Motion assessment method and system based on closed-loop cardiopulmonary function digital twin model

Abstract

The present invention provides a motion assessment method and system based on a closed-loop cardiopulmonary function digital twin model. The motion assessment method includes the following steps: obtaining multi-dimensional physiological signals collected synchronously based on a single contact point; constructing a time dynamic feature library using long-term continuous monitoring data; analyzing individual physiological pattern characteristics in combination with a deep learning algorithm; establishing a progressively optimized closed-loop cardiopulmonary function digital twin model; achieving maximum exercise volume, heart rate variation range and cardiac output changes in a virtual space based on the closed-loop cardiopulmonary function digital twin model, and issuing a warning signal before reaching the patient's cardiopulmonary function limit. The present invention has the advantages of continuous monitoring and early prediction of danger, especially the ability to accurately predict activity restrictions caused by cardiac function, lung function, and metabolism, which is of great benefit to disease management and the implementation of daily exercise prescriptions, ensuring that patients can obtain maximum benefits while avoiding injuries caused by excessive exercise.

Description

Technical Field

[0001] The present invention relates to the field of cardiopulmonary coupling technology, and in particular to a motion assessment method and system based on a closed-loop cardiopulmonary function digital twin model. Background Art

[0002] Cardiopulmonary dysfunction is a core issue in many chronic diseases, such as chronic obstructive pulmonary disease (COPD), coronary artery disease, and heart failure. These conditions not only impact patients' daily quality of life but can also lead to serious complications and even death. Therefore, personalized treatment plans are crucial for individuals with cardiopulmonary diseases, and effective monitoring of patients with cardiopulmonary limitations is particularly important. Cardiopulmonary exercise testing (CPET) or the six-minute walk test (6MWT) allow physicians to accurately assess a patient's cardiopulmonary function and tailor rehabilitation or medication regimens accordingly. However, this assessment is intermittent and does not allow for timely adjustments. Intermittent assessments are significantly limited in situations such as patients with cardiopulmonary limitations who are at high risk of acute exacerbations, COPD patients experiencing acute exacerbations, or heart disease patients experiencing sudden myocardial infarction. In such situations, real-time monitoring and closed-loop control mechanisms can adjust treatment strategies based on real-time data, thereby improving efficacy and minimizing side effects.

[0003] With advances in technology, digital twin technology has been widely used in various fields, particularly in healthcare, for simulating and optimizing patient treatment plans, surgical planning, and drug development. It also has potential for assessing exercise safety in patients with cardiopulmonary limitations. However, current digital twin models fail to cover the cardiopulmonary coupling, metabolic regulation, circulatory system, and closed-loop control of patients with cardiopulmonary limitations (such as those with chronic obstructive pulmonary disease (COPD) and heart failure), making it impossible to build a corresponding monitoring system. Specifically, the related technologies have the following shortcomings:

[0004] Lack of cardiopulmonary coupling analysis: Although previous studies have explored the application of cardiopulmonary coupling in sleep monitoring and demonstrated its ability to effectively reflect the interaction between the heart and respiratory system, there is currently a lack of mature solutions for how to use cardiopulmonary coupling for dynamic monitoring and immediate feedback in exercise or rehabilitation scenarios, especially in patients with limited cardiopulmonary function. This limits the ability to design personalized training programs for patients with cardiopulmonary limitations.

[0005] Insufficient understanding of metabolic regulation mechanisms: Changes in energy metabolism are a key characteristic of diseases such as heart failure. However, existing digital twin models are often static and fail to fully account for these complex metabolic regulation mechanisms, making it difficult to accurately predict the specific effects of different interventions on patients.

[0006] Simplified treatment of the circulatory system: Most existing digital twin platforms simplify the details of the circulatory system when building human models. This simplification can lead to inaccurate simulation of important physiological parameters (such as blood oxygen saturation and blood pressure fluctuations), especially for patients with cardiopulmonary limitations, who require special attention. This can lead to deviations when developing treatment strategies for such patients.

[0007] Lack of closed-loop control: In practical applications, closed-loop control mechanisms link the cardiopulmonary system, exercise, metabolism, and circulation. However, current digital twin technology applications for patients with cardiopulmonary limitations do not fully consider the importance of closed-loop control. For example, when cardiopulmonary function reaches its limits, a comprehensive digital twin model must gradually reduce exercise intensity to ensure the continued operation of the virtual closed-loop system and prevent collapse. Alternatively, the model must accurately predict the limits of quantitative exercise and promptly terminate potentially harmful exercise.

[0008] Challenges of personalized healthcare: While digital twin technology offers the potential for precision medicine, achieving true personalization requires overcoming challenges such as data integration and model validation. Especially for patients with complex cardiopulmonary limitations and complex pathologies, ensuring that the established digital twin models are both sufficiently detailed and responsive to individual differences remains a pressing issue.

[0009] In summary, although the current digital twin technology has made significant progress in many aspects, it is still not perfect in terms of cardiopulmonary coupling, metabolic regulation, and simulation and support of the circulatory system for patients with cardiopulmonary limitations. In particular, the lack of closed-loop regulation makes it difficult for this technology to meet clinical needs.

[0010] Therefore, it is necessary to provide a new method to solve the above technical problems. Summary of the Invention

[0011] In order to achieve the above-mentioned purpose and other advantages of the present invention, the first object of the present invention is to provide a motion assessment method based on a closed-loop cardiopulmonary function digital twin model, comprising the following steps:

[0012] Acquire multi-dimensional physiological signals collected synchronously based on a single contact point;

[0013] Use long-term continuous monitoring data to build a temporal dynamic feature library;

[0014] Combined with deep learning algorithms to analyze individual physiological pattern characteristics;

[0015] Establish a progressively optimized closed-loop cardiopulmonary function digital twin model;

[0016] Based on the closed-loop cardiopulmonary function digital twin model, the maximum amount of exercise, the range of heart rate changes and the changes in cardiac output are achieved in the virtual space, and a warning signal is issued before the patient's cardiopulmonary function limit is reached.

[0017] Furthermore, the multi-dimensional physiological signal is configured to be obtained by synchronously collecting signals from an ECG sensor, a PPG sensor and a pressure sensor.

[0018] Furthermore, the steps of establishing a progressively optimized closed-loop cardiopulmonary function digital twin model include:

[0019] During the cold start phase, the steady-state parameters of the model are gradually optimized through long-term continuous monitoring data;

[0020] Real-time update of dynamic parameters based on autoregressive plus exogenous variable model;

[0021] Through the multi-time scale parameter coupling algorithm, accurate simulation of physiological responses from seconds to long-term can be achieved.

