probability of postoperative pulmonary complications.

[0037] Optionally, the prediction function of the model obtained by training the preoperative clinical physiological feature data including the feature data using the stochastic gradient algorithm is:

[0038] h θ (x)=θ0x0+θ1x1+…+θ j x j

[0039] Where θ=[θ0θ1…θ j ],x=]x0x1…x j ], h θ(x) represents the complication probability of the array x formed by the input feature data of the evaluated person, θ is the vector formed by the feature coefficients, θ0 is a constant, and θ j Let x be the feature coefficient corresponding to the j-th feature data, and x be the vector formed by the feature data, x0 = 1, x j Let j be the feature data of the jth term.

[0040] Optionally, the loss function based on the prediction function is expressed as:

[0041]

[0042] Where y represents the actual label.

[0043] Optionally, the cost function obtained based on the loss function is expressed as:

[0044]

[0045] Where m is the number of consecutive preoperative clinical physiological parameter arrays in the dataset, and x (i) Let y be the i-th continuous preoperative clinical physiological parameter array. (i) For x (i) The corresponding label has a value of 0 or 1 for y.

[0046] Optionally, the feature coefficients are initialized using gradient descent and updated incrementally until the optimal feature coefficients θ for the feature data in the continuous preoperative clinical physiological parameter array are obtained. j , is represented as:

[0047]

[0048] Where, θ (t) θ represents the parameter in the current t-th iteration. (t+1) Represents the model coefficients for iteration t+1.

[0049] Compared to existing technologies, the system of this invention, through the collaborative interaction of its various modules, can predict the probability of pulmonary complications. Utilizing wearable devices to collect continuous respiratory physiological data during exercise to predict postoperative pulmonary complications in perioperative patients shows great promise and overcomes the limitations of traditional pulmonary function testing, which is hampered by the subjective cooperation of the assessee, and cardiopulmonary exercise testing, which is difficult to promote due to expensive equipment and complex operation. Furthermore, through theoretical analysis, empirical summarization, and model calculation, the system determines the correlation between characteristic data and the probability of postoperative pulmonary complications during submaximal exercise experiments. This allows for highly accurate probabilistic results in submaximal exercise experiments, facilitating accurate assessments for patients and providing guidance for surgical treatment planning, thereby improving surgical outcomes and mitigating risks. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings.

[0051] Figure 1 This is a schematic diagram of the framework of a wearable system for assessing the risk of postoperative pulmonary complications, provided by a specific embodiment of the present invention.

[0052] Figure 2 This is a schematic diagram of the vest structure of a wearable system for assessing the risk of postoperative pulmonary complications provided by a specific embodiment of the present invention;

[0053] Figure 3 This is a schematic diagram of the signal recording box structure of a wearable system for assessing the risk of postoperative pulmonary complications, provided in a specific embodiment of the present invention.

[0054] Figure 4 This is a schematic diagram of the signal acquisition of a wearable system for assessing the risk of postoperative pulmonary complications provided in a specific embodiment of the present invention;

[0055] Figure label:

[0056] 1-Equipment interface; 2-Chest breathing belt; 3-Abdominal breathing belt; 4-Power button; 5-Blue indicator light; 6-Green indicator light. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] To facilitate understanding of the embodiments of the present invention, further explanations and descriptions will be provided below with reference to the accompanying drawings and specific embodiments. These embodiments do not constitute a limitation on the scope of protection of the present invention.

[0059] This embodiment provides a wearable system for assessing the risk of postoperative pulmonary complications, such as... Figure 1 As shown, it includes:

[0060] The data acquisition module includes a vest, a pulse oximeter, and a signal recording box, used to collect physiological signals of the subject during a submaximal exercise experiment. The vest is connected to the signal recording box to collect respiratory signals, including thoracic and abdominal respiratory signals. The pulse oximeter is connected to the signal recording box to collect blood oxygenation signals. The signal recording box is equipped with an electrocardiogram (ECG) sensor to collect ECG signals. The submaximal exercise experiment includes a first stage of performing the submaximal exercise experiment and a second stage after performing the submaximal exercise experiment.

[0061] A data processing module, connected to the data acquisition module, is used to process the physiological signal based on the peak value and slope of the physiological signal to obtain feature data, wherein the feature data includes the maximum respiratory rate in the first stage, the average abdominal breathing contribution ratio in the first stage, and the average abdominal breathing contribution ratio in the second stage.

[0062] The model calculation module, connected to the data processing module, is used to input the feature data into a model obtained by training a preoperative clinical and physiological feature array including the feature data using a stochastic gradient algorithm, and to calculate the incidence of postoperative pulmonary complications.

[0063] The model evaluation module, connected to the model calculation module, determines whether the evaluated subject has experienced postoperative pulmonary complications based on the Melbourne Group Rating Scale, assesses the model accuracy, and optimizes the model parameters.

[0064] In this embodiment of the invention, the selected preoperative clinical physiological characteristic data include surgical method, age, cardiac function classification standard, European cardiovascular surgery risk factor score, left ventricular diameter, left ventricular enlargement, left atrial diameter, left atrial enlargement, right ventricular diameter, right ventricular enlargement, right atrial diameter, right atrial enlargement, left ventricular end-diastolic volume, left ventricular end-diastolic structural change parameters, left ventricular end-systolic volume, left ventricular end-systolic structural change parameters, left ventricular ejection fraction, left ventricular systolic function change parameters, pulmonary artery diameter, tidal volume characteristic data, ventilation characteristic data, respiratory rate characteristic data, and inspiratory and expiratory time characteristic data.

[0065] Preferably, respiratory signals, blood oxygen signals, and electrocardiogram signals are selected as parameters that are strongly correlated with the risk of postoperative pulmonary complications in the characteristic data of the assessed person. By selecting the above data, the amount of calculation required to predict postoperative pulmonary complications can be significantly reduced, the prediction speed can be improved, and the accuracy of the prediction can be guaranteed.

[0066] The contribution ratio of abdominal breathing can indicate the lung function of the person being assessed. Good abdominal breathing can increase the range of motion of the diaphragm, and the movement of the diaphragm directly affects the lung ventilation, increases vital capacity, and reduces lung infection. It is used to better characterize the physiological function indicators related to postoperative pulmonary complications.

[0067] Respiratory rate is an important parameter of respiratory function and can also effectively characterize physiological indicators related to postoperative pulmonary complications.

[0068] Optionally, the signal recording box includes a buffer and a memory for storing the acquired data in a specific format to completely preserve the entire data.

[0069] Optionally, the vest includes a device interface, a chest breathing belt, an abdominal breathing belt, and a data acquisition device, wherein both the chest and abdominal breathing belts include breathing cables, such as... Figure 2 As shown.

[0070] In this embodiment of the invention, a breathing belt with a breathing line can be embedded in the chest and abdomen of the vest. It is connected to a data acquisition device through a slot. The breathing signal is acquired at a frequency of 25 Hz by breathing induction volume plethysmography. The breathing signal is processed by a breathing signal conditioning circuit and the processed data is transmitted to the memory of the signal recording box to realize data transmission.

[0071] Optionally, the vest is connected to the signal recording box via magnetic attraction.

[0072] The structure design, which uses magnets to attract each other, facilitates installation and increases the adhesion strength of the connection.

[0073] In this embodiment of the invention, the device interface of the vest is connected to the signal recording box via magnetic attraction.

[0074] Optionally, the breathing band is made of fabric electrode material and soft skin-friendly material, which has good elasticity and comfort, and the sizes include XS, S, M, L, XL, XXL and XXXL to meet the needs of users with different body types.

[0075] In this embodiment of the invention, the electrocardiogram sensor acquires electrocardiogram signals at a frequency of 200 Hz, processes the electrocardiogram signals through an electrocardiogram signal conditioning circuit, and transmits the processed data to the memory of the signal recording box.

[0076] Optionally, the pulse oximeter is connected to the signal recording box via Bluetooth, and transmits the blood sample saturation data to the signal recording box in real time according to the data transmission protocol, and temporarily stores it in the buffer of the signal recording box.

[0077] Optionally, the signal recording box includes a wireless communication device connected to the data processing module, wherein the wireless communication device includes WIFI, a carrier network, or Bluetooth.