[0022] Furthermore, the step of updating dynamic parameters in real time based on the autoregressive plus exogenous variable model includes:

[0023] Through the dynamic changes of the measured physiological parameters, the CMA-ES method is used to fit the physiological parameters and dynamically identify the model parameters that have the greatest impact on the system response and the model parameter values that minimize the difference between the model prediction and the experimental data at a specific time point;

[0024] Evaluate dynamic parameters;

[0025] Implement a deterministic and parameterized multi-input single-output model for each dynamic parameter so that the model can capture the trend of the parameter over time;

[0026] The autoregressive plus exogenous variable model uses the trend of parameter changes over time to predict future states;

[0027] Model performance was evaluated by calculating the prediction error, which is the ratio of the difference between the experimental variable value and the simulated predicted value.

[0028] Furthermore, the step of evaluating the dynamic parameters includes:

[0029] Perform time-specific fitting of the model using sequential applications of a static fitting strategy;

[0030] Dynamic modeling of model parameters to reflect their time trends;

[0031] The strategy is validated by considering the future steady-state response of the modeled physiological system.

[0032] Furthermore, the step of using the autoregressive plus exogenous variable model to predict future states by utilizing the trend of parameter changes over time includes:

[0033] For each model, the following inputs were used to initialize the model: time difference between recordings, average weekly physical activity, average daily sleep hours, and anaerobic threshold;

[0034] After the start of continuous measurement, the model was updated at a set frequency, the polynomial coefficients of each model were estimated using the least squares method, and the optimal order of the autoregressive plus exogenous variable model was selected according to the Akaike Information Criterion.

[0035] Furthermore, the cardiopulmonary function limit is set according to the cardiac function reserve and the pulmonary function reserve.

[0036] Furthermore, the closed-loop cardiopulmonary function digital twin model includes a respiratory control system, an autonomic nervous system, a respiratory system, a gas transport and exchange system, and a circulatory system; wherein,

[0037] The respiratory control system monitors the oxygen and carbon dioxide levels in the blood through peripheral chemoreceptors and the degree of lung expansion through lung expansion receptors;

[0038] The autonomic nervous system regulates heart rate and vasoconstriction through the sympathetic nerves, regulates heart rate and vasodilation through the parasympathetic nerves, and regulates peripheral blood flow resistance through the perception of metabolites;

[0039] The respiratory system is neurally regulated, and restricted breathing in the lungs also causes changes in the levels of carbon dioxide and oxygen in the blood, which in turn affects various chemical receptors;

[0040] The gas transport exchange pairs exchange oxygen and carbon dioxide in tissues and alveoli;

[0041] The heart in the circulatory system is responsible for pumping blood, and the flow of blood throughout the body is regulated by vascular tension and blood flow resistance.

[0042] Furthermore, the arteries in the circulatory system are represented by RC circuit connections, the veins in the circulatory system are described by RC circuits and unidirectional diodes, and the capillaries in the circulatory system are represented by RC circuits connected in series with resistors or by L and T topological structures.

[0043] Furthermore, the overall blood flow pattern in the autonomic nervous system is:

[0044] ;

[0045] in, represents the venous oxygen concentration in the ith circulation region, is the reference venous oxygen concentration value, is a static function, indicating the state of metabolic control; is the slope coefficient;

[0046] This static function is used in a first-order dynamic control block to control the peripheral arterial and venous resistance in each circulatory region:

[0047] ;

[0048] in, is the change in resistance caused by metabolic control, is the time constant of metabolic control, To control the gain;

[0049] The venous resistance of the i-th vascular segment , and its final control formula is:

[0050] ;

[0051] in, is the set point value of the venous resistance of the i-th vascular segment, It is a change in venous resistance caused by metabolic control;

[0052] The control formula of peripheral arterial resistance is:

[0053] ;

[0054] in, is the time-varying peripheral arterial resistance, is the set arterial peripheral resistance, is a constant parameter that represents the arterial resistance when the sympathetic vasoconstrictor effect is completely eliminated. is the change caused by sympathetic control, is the effect of metabolic control on resistance, is the amount of change caused by metabolic control;

[0055] The relationship between metabolism and respiration is mediated by oxygen consumption:

[0056] ;

[0057] in, is the resting oxygen consumption, and are the oxygen consumption of the left and right legs respectively. is the respiratory quotient, and the subscript HF represents the corresponding changes in heart failure. The power used to do work for movement.

[0058] Furthermore, the exchange process of carbon dioxide between the alveoli and veins during the breathing process of the respiratory control system is:

[0059] ;

[0060] in, is the oxygen concentration in the arterial blood flow, is the oxygen concentration in the venous blood flow, is the blood volume of the ith region, is the arterial blood flow through the ith region, is the venous blood flow through the ith region, is the oxygen consumption of the ith region;

[0061] By introducing arterial and venous blood flow, as well as oxygen consumption, the change of carbon dioxide over time during breathing is simulated:

[0062] ;

[0063] in, is the concentration of carbon dioxide in the arterial blood stream, is the concentration of carbon dioxide in the venous bloodstream;

[0064] Starting from the mass balance equation, the oxygen concentration in arterial blood is derived by combining the oxygen exchange process between tissues and blood. It is described by the following mass balance equation:

[0065] ;

[0066] The rate of change of oxygen concentration in arterial blood is approximated using the mass balance equation under steady-state conditions:

[0067] ;

[0068] in, represents the ventilation volume calculated by the model, is the partial pressure of oxygen in the arterial blood of the upper body, is the partial pressure of carbon dioxide in the arterial blood of the upper body, Indicates that when the partial pressure of carbon dioxide in arterial blood exceeds this value, a new ventilation cycle is triggered. represent the ventilation control gains for oxygen and carbon dioxide, respectively, is a constant parameter;

[0069] ventilation Expressed as frequency and tidal volume :

[0070] ;

[0071] in, and is the fitting constant;

[0072] The effective tidal volume for alveolar ventilation is:

[0073] ;

[0074] in, is the dead space ratio.

[0075] Furthermore, the construction of the closed-loop cardiopulmonary function digital twin model also includes:

[0076] The initial parameter values are set by using a parameter initial value library based on scientific research data of different disease populations to make the fitting starting point close to the physiological reasonable range;

[0077] By presetting physiological correlation constraints between parameters, the convergence stability of the model and the reliability of the results are improved.