[0078] In this embodiment of the invention, the signal recording box packages the data stored in the buffer once per second and transmits it to the data processing module via a wireless communication device. In the data processing module, the acquired data is parsed, processed, and analyzed to achieve the system's real-time monitoring function. Simultaneously, this data is stored in a memory according to a specified format. By packaging the data once per second and transmitting it to the data processing module, the signal recording box achieves temporal alignment of the various physiological signals, reaching millisecond-level alignment, thus avoiding network latency caused by separate transmissions over the network.

[0079] Optionally, the system further includes a data display module connected to the data processing module for real-time display of the data received and processed by the data processing module.

[0080] The data display module can be a mobile phone, tablet, or computer.

[0081] In this embodiment of the invention, respiratory rate, electrocardiogram (ECG), and blood oxygen saturation values ​​are displayed in real time, along with trend graphs of these parameters. Based on the collected ECG and respiratory signals, respiratory rate and heart rate are calculated every second, and blood oxygen saturation values ​​are obtained from the pulse oximeter every second. By displaying these data, it is easy to understand the physiological parameters of the person being assessed during exercise in real time. If the person experiencing discomfort, timely measures can be taken to prevent accidents. For example, if the heart rate or blood oxygen saturation value exceeds the normal range, the device activates an alarm system, emitting a beeping sound to attract the attention of medical personnel. If necessary, the person being assessed can be instructed to stop exercising.

[0082] Optionally, the data processing module can be a remote server or a cloud server.

[0083] Optionally, the signal recording box further includes a power supply unit, which is connected to other modules in the signal recording box and has a lithium battery inside to provide power to the other modules.

[0084] Optionally, the signal recording box also includes indicator lights to indicate the working status. For example... Figure 3 As shown.

[0085] In this embodiment of the invention, pressing and holding the power button of the signal recording box for 3 seconds will cause the blue and green indicator lights to illuminate simultaneously, indicating that the power-on is normal. Subsequently, the blue indicator light will flash once every 3 seconds, and the green indicator light will turn off, indicating that the device is working normally. Then, vital sign signal data will be collected and stored.

[0086] Optionally, the step of processing the physiological signal based on its peak value and slope to obtain feature data includes:

[0087] The chest breathing signal and the abdominal breathing signal are superimposed, and the peak value and slope of the superimposed signal are used to determine the respiratory cycle and corresponding respiratory time period of the subject in each stage of the submaximal exercise experiment.

[0088] Determine the maximum respiratory rate based on the respiratory cycle of the first stage;

[0089] Based on the respiratory time period corresponding to the respiratory cycle in the first stage, the abdominal respiratory amplitude is integrated with the sum of the chest respiratory amplitude and the abdominal respiratory amplitude at the corresponding time to obtain the average abdominal respiratory contribution ratio in the first stage and the average abdominal respiratory contribution ratio in the second stage. The respiratory amplitude is the impedance amplitude obtained by processing the respiratory signal through electrical impedance tomography.

[0090] Optionally, after superimposing the chest breathing signal and the abdominal breathing signal, the process includes: using a bandpass filter to filter the signal to be processed to remove DC and heart-related impedance change data.

[0091] Optionally, after superimposing the chest breathing signal and the abdominal breathing signal, the process includes: preprocessing the signal to be processed using a digital high-pass filter to remove data below a preset frequency threshold.

[0092] Optionally, determining the respiratory cycle and corresponding respiratory time period of the subject at each stage of the submaximal exercise experiment using the peak value and slope of the superimposed signal includes:

[0093] The superimposed signal is divided into a waiting rising crossover state and a waiting falling crossover state;

[0094] Using the following formula, while waiting for the rising crossover state, the current minimum value of the superimposed signal is continuously updated to obtain the exhalation endpoint; while waiting for the falling crossover state, the current maximum value of the superimposed signal is continuously updated using the following formula to obtain the inhalation endpoint.

[0095] MICS = MICSfact-60 / maxBR

[0096] MDCS = MDCSfact - 60 / maxBR

[0097] Where MICSfact and MDCSfact are normalized parameters, maxBR is the assumed maximum respiratory rate, MICS represents the minimum identical cross interval, and MDCS represents the minimum different cross interval.

[0098] The respiratory cycle and the corresponding respiratory time period are determined based on adjacent respiratory endpoints and inspiratory endpoints.

[0099] In this embodiment of the invention, the same cross interval refers to the difference in consecutive zero-crossing times with the same slope, and different cross intervals refer to the difference in consecutive zero-crossing times with different slopes.

[0100] While waiting for an ascending crossover, the current minimum value of the signal is continuously updated using a formula. If an ascending zero-crossing point is detected and the conditions are met: same crossover interval > minimum same crossover interval, different crossover interval > minimum different crossover interval, then the final minimum value is confirmed as the final failure stage, and the algorithm's state is changed to waiting for a descending crossover.

[0101] While waiting for a falling crossover, the current maximum value of the signal is continuously updated using a formula. If a falling zero crossover point is detected and the conditions are met: same crossover interval > minimum same crossover interval, different crossover interval > minimum different crossover interval, then the final maximum value is confirmed as the final excitation phase, and the algorithm's state is changed to waiting for a rising crossover.

[0102] The time from the maximum impedance amplitude during the inspiratory phase to the minimum impedance amplitude during the corresponding expiratory phase is called a respiratory cycle.

[0103] Optionally, the submaximal exercise experiment also includes a preparation phase before conducting the submaximal exercise experiment.

[0104] This setup allows for comparison with vital signs during exercise to understand the changes in vital signs from resting to exercise states. It also prepares the body for the next stage, adjusting it to the optimal state for submaximal exercise testing.

[0105] Optionally, training a model using preoperative clinical and physiological characteristic data including the said feature data using a stochastic gradient descent algorithm includes:

[0106] Using preoperative clinical physiological characteristics as variables, and the probability of postoperative pulmonary complications as a function, we predicted the probability of postoperative pulmonary complications.

[0107] Further, optionally, the prediction function of the model obtained by training the preoperative clinical physiological characteristic data including the feature data using the stochastic gradient algorithm is:

[0108] h θ (x)=θ0x0+θ1x1+…+θ j x j

[0109] Where θ=[θ0θ1…θ j ], x = [x0x1…x j ], h θ(x) represents the complication probability of the array x formed by the input feature data of the evaluated person, θ is the vector formed by the feature coefficients, θ0 is a constant, and θ j Let x be the feature coefficient corresponding to the j-th feature data, and x be the vector formed by the feature data, x0 = 1, x j Let j be the feature data of the jth term.

[0110] Further, optionally, the loss function based on the prediction function is expressed as:

[0111]

[0112] Where y represents the actual label.

[0113] Further, optionally, the cost function obtained based on the loss function is expressed as:

[0114]

[0115] Where m is the number of consecutive preoperative clinical physiological parameter arrays in the dataset, and x (i) Let y be the i-th continuous preoperative clinical physiological parameter array. (i) For x (i) The corresponding label has a value of 0 or 1 for y.

[0116] Furthermore, optionally, the feature coefficients are initialized using gradient descent and updated incrementally until the optimal feature coefficients θ for the feature data in the continuous preoperative clinical physiological parameter array are obtained. j , is represented as:

[0117]

[0118] Where, θ (t) θ represents the parameter in the current t-th iteration. (t+1) Represents the model coefficients for iteration t+1.

[0119] In this embodiment of the invention, respiratory signals from eight minutes—one minute before and one minute after the six-minute walk test—are used for high-risk screening of postoperative pulmonary complications. The following three parameters are extracted: BR_max, Abo_contribute_wt, and Abo_contribute_recovery. BR_max represents the maximum respiratory rate during the first phase of the walk test; Abo_contribute_wt represents the average abdominal respiratory contribution ratio during the first phase of the walk test; and Abo_contribute_recovery represents the average abdominal respiratory contribution ratio during the second phase of the recovery test. Figure 4A1 is the chest breathing amplitude curve, and A2 is the abdominal breathing amplitude curve. The abdominal breathing contribution ratio of the current breath is calculated as A2 / (A1+A2). The average abdominal breathing contribution ratio of the first stage and the second stage is obtained by summing the abdominal breathing contribution ratios of the first stage and the second stage respectively.