[0078] Furthermore, the step of improving the convergence stability and result reliability of the model by presetting physiological correlation constraints between parameters includes:

[0079] Simulation of patients with cardiopulmonary restriction, including chronic obstructive pulmonary disease simulation, heart failure simulation, and pulmonary hypertension simulation;

[0080] The peripheral blood flow resistance is regulated by the formula in the autonomic nervous system, and the oxygen content in the blood is monitored in real time, thereby triggering changes in ventilation. By activating the sympathetic nerves, changes in heart rate, blood pressure and blood flow are simultaneously caused to achieve exercise simulation;

[0081] Gravity simulation is achieved by adding the influence of gravity to the blood force.

[0082] The second object of the present invention is to provide a motion assessment system based on a closed-loop cardiopulmonary function digital twin model, which applies the above-mentioned motion assessment method and includes a multi-dimensional physiological signal acquisition module and a motion assessment module; wherein,

[0083] The multi-dimensional physiological signal acquisition module synchronously acquires multi-dimensional physiological signals based on a single contact point;

[0084] The motion assessment module is used to obtain multi-dimensional physiological signals collected synchronously based on a single contact point, use long-term continuous monitoring data to build a time dynamic feature library, combine deep learning algorithms to analyze individual physiological pattern characteristics, and establish a progressively optimized closed-loop cardiopulmonary function digital twin model. Based on the closed-loop cardiopulmonary function digital twin model, the maximum amount of exercise, heart rate variation range and cardiac output changes are achieved in the virtual space, and a warning signal is issued before the patient's cardiopulmonary function limit is reached.

[0085] Compared with the prior art, the present invention has the following beneficial effects:

[0086] The present invention provides a motion assessment method and system based on a closed-loop digital twin model of cardiopulmonary function, which has the advantages of continuous monitoring and early prediction of danger. In particular, it can accurately predict activity restrictions caused by cardiac function, lung function, and metabolism, which is of great benefit to disease management and the implementation of daily exercise prescriptions, ensuring that patients can obtain maximum benefits while avoiding injuries caused by excessive exercise.

[0087] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and to implement it according to the contents of the description, the following preferred embodiments of the present invention are described in detail with reference to the accompanying drawings. The specific implementation methods of the present invention are given in detail by the following embodiments and the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0089] Figure 1 The process of motion assessment method based on closed-loop cardiopulmonary function digital twin model Figure 1 ;

[0090] Figure 2 The process of motion assessment method based on closed-loop cardiopulmonary function digital twin model Figure 2 ;

[0091] Figure 3 This is a schematic diagram of the multi-dimensional physiological signal acquisition module;

[0092] Figure 4 Schematic diagram of the cardiopulmonary-circulatory-metabolic closed-loop interactive system;

[0093] Figure 5 Schematic diagram of the distributed hemodynamic model;

[0094] Figure 6 Schematic diagram of modeling the small veins in the legs of S21;

[0095] Figure 7 Schematic diagram of modeling the S4 inferior carotid artery;

[0096] Figure 8 This is a simulation diagram of the neuromodulation system;

[0097] Figure 9 This is a diagram showing changes in blood pressure after posture changes in healthy individuals and heart failure patients;

[0098] Figure 10 Schematic diagram of changes in blood pressure after exercise in healthy people and patients with heart failure;

[0099] Figure 11 Schematic diagram of a virtual test evaluation for heart failure patients at 70W exercise volume;

[0100] Figure 12 This is a schematic diagram of the virtual test response of a healthy person under the same amount of exercise;

[0101] Figure 13 Schematic diagram of the motion assessment system based on the closed-loop cardiopulmonary function digital twin model;

[0102] Figure 14 It is a schematic diagram of computer equipment;

[0103] Figure 15 A schematic diagram of a computer-readable storage medium. DETAILED DESCRIPTION

[0104] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. It should be noted that, without conflict, the embodiments or technical features described below can be arbitrarily combined to form new embodiments.

[0105] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0106] The figure numbers in this application are only used to distinguish the various steps in the scheme and are not used to limit the execution order of the various steps. The specific execution order is subject to the description in the specification.

[0107] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0108] The present invention can collect data in real time through a closed-loop model of the cardiopulmonary, metabolic, and circulatory systems and multiple modalities, predict the direction of cardiopulmonary function, and issue timely warnings when the cardiopulmonary limit is approaching or about to be reached.

[0109] Example 1

[0110] This embodiment provides a motion assessment method based on a closed-loop digital twin model of cardiopulmonary function. By using a closed-loop model of the cardiopulmonary, metabolic, and circulatory systems and multimodal real-time data collection, it predicts the direction of cardiopulmonary function and issues a timely warning when the cardiopulmonary limit is approaching or about to be reached.

[0111] Specifically, a motion assessment method based on a closed-loop cardiopulmonary function digital twin model, such as Figure 1-Figure 2 As shown, the following steps are included:

[0112] S100, acquiring multi-dimensional physiological signals synchronously collected based on a single contact point;

[0113] In some embodiments, the multi-dimensional physiological signal is configured to be obtained by synchronously collecting signals from an ECG sensor, a PPG sensor, and a pressure sensor.

[0114] Despite significant progress in wearable monitoring technology, deploying sensors at multiple locations still faces limitations such as complex operation and poor user experience. To address this, this embodiment innovatively develops a single-point multimodal sensing system, achieving efficient individualized modeling through the following technical approaches: synchronously collecting multi-dimensional physiological signals from a single contact point; constructing a temporal dynamic feature library using long-term continuous monitoring data; combining deep learning algorithms to analyze individual physiological pattern characteristics; and establishing a progressively optimized digital twin model.

[0115] This single-point integration solution significantly improves wearability and user compliance while ensuring data quality.

[0116] like Figure 3 As shown, the hardware system includes a PPG (photoplethysmography) sensor, an ECG (electrocardiogram) sensor, and a pressure sensor.

[0117] The ECG (electrocardiogram) sensor has two leads located on the wrist, detecting the heart's electrical signals through the skin's surface. More leads allow for more detailed determination of ECG amplitude and direction, providing an assessment of heart function in patients with heart failure.

[0118] The PPG (photoplethysmography) sensor, located on the inside of the wrist, measures changes in blood volume by emitting light and detecting reflected light. It is sensitive to the model's volume parameters. In this embodiment, the sensor primarily measures wrist blood vessels. It can also monitor physiological parameters such as heart rate, blood oxygen saturation, and respiratory rate.

[0119] The pressure sensor is located on the inside of the wrist, close to the PPG sensor, and is used to assess wrist blood pressure and obtain key information such as cardiac output.

[0120] All sensors are non-invasive, comfortable to wear, and suitable for long-term monitoring.

[0121] The assessment of individual cardiopulmonary health can be determined by posture changes and isopower exercises. By obtaining individual cardiopulmonary health parameters in a relatively standard and safe way and forming a digital twin model, more intense and dangerous "virtual space tests" can be performed. The cardiopulmonary-metabolism-circulatory model has significantly different responses for healthy people and patients with heart failure, such as Figure 9-10 shown.