[0120] The above three parameters are calculated according to the following formula:

[0121] ppcs=6.888+3324.691*Abo_contribute_wt-3325.002*Abo_contribute_recovery–0.238*Br_max

[0122] In this formula, ppcs represents postoperative pulmonary complications. The value of ppcs in the formula ranges from 0 to 1, with values ​​closer to 1 indicating a higher probability of postoperative pulmonary complications. For example, 0.7 represents a 70% probability of postoperative pulmonary complications.

[0123] Optionally, the system further includes:

[0124] The user management module, connected to the model evaluation module, is used to store user information, identify the person being evaluated, perform the submaximal exercise experiment, and store the corresponding experimental data.

[0125] In this embodiment of the invention, after searching for or selecting a corresponding person to be evaluated, the user management module assigns a 6-minute walking test task to that person. In addition, the module can also view the user's medical record information, such as basic information: outpatient number, height, weight, age, etc.; allergy history; diagnostic information; examination reports, etc., to understand the user's basic situation.

[0126] Optionally, the system further includes:

[0127] An alarm device, connected to the data acquisition module, is used to determine whether the electrocardiogram signal exceeds a preset threshold. If it exceeds the preset threshold, the alarm device issues a prompt signal, wherein the prompt signal includes at least one of the following or a combination thereof: sound, text, image, and light.

[0128] Optionally, the system further includes:

[0129] The experimental assessment module acquires the subject's Borg dyspnea score, PRE score for exertion, and New York heart function classification to determine whether the target subject can perform submaximal exercise experiments.

[0130] Use the system to conduct a 6-minute walking test task.

[0131] In this embodiment of the invention, the subject performs a 6-minute walking test task. This task is divided into three phases: preparation phase, testing phase, and recovery phase.

[0132] Preparation Phase: After the subject puts on and turns on the device, the preparation phase begins. The subject must remain seated for one minute while vital signs data are collected, including ECG, respiration, and blood oxygen saturation. This data is packaged and uploaded in real-time to the data processing module via a data network or Wi-Fi network. The data processing module analyzes and processes the data, displaying the heart rate, respiratory rate, and blood oxygen saturation data and their trends in real-time on the data display module. This objectively assesses the subject's health status. Simultaneously, the subject is assessed using the Borg dyspnea score, PRE score for exertion, and New York Heart Function Classification. The subject's subjective feelings are also considered to assess their health status. This combined subjective and objective approach determines whether the subject is suitable for the 6-minute walk test. Additionally, the system has upper and lower limits for heart rate alarms. If the subject's heart rate falls outside these limits during the test, the system emits a beeping alarm to alert medical personnel and prompt them to take timely action, such as stopping the walk test.

[0133] Testing phase: The subject walks as fast as possible for 6 minutes, and the heart rate, respiratory rate and blood oxygen saturation data and their trend graphs are displayed in real time. The vital signs of the subject are monitored in real time to confirm whether the subject's physical condition can continue the walking test. If discomfort occurs, measures are taken in time to avoid accidents. If the subject feels dizzy or weak in the limbs, the test is stopped and the patient is asked to sit down and rest.

[0134] Optionally, this stage can include adding laps, and the distance walked by the person being evaluated can be roughly estimated and understood by marking the number of laps.

[0135] The subject walks one lap in the 6-minute walking test area. Clicking the "Add Lap" button adds laps and marks them in the system, thus obtaining the approximate distance walked during the test. The walking distance is greater than or equal to the number of laps added multiplied by the number of meters per lap.

[0136] Recovery Phase: After walking for 6 minutes, the person being assessed maintains a seated posture for 1 minute. Monitoring data allows us to understand the person's health and recovery progress during this phase. For example, heart rate and respiratory rate should show a decreasing trend, while blood oxygen saturation should show an increasing trend. If any abnormalities occur, timely measures can be taken to prevent accidents. Simultaneously, the patient's heart rate recovery within 1 minute can also indirectly reflect their cardiopulmonary function.