[0122] A core goal of this embodiment is to enable low-load cardiopulmonary function monitoring and cardiopulmonary hemodynamic analysis during exercise, focusing on solving the coupling mechanism between autonomic nervous system regulation and vascular autoregulation. To this end, this embodiment innovatively develops a multi-system coupled regulation prediction model with three major technical features:

[0123] Digital twin infrastructure: Supports progressive learning of long-term wearable monitoring data, enabling personalized modeling of the cardiopulmonary-metabolic-circulatory system, including over 400 dynamic parameters with physiological relevance;

[0124] High-dimensional parameter optimization technology: Based on the time series data of limited sensor nodes, a global optimization algorithm is used for parameter fitting;

[0125] Mechanism analysis function: Quantify the interaction between autonomic nervous system and hemodynamic regulation, dynamically reflect the nonlinear response characteristics of the physiological system, and provide theoretical model support for personalized health assessment.

[0126] Figure 4 Demonstrates the interaction between the respiratory control system and the cardiovascular system, specifically involving the following key components:

[0127] In some embodiments, the closed-loop cardiopulmonary function digital twin model includes a respiratory control system, an autonomic nervous system, a respiratory system, a gas transport and exchange system, and a circulatory system; wherein,

[0128] Respiratory control system: monitors blood oxygen (O2) and carbon dioxide (CO2) levels through peripheral chemoreceptors and lung expansion through lung expansion receptors, which are directly related to respiratory rate;

[0129] Autonomic nervous system: regulates heart rate and vasoconstriction through the sympathetic nervous system, regulates heart rate and vasodilation through the parasympathetic nervous system, and regulates peripheral blood flow resistance through the perception of metabolites;

[0130] Respiratory system: regulated by nerves. At the same time, restricted lung breathing will also cause CO2 and O2 levels in the blood, thereby affecting various chemical receptors;

[0131] Gas transport and exchange: the exchange of oxygen and carbon dioxide in tissues and alveoli;

[0132] Circulatory system: The heart is responsible for pumping blood, and the flow of blood throughout the body is regulated by vascular tension and blood flow resistance.

[0133] Closed-loop pathways are achieved through nerves and various receptors. Blood gas levels are monitored peripherally, while lung expansion and pressure receptors monitor lung status and blood pressure. Sympathetic and parasympathetic nerves regulate the cardiovascular system to adapt to respiratory demands. Oxygen and carbon dioxide are exchanged in the lungs and tissues, ensuring that blood gas concentrations remain within normal ranges. Effective gas exchange is achieved through respiratory muscles and lung power, while mechanical ventilation can assist breathing.

[0134] This example describes the model in detail from three perspectives: the circulatory system, the metabolic system, and the cardiopulmonary coupling.

[0135] Circulatory system: To explain the multi-component design of the systemic circulation, this example divides the blood flow of the body into four major anatomical parts. Figure 5 The circuit scheme described in

[15] is used to explain the head, chest, abdomen, and legs. In this design, arteries, capillaries, and veins are represented by different circuit models depending on the diversity of the segments.

[0136] The artery is divided into seven parts, which are represented by RC circuit connections. The specific parameters are shown in Table 2-3. The specific implementation method is referenced Figure 5 Arteries are generally modeled as vascular elasticity (capacitance C) and blood flow resistance (R). Figure 7 Schematic diagram of modeling the S4 lower carotid artery. In addition to these two parameters, the vein also needs to consider the venous valve that prevents backflow, which is described by a unidirectional diode. Figure 6 Schematic diagram of modeling the leg venules in S21. Each vascular segment is described by 2-6 parameters. In the upper body, these include the lower carotid artery (LOC) through S4 and the superior carotid artery (UPC) through S5; in the chest, the thoracic aorta (ThA) through S10; in the abdomen, the abdominal aorta (UM) through S11; and in the legs through S12, S13, and S14, corresponding to the femoral artery (FA), the leg artery (LSA), and the leg artery (LAR), respectively.

[0137] Capillaries are represented by RC circuits connected in series with a resistor, or by a topological structure of L and T. Their connections include the head, liver, upper mesenteric (UM), lower mesenteric (UM), kidney (KIDN), and leg capillaries (LCAP), corresponding to the seven regions represented by S2, S6, S15, and S16. Finally, they connect to the heart through venous closures. In the upper body, these capillaries are supplied by the head venule (HSV) in S7, the jugular vein (JV) in S8, and the superior vena cava (SVC) in S9. In the chest, the thoracic vein (TVC) is represented by S24, and in the abdomen, the abdominal vena cava (AVC) is represented by S23. In the legs, the femoral vein (FV), the leg venule (LSV), and the leg venule (LVE) are represented by S20, S21, and S22, respectively.

[0138] The cardiopulmonary circuit is designed in the S1 to S3 blocks, including their connection to the circulatory system block. The whole body circuit is implemented by connecting to the designated electrical components S24 and S4 blocks.

[0139] Fluctuations in hemodynamic parameters such as blood pressure, blood flow, and heart rate are measured using a wristwatch placed close to the thoracic artery. Upper arm and wrist blood pressure can be synchronized with the thoracic artery through a transfer function.

[0140] like Figure 8 As shown, neuroregulation (metabolic closed-loop regulation): simulates the neuroregulatory mechanisms of the cardiovascular system, specifically how the carotid sinus reflex, sympathetic nerve, and vagal nerve activity affect cardiac function and vascular tone. Specifically, this design includes parameter settings related to the carotid sinus nerve pathway, sympathetic nerve efferent pathway, and vagal nerve efferent pathway.

[0141] Carotid sinus afferent pathway (Pn, fmin, fmax, ka, touz, toup): These variables represent the characteristics of the carotid sinus baroreceptors, such as basal pressure (Pn), minimum and maximum frequency responses (fmin, fmax), gain (ka), and time constants (touz, toup). The carotid sinus is a baroreceptor located at the carotid bifurcation that is sensitive to changes in blood pressure and provides feedback to the central nervous system via neural signals to help regulate blood pressure.

[0142] Sympathetic efferent pathways (fes1, fes0, kes, fesmin) and vagal efferent pathways (fev0, fev1, kev, fcs0): Parameters in these two components define the efferent response characteristics of the sympathetic and vagal nerves. Sympathetic nerves typically increase heart rate and contractility, while vagal nerves have the opposite effect, slowing heart rate and reducing contractility. These parameters can be used to calculate the impact of these two neural activities on cardiac function.

[0143] Effector parameters (Gx, toux, Dx): This section defines the impact of different physiological parameters on the heart's pumping capacity (Emaxlv, Emaxrv), blood flow resistance (Rsp, Rep), uncompressed volume (Vusv, Vuev), and the time constants of sympathetic and vagal action (Ts, Tv).

[0144] The relationship between Fes and Emaxlv, Emaxrv Rsp, Rep, Vusv, and Vuev implements a neural feedback control system with a very similar control loop. For example, for Emaxlv, it works by simulating the effects of the sympathetic and vagus nerve efferent pathways on cardiac function. Specifically, the sympathetic signals responsible for regulation After a series of processing steps, including sinusoidal wave generation to simulate periodic neural activity, nonlinear transformation to reflect the characteristics of the physiological system, and dynamic response adjustment to reflect the heart's response to neural modulation, the final output, Emaxlv, represents the maximum contractile force of the left ventricle, reflecting the regulatory effect of neural feedback mechanisms on the heart's pumping capacity. This model integrates neural signal generation and transformation with the simulation of cardiac effects, demonstrating the complex interactions of neural modulation in the cardiovascular system.

[0145] In the vagus nerve pathway module, and master control sympathetic signal The relationship is as follows:

[0146] (1);

[0147] in, and The heart rate is controlled by stimulating the sinoatrial node. is a sensitive parameter of the efferent vagus nerve, and is the benchmark value of the fitting function, is the fitting coefficient, The pressure-volume relationship and dynamic behavior of the left and right ventricular (RV) modules are parameters at the edge of the value range and can be used to simulate the functional changes of the right ventricle during the cardiac cycle.

[0148] When metabolic activity increases, metabolic relaxation reduces peripheral resistance, thereby increasing blood flow. This mechanism ensures that blood flow increases in tissues that require more oxygen. In health, cardiac output (CO) and the alveolar-venous oxygen difference increase significantly, indicating that the heart's pumping function is enhanced to support the increased oxygen demand. In heart failure (HF), despite changes in total peripheral resistance (TPR) and single-leg resistance (Rlla), the overall blood flow pattern still reflects the body's attempt to adapt to oxygen demand through increased metabolic relaxation.

[0149] (2);

[0150] in, represents the venous oxygen concentration in the ith circulation region, is the reference venous oxygen concentration value, is a static function, indicating the state of metabolic control; is the slope coefficient; the static function It is used in the first-order dynamic control block to control the peripheral arterial and venous resistance of each circulation region.

[0151] (3);

[0152] in, is the change in resistance caused by metabolic control, is the time constant of metabolic control, is the control gain; for the venous resistance of the i-th vascular segment , and its final control formula is:

[0153] (4);

[0154] in, is the set point value of the venous resistance of the i-th vascular segment, Changes in venous resistance are caused by metabolic control; when the venous oxygen concentration is equal to the reference value, venous resistance is maintained at the set point value.

[0155] A key mechanism during exercise is sympathetic blockade, which determines the interplay between the baroreflex and metabolic systems in regulating peripheral circulation. Metabolic control can counteract sympathetic vasoconstriction in the exercised area because local factors and substances reduce the sensitivity of vascular smooth muscle to sympathetic tone. To simulate this sympathetic blockade, this embodiment implements control of peripheral arterial resistance as follows:

[0156] (5);

[0157] in, is the time-varying peripheral arterial resistance, is the set arterial peripheral resistance, is a constant parameter that represents the arterial resistance when the sympathetic vasoconstrictor effect is completely eliminated. is the change caused by sympathetic control, is the effect of metabolic control on resistance, is the amount of change caused by metabolic control;

[0158] This model accurately simulates the effects of metabolic control on peripheral vascular resistance, helping to understand the regulatory mechanisms of the cardiovascular system under different physiological conditions. The regulatory effects of metabolic control on hemodynamic parameters are particularly important during exercise or pathological conditions.

[0159] The relationship between metabolism and respiration is mediated by oxygen consumption, as shown in formula (6) to formula (10):

[0160] ;

[0161] in, is the resting oxygen consumption, and are the oxygen consumption of the left and right legs respectively. The respiratory quotient is the ratio of oxygen consumption to carbon dioxide production; it takes into account the oxygen consumption ratio of different energy sources (such as glucose, fat, etc.). The subscript HF represents the corresponding changes in heart failure. The power used to do work for movement.

[0162] Respiratory control: Equation (11) describes the exchange of carbon dioxide between the alveoli and veins during respiration. By introducing the pulmonary shunt rate, pulmonary venous blood flow, and alveolar carbon dioxide concentration, we can simulate how carbon dioxide changes over time during respiration.

[0163] (11);

[0164] in, is the oxygen concentration in the arterial blood flow, is the oxygen concentration in the venous blood flow, is the blood volume of the ith region, is the arterial blood flow through the ith region, is the venous blood flow through the ith region, is the oxygen consumption in the i-th region. This formula describes the exchange of carbon dioxide and oxygen in tissues. By introducing arterial and venous blood flow, as well as oxygen consumption, it is possible to simulate how carbon dioxide changes over time during respiration.

[0165] (12);

[0166] in, is the concentration of carbon dioxide in the arterial blood stream, is the concentration of carbon dioxide in the venous bloodstream;

[0167] This formula also describes the change in carbon dioxide concentration over time. The RQ of pure carbohydrate metabolism is close to 1, while the RQ of fat metabolism is close to 0.7.

[0168] Starting from the mass balance equation, the oxygen concentration in arterial blood is derived by combining the oxygen exchange process between tissues and blood. It can be described by the following mass balance equation:

[0169] (13);

[0170] In practical applications, it is usually assumed and are known, and their rates of change can be estimated or measured by other means. Therefore, this embodiment directly uses the mass balance equation under steady-state conditions to approximately calculate the rate of change of oxygen concentration in arterial blood.

[0171] ;

[0172] in, Represents the ventilation volume calculated by the model, usually in liters per minute (L / min), which is the output of the model. It is the partial pressure of oxygen in the upper body arterial blood, reflecting the concentration of oxygen in the blood. It is the partial pressure of carbon dioxide in the upper body arterial blood, reflecting the concentration of carbon dioxide in the blood. It means that when the partial pressure of carbon dioxide in the arterial blood exceeds this value, a new ventilation cycle is triggered, which can be understood as a critical point for starting the respiratory regulation mechanism. represent the ventilatory control gains for oxygen and carbon dioxide, respectively, i.e. how they affect changes in ventilation. is a constant parameter that determines the intensity of the effect of oxygen partial pressure on ventilation. ) can be expressed as a frequency ( ) and tidal volume ( ):

[0173] ;

[0174] in, and is the fitting constant;

[0175] Effective tidal volume of alveolar ventilation ( )as follows:

[0176] (20);

[0177] in, is the dead space ratio, which accounts for the proportion of air in the airway that is dead space. Dead space / TV = −0.012⋅(peak O₂ uptake) + 0.611. For healthy individuals, a peak oxygen uptake of 34 ml / min / kg is assumed; for patients with heart failure, a peak oxygen uptake of 15 ml / min / kg is assumed. These values are used to calculate the corresponding dead space ratio and, in turn, the effective tidal volume.