[0137] Throughout this module, if the person being assessed experiences discomfort, they can describe the event and mark the handling method in the system. Events include chest tightness, dizziness, palpitations, angina, etc. Handling methods include resting, continuing execution, and ending execution to ensure the safety of the person being assessed.

[0138] Optionally, after the test, the entire walking test process can be reviewed to understand the vital signs of the subjects being assessed, including changes in the subjects' heart rate, respiratory rate and blood oxygen saturation based on the collected vital signals, the walking distance of the subjects, whether any discomfort symptoms occurred and the treatment methods, etc.

[0139] Furthermore, throughout the trial phase, the system can provide voice prompts to the assessee regarding which phase they are entering and how to proceed, reducing the workload of medical staff. For example, it may say, "Now entering the preparation phase, please remain seated for 1 minute." Ten seconds before entering the testing phase, it may say, "The testing phase is about to begin." After the preparation phase, it may say, "Entering the testing phase, this phase lasts 6 minutes, please walk as fast as possible." Ten seconds before entering the recovery phase, it may say, "The recovery phase is about to begin." After the testing phase, it may say, "Now entering the recovery phase, please remain seated for 1 minute."

[0140] After the 6-minute walk test, the data processing module processes and analyzes the monitored data to calculate characteristic data, which is then input into the evaluation model to calculate the incidence of postoperative pulmonary complications. Based on the predicted incidence, the doctor can determine the optimal surgical time and whether preoperative conditioning is necessary for the patient. For example, if the postoperative pulmonary complication rate is 70%, the doctor might decide to have the patient undergo breathing training for a period to improve their physical function. After a period of time, another 6-minute walk test would be conducted, and if the postoperative pulmonary complication rate is 10%, the doctor can then determine the appropriate time for surgery.

[0141] Model Evaluation Module: Postoperatively, the system assesses the surgical population by selecting corresponding options on the Melbourne Rating Scale to determine whether postoperative pulmonary complications occurred, evaluate model accuracy, and optimize model parameters. This evaluation method defines clinically significant PPC as meeting four or more positive indicators, including: body temperature >38℃; white blood cell count >11.2×10⁻⁶. 9 Or use of respiratory antibiotics; clinical diagnosis of pneumonia or lung infection; chest X-ray showing atelectasis / lung shadows; production of purulent (yellow / green) sputum different from preoperative sputum; positive sputum microbial culture; SpO2 <90% without oxygen inhalation; readmission to the intensive care unit due to respiratory problems or stay in the intensive care unit for >36 hours.

[0142] Compared with existing technologies, the system of this invention, through the collaborative interaction of its various modules, can predict the probability of pulmonary complications. Utilizing wearable devices to collect continuous respiratory physiological data during exercise to predict postoperative pulmonary complications in perioperative patients shows great promise and can overcome the limitations of traditional pulmonary function testing, which is constrained by the subjective cooperation of the assessee, and cardiopulmonary exercise testing, which is difficult to promote due to expensive equipment and complex operation. Furthermore, through theoretical analysis, empirical summarization, and model calculation, the system determines the correlation between characteristic data and the probability of postoperative pulmonary complications during submaximal exercise experiments. This allows for highly accurate probabilistic results in submaximal exercise experiments, facilitating accurate assessments for assessees, thereby providing guidance for surgical treatment planning, improving surgical outcomes, and mitigating risks.