[0178] Although this embodiment constructs a precise model containing more than 400 parameters, the parameter fitting process may face challenges such as local optimal solutions or slow convergence. To this end, this embodiment adopts two key optimization strategies: first, a parameter initial value library is established based on scientific research data of different disease populations to ensure that the fitting starting point is close to the physiologically reasonable range; specifically, the initial parameter values are set by using the parameter initial value library established based on scientific research data of different disease populations to ensure that the fitting starting point is close to the physiologically reasonable range;

[0179] Secondly, by presetting physiological correlation constraints between parameters, the model's convergence stability and result reliability are significantly improved. This dual optimization design enables personalized fitting of complex models while maintaining physiological significance and clinical practicality.

[0180] Furthermore, the step of improving the convergence stability and result reliability of the model by presetting physiological correlation constraints between parameters includes:

[0181] Simulation of patients with cardiopulmonary restriction, including chronic obstructive pulmonary disease simulation, heart failure simulation, and pulmonary hypertension simulation;

[0182] Specifically, in chronic obstructive pulmonary disease (COPD): respiratory rate increases to compensate for hypoventilation; tidal volume decreases because the patient's lung elasticity is reduced, resulting in an increase in end-expiratory lung volume.

[0183] Heart failure: respiratory rate increases as a compensatory mechanism to reduce cardiac output; tidal volume may be normal or slightly lower depending on the presence of pulmonary congestion; fluid: simulates edema and cardiac load by adjusting the PV curve.

[0184] Pulmonary hypertension: PAP (pulmonary artery pressure) increases, reflecting increased pulmonary vascular resistance; HR (heart rate) increases to compensate for the decrease in stroke volume.

[0185] This example specifically focuses on the physiological impairments associated with heart failure, such as reduced baroreflex sensitivity, weakened metabolic control, impaired vasodilation, and hyperventilation. By adjusting model parameters, the cardiopulmonary dysfunction exhibited by heart failure patients during exercise can be reproduced.

[0186] Exercise simulation: Regulate peripheral blood flow resistance using formulas (2) to (10), monitor blood oxygen levels in real time, and trigger changes in ventilation. By activating the sympathetic nerves, changes in heart rate, blood pressure, and blood flow are simultaneously caused. When oxygen consumption and oxygen carrying capacity reach a balance, the subject is considered safe. If ventilation is insufficient to compensate for the oxygen consumption gap and heart rate continues to rise significantly, a warning of the current status is required.

[0187] Gravity simulation: During the transition from recumbent to standing, gravity forces blood to pool in the lower extremities. Venous return is reduced, leading to a decrease in cardiac output, a drop in arterial blood pressure, and an immediate decrease in blood flow to the brain. This decrease in arterial blood pressure acts on baroreceptors in the carotid and aortic walls, resulting in a decrease in parasympathetic nerve activity and an increase in sympathetic nerve activation, completing baroreflex-mediated autonomic regulation. This decrease in parasympathetic nerve activity induces a rapid increase in heart rate (within 1-2 cardiac cycles), while sympathetic nerve activation produces a slower increase (within 6-8 cardiac cycles), increasing vascular resistance, vascular tone, and cardiac contractility, further increasing heart rate. Gravity simulation can be achieved by adding gravity to the forces acting on blood. Gravity simulation can be used to assess orthostatic hypotension and its severity.

[0188] The specific parameters are shown in Tables 1 to 5.

[0189] Table 1 Common symbols and their meanings

[0190]

[0191] Table 2 Parameter settings for different blood vessels

[0192]

[0193] Table 3 Cardiovascular parameters of exercise simulation in healthy state

[0194]

[0195] Table 4 Parameters of ventilation and muscle contraction models

[0196]

[0197] Table 5 Index table of patients with heart failure

[0198]

[0199] S110, constructing a time dynamic feature library using long-term continuous monitoring data;

[0200] S120, combining deep learning algorithms to analyze individual physiological pattern characteristics;

[0201] S130, establish a progressively optimized closed-loop cardiopulmonary function digital twin model;

[0202] It should be noted that this digital twin model adopts a dynamic iteration mechanism:

[0203] In some embodiments, the step of establishing a progressively optimized closed-loop cardiopulmonary function digital twin model includes:

[0204] During the cold start phase, steady-state parameters such as nervous system sensitivity are gradually optimized through long-term continuous monitoring data;

[0205] Real-time updates of dynamic parameters such as vascular elasticity (e.g., second-by-second heart rate regulation under stress) based on an autoregressive plus exogenous variable (ARX) model;

[0206] Through the multi-time scale parameter coupling algorithm, accurate simulation of physiological responses from seconds to long-term can be achieved.

[0207] This hierarchical iterative architecture not only ensures the stability of basic physiological parameters, but also captures rapid stress responses, making the digital twin simulation closer to the real physiological process. After fitting, each test subject generates a set of digital twin models that can achieve the maximum amount of exercise, heart rate range and cardiac output changes in the virtual space, such as Figure 11-12 As shown, it can send out early warning signals before the patient's cardiopulmonary function limits are reached.

[0208] The specific fitting process of the model is as follows:

[0209] The algorithm flow of the dynamic fitting strategy is mainly divided into several key steps, including data processing, model parameter estimation, optimal order selection and model verification. The following is a detailed description of the algorithm flow:

[0210] In some embodiments, the step of updating dynamic parameters in real time based on the autoregressive plus exogenous variable model includes:

[0211] First, since the cardiopulmonary-metabolism-circulation model has many parameters, this embodiment measures the dynamic changes of parameters such as heart rate, respiration, cardiac output, blood pressure, and blood oxygen, fits the physiological parameters through the CMA-ES method, and dynamically identifies those model parameters that have the greatest impact on the system response, and the model parameter values that minimize the difference between the physiological model prediction and the experimental data at a specific time point.

[0212] CMA-ES is a stochastic global optimization algorithm based on adaptive and evolutionary principles, offering advantages in multi-parameter fitting speed and accuracy. The algorithm demonstrated excellent convergence speed, precision, and accuracy in the application scenario of this embodiment. By using experimental data (incremental step sizes of VO2 and VCO2 from resting to anaerobic threshold) and employing the root mean square error (RMSE) as a cost function, the CMA-ES algorithm effectively optimizes model parameters and improves prediction accuracy.

[0213] Evaluate dynamic parameters;

[0214] Furthermore, dynamic parameter evaluation includes the following three key steps:

[0215] Perform time-specific fitting of the model using sequential applications of a static fitting strategy;

[0216] Dynamic modeling of model parameters to reflect their time trends;

[0217] The strategy is validated by considering the future steady-state response of the modeled physiological system.

[0218] In order to evaluate a dynamic parameter for plausibility, it must be optimized at least twice at two consecutive time points. Only those parameters that show a significant change during the optimization process are modeled as dynamic parameters, while the other parameters retain their previous values, whether nominal, normalized or previously optimized.

[0219] Implement a deterministic and parameterized multi-input single-output model for each dynamic parameter so that the model can capture the trend of the parameter over time;

[0220] Each dynamic parameter implements a deterministic and parameterized Multiple-Input Single-Output (MISO) model. The MISO model design allows for parameter prediction based on past values and temporal history associated with important factors, such as habits, events, or illnesses. This design enables the model to capture trends in parameter changes over time and use these trends to predict future states.

[0221] The autoregressive plus exogenous variable model uses the trend of parameter changes over time to predict future states;

[0222] Trend prediction corresponds to an autoregressive plus exogenous variable (ARX) model, which considers the influence of past values of outputs and inputs on current values. Each model is initialized using the following inputs: the time difference between recordings, average weekly physical activity, average daily sleep hours, and anaerobic threshold (AT). After the start of continuous measurement, the model is updated at a set frequency (e.g., once every minute). The polynomial coefficients of each model are estimated using the least squares method, and the optimal ARX model order is selected based on the Akaike Information Criterion (AIC).

[0223] Finally, the model performance was evaluated by calculating the prediction error (PE), which is the ratio of the difference between the experimental variable value and the simulation prediction value.

[0224] S140. Based on the closed-loop cardiopulmonary function digital twin model, the maximum amount of exercise, the range of heart rate changes, and the changes in cardiac output are achieved in the virtual space, and a warning signal is issued before the patient's cardiopulmonary function limit is reached.

[0225] Furthermore, the cardiopulmonary function limit can be set by cardiac reserve (obtained from regular testing) and pulmonary reserve (maximum ventilation). The method and model provided in this embodiment can continuously and dynamically adjust the warning value under different exercise levels and changing hemodynamic parameters.

[0226] This embodiment addresses the shortcomings of existing technologies by introducing advanced sensing technology, improved data processing algorithms, and a deeper understanding of physiology, thereby providing a safer and more effective health management solution for this special population. It also utilizes a closed-loop control mechanism to monitor and adjust the patient's treatment process in real time, improving overall treatment effectiveness.

[0227] Example 2

[0228] This embodiment provides a motion assessment system based on a closed-loop digital twin model of cardiopulmonary function. Through closed-loop models of the cardiopulmonary, metabolic, and circulatory systems and multimodal real-time data collection, it predicts the direction of cardiopulmonary function and issues timely warnings when approaching or about to approach cardiopulmonary limits.

[0229] Specifically, a motion assessment system based on a closed-loop cardiopulmonary function digital twin model applies the above-mentioned motion assessment method. For a detailed description of the method, please refer to the corresponding description in the above-mentioned method embodiment, which will not be repeated here. Figure 13 As shown, the system 200 includes a multi-dimensional physiological signal acquisition module 210 and a motion assessment module 220; wherein,

[0230] The multi-dimensional physiological signal acquisition module synchronously acquires multi-dimensional physiological signals based on a single contact point;

[0231] The motion assessment module is used to obtain multi-dimensional physiological signals collected synchronously based on a single contact point, use long-term continuous monitoring data to build a time dynamic feature library, combine deep learning algorithms to analyze individual physiological pattern characteristics, and establish a progressively optimized closed-loop cardiopulmonary function digital twin model. Based on the closed-loop cardiopulmonary function digital twin model, the maximum amount of exercise, heart rate variation range and cardiac output changes are achieved in the virtual space, and a warning signal is issued before the patient's cardiopulmonary function limit is reached.

[0232] Based on the technical solutions of the above embodiments, optionally, the multi-dimensional physiological signals are configured to be obtained by synchronously collecting signals from an ECG sensor, a PPG sensor, and a pressure sensor.

[0233] Based on the technical solution of the above embodiment, optionally, the step of establishing a progressively optimized closed-loop cardiopulmonary function digital twin model includes:

[0234] During the cold start phase, the steady-state parameters of the model are gradually optimized through long-term continuous monitoring data;

[0235] Real-time update of dynamic parameters based on autoregressive plus exogenous variable model;

[0236] Through the multi-time scale parameter coupling algorithm, accurate simulation of physiological responses from seconds to long-term can be achieved.

[0237] Based on the technical solution of the above embodiment, optionally, the step of updating the dynamic parameters in real time based on the autoregressive plus exogenous variable model includes:

[0238] Through the dynamic changes of the measured physiological parameters, the CMA-ES method is used to fit the physiological parameters and dynamically identify the model parameters that have the greatest impact on the system response and the model parameter values that minimize the difference between the model prediction and the experimental data at a specific time point;

[0239] Evaluate dynamic parameters;

[0240] Implement a deterministic and parameterized multi-input single-output model for each dynamic parameter so that the model can capture the trend of the parameter over time;

[0241] The autoregressive plus exogenous variable model uses the trend of parameter changes over time to predict future states;

[0242] Model performance was evaluated by calculating the prediction error, which is the ratio of the difference between the experimental variable value and the simulated predicted value.

[0243] Based on the technical solution of the above embodiment, optionally, the step of evaluating the dynamic parameters includes:

[0244] Perform time-specific fitting of the model using sequential applications of a static fitting strategy;

[0245] Dynamic modeling of model parameters to reflect their time trends;

[0246] The strategy is validated by considering the future steady-state response of the modeled physiological system.

[0247] Based on the technical solution of the above embodiment, optionally, the step of predicting the future state by using the trend of parameter changes over time through the autoregressive plus exogenous variable model includes:

[0248] For each model, the following inputs were used to initialize the model: time difference between recordings, average weekly physical activity, average daily sleep hours, and anaerobic threshold;

[0249] After the start of continuous measurement, the model was updated at a set frequency, the polynomial coefficients of each model were estimated using the least squares method, and the optimal order of the autoregressive plus exogenous variable model was selected according to the Akaike Information Criterion.