[0143] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A wearable system for assessing the risk of postoperative pulmonary complications, characterized in that, include: The data acquisition module includes a vest, a pulse oximeter, and a signal recording box, used to collect physiological signals of the subject during a submaximal exercise experiment. The vest is connected to the signal recording box to collect respiratory signals, including thoracic and abdominal respiratory signals. The pulse oximeter is connected to the signal recording box to collect blood oxygenation signals. The signal recording box is equipped with an electrocardiogram (ECG) sensor to collect ECG signals. The submaximal exercise experiment includes a first stage of performing the submaximal exercise experiment and a second stage after performing the submaximal exercise experiment. A data processing module, connected to the data acquisition module, is used to process the physiological signal based on the peak value and slope of the physiological signal to obtain feature data, wherein the feature data includes the maximum respiratory rate in the first stage, the average abdominal breathing contribution ratio in the first stage, and the average abdominal breathing contribution ratio in the second stage. The model calculation module, connected to the data processing module, is used to input the feature data into a model obtained by training preoperative clinical and physiological feature data including the feature data using a stochastic gradient algorithm, and to calculate the incidence of postoperative pulmonary complications. The model evaluation module, connected to the model calculation module, determines whether the evaluated subject has postoperative pulmonary complications based on the Melbourne Group Rating Scale, evaluates the model accuracy, and optimizes the model parameters. The physiological signal is processed based on its peak value and slope to obtain feature data, including: The chest breathing signal and the abdominal breathing signal are superimposed, and the peak value and slope of the superimposed signal are used to determine the respiratory cycle and corresponding respiratory time period of the subject in each stage of the submaximal exercise experiment. Determine the maximum respiratory rate based on the respiratory cycle of the first stage; Based on the respiratory time period corresponding to the respiratory cycle in the first stage, the abdominal respiratory amplitude is integrated with the sum of the chest respiratory amplitude and the abdominal respiratory amplitude at the corresponding time to obtain the average abdominal respiratory contribution ratio in the first stage and the average abdominal respiratory contribution ratio in the second stage. The respiratory amplitude is the impedance amplitude obtained by processing the respiratory signal through electrical impedance tomography.

2. The wearable system for assessing the risk of postoperative pulmonary complications according to claim 1, characterized in that, The system also includes: The experimental assessment module obtains the subject's Borg dyspnea score, PRE score for exertion, and New York heart function classification to determine whether the subject can perform submaximal exercise experiments.

3. The wearable system for assessing the risk of postoperative pulmonary complications according to claim 1, characterized in that, After superimposing the chest breathing signal and the abdominal breathing signal, the process includes: using a bandpass filter to filter the signal to be processed to remove DC and heart-related impedance change data.

4. The wearable system for assessing the risk of postoperative pulmonary complications according to claim 1, characterized in that, After superimposing the chest breathing signal and the abdominal breathing signal, the process includes: preprocessing the signal to be processed using a digital high-pass filter to remove data below a preset frequency threshold.

5. The wearable system for assessing the risk of postoperative pulmonary complications according to claim 1, characterized in that, The method of using the peak value and slope of the superimposed signal to determine the respiratory cycle and corresponding respiratory time period of the subject at each stage of the submaximal exercise experiment includes: The superimposed signal is divided into a waiting rising crossover state and a waiting falling crossover state; Using the following formula, while waiting for the rising crossover state, the current minimum value of the superimposed signal is continuously updated to obtain the exhalation endpoint; while waiting for the falling crossover state, the current maximum value of the superimposed signal is continuously updated using the following formula to obtain the inhalation endpoint. MICS=MICSfact-60 / maxBR MDCS = MDCSfact - 60 / maxBR Where MICSfact and MDCSfact are normalized parameters, maxBR is the assumed maximum respiratory rate, MICS represents the minimum identical cross interval, and MDCS represents the minimum different cross interval. The respiratory cycle and the corresponding respiratory time period are determined based on adjacent respiratory endpoints and inspiratory endpoints.

6. The wearable system for assessing the risk of postoperative pulmonary complications according to claim 1, characterized in that, The system also includes: The data display module is connected to the data processing module and is used to display the data received and processed by the data processing module in real time.

7. The wearable system for assessing the risk of postoperative pulmonary complications according to claim 1, characterized in that, The submaximal exercise experiment also includes a preparation phase before conducting the submaximal exercise experiment.

8. A wearable system for assessing the risk of postoperative pulmonary complications according to claim 1 or 4, characterized in that, The system also includes: The user management module, connected to the model evaluation module, is used to store user information, identify the person being evaluated, perform the submaximal exercise experiment, and store the corresponding experimental data.

9. A wearable system for assessing the risk of postoperative pulmonary complications according to claim 1, characterized in that, The system also includes: An alarm device, connected to the data acquisition module, is used to determine whether the electrocardiogram signal exceeds a preset threshold. If it exceeds the preset threshold, the alarm device issues a prompt signal, wherein the prompt signal includes at least one of the following or a combination thereof: sound, text, image, and light.

Images