[0250] Based on the technical solution of the above embodiment, optionally, the cardiopulmonary function limit is set according to the cardiac function reserve and the pulmonary function reserve.

[0251] Based on the technical solution of the above embodiment, optionally, the closed-loop cardiopulmonary function digital twin model includes a respiratory control system, an autonomic nervous system, a respiratory system, a gas transport and exchange system, and a circulatory system; wherein,

[0252] The respiratory control system monitors the oxygen and carbon dioxide levels in the blood through peripheral chemoreceptors and the degree of lung expansion through lung expansion receptors;

[0253] The autonomic nervous system regulates heart rate and vasoconstriction through the sympathetic nerves, regulates heart rate and vasodilation through the parasympathetic nerves, and regulates peripheral blood flow resistance through the perception of metabolites;

[0254] The respiratory system is neurally regulated, and restricted breathing in the lungs also causes changes in the levels of carbon dioxide and oxygen in the blood, which in turn affects various chemical receptors;

[0255] The gas transport exchange pairs exchange oxygen and carbon dioxide in tissues and alveoli;

[0256] The heart in the circulatory system is responsible for pumping blood, and the flow of blood throughout the body is regulated by vascular tension and blood flow resistance.

[0257] Based on the technical solutions of the above embodiments, optionally, the arteries in the circulatory system are represented by RC circuit connections, the veins in the circulatory system are described by RC circuits and unidirectional diodes, and the capillaries in the circulatory system are represented by RC circuits connected in series with resistors or by a topological structure of L and T.

[0258] Based on the technical solution of the above embodiment, optionally, the overall blood flow pattern in the autonomic nervous system is:

[0259] ;

[0260] in, represents the venous oxygen concentration in the ith circulation region, is the reference venous oxygen concentration value, is a static function, indicating the state of metabolic control; is the slope coefficient;

[0261] This static function is used in a first-order dynamic control block to control the peripheral arterial and venous resistance in each circulatory region:

[0262] ;

[0263] in, is the change in resistance caused by metabolic control, is the time constant of metabolic control, To control the gain;

[0264] The venous resistance of the i-th vascular segment , and its final control formula is:

[0265] ;

[0266] in, is the set point value of the venous resistance of the i-th vascular segment, It is a change in venous resistance caused by metabolic control;

[0267] The control formula of peripheral arterial resistance is:

[0268] ;

[0269] in, is the time-varying peripheral arterial resistance, is the set arterial peripheral resistance, is a constant parameter that represents the arterial resistance when the sympathetic vasoconstrictor effect is completely eliminated. is the change caused by sympathetic control, is the effect of metabolic control on resistance, is the amount of change caused by metabolic control;

[0270] The relationship between metabolism and respiration is mediated by oxygen consumption:

[0271] ;

[0272] in, is the resting oxygen consumption, and are the oxygen consumption of the left and right legs respectively. is the respiratory quotient, and the subscript HF represents the corresponding changes in heart failure. The power used to do work for movement.

[0273] Based on the technical solution of the above embodiment, optionally, the exchange process of carbon dioxide between the alveoli and veins during the breathing process of the breathing control system is:

[0274] ;

[0275] in, is the oxygen concentration in the arterial blood flow, is the oxygen concentration in the venous blood flow, is the blood volume of the ith region, is the arterial blood flow through the ith region, is the venous blood flow through the ith region, is the oxygen consumption of the ith region;

[0276] By introducing arterial and venous blood flow, as well as oxygen consumption, the change of carbon dioxide over time during breathing is simulated:

[0277] ;

[0278] in, is the concentration of carbon dioxide in the arterial blood stream, is the concentration of carbon dioxide in the venous bloodstream;

[0279] Starting from the mass balance equation, the oxygen concentration in arterial blood is derived by combining the oxygen exchange process between tissues and blood. It is described by the following mass balance equation:

[0280] ;

[0281] The rate of change of oxygen concentration in arterial blood is approximated using the mass balance equation under steady-state conditions:

[0282] ;

[0283] ;

[0284] in, represents the ventilation volume calculated by the model, is the partial pressure of oxygen in the arterial blood of the upper body, is the partial pressure of carbon dioxide in the arterial blood of the upper body, Indicates that when the partial pressure of carbon dioxide in arterial blood exceeds this value, a new ventilation cycle is triggered. represent the ventilation control gains for oxygen and carbon dioxide, respectively, is a constant parameter;

[0285] ventilation Expressed as frequency and tidal volume :

[0286] ;

[0287] in, and is the fitting constant;

[0288] The effective tidal volume for alveolar ventilation is:

[0289] ;

[0290] in, is the dead space ratio.

[0291] Based on the technical solution of the above embodiment, optionally, the construction of the closed-loop cardiopulmonary function digital twin model further includes:

[0292] The initial parameter values are set by using a parameter initial value library based on scientific research data of different disease populations to make the fitting starting point close to the physiological reasonable range;

[0293] By presetting physiological correlation constraints between parameters, the convergence stability of the model and the reliability of the results are improved.

[0294] Based on the technical solution of the above embodiment, optionally, the step of improving the convergence stability and result reliability of the model by presetting physiological correlation constraints between parameters includes:

[0295] Simulation of patients with cardiopulmonary restriction, including chronic obstructive pulmonary disease simulation, heart failure simulation, and pulmonary hypertension simulation;

[0296] The peripheral blood flow resistance is regulated by the formula in the autonomic nervous system, and the oxygen content in the blood is monitored in real time, thereby triggering changes in ventilation. By activating the sympathetic nerves, changes in heart rate, blood pressure and blood flow are simultaneously caused to achieve exercise simulation;

[0297] Gravity simulation is achieved by adding the influence of gravity to the blood force.

[0298] This embodiment addresses the shortcomings of existing technologies by introducing advanced sensing technology, improved data processing algorithms, and a deeper understanding of physiology, thereby providing a safer and more effective health management solution for this special population. It also utilizes a closed-loop control mechanism to monitor and adjust the patient's treatment process in real time, improving overall treatment effectiveness.

[0299] Example 3

[0300] A computer device 300, such as Figure 14 As shown, the system includes a memory 310, a processor 320, and a computer program 330 stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a motion assessment method based on a closed-loop cardiopulmonary function digital twin model are implemented. For a detailed description of the method, please refer to the corresponding description in the above method embodiment and will not be repeated here.

[0301] Example 4

[0302] A computer-readable storage medium such as Figure 15 As shown, a computer program is stored thereon, and when the computer program is executed by the processor, the steps of a motion assessment method based on a closed-loop cardiopulmonary function digital twin model are implemented. For a detailed description of the method, please refer to the corresponding description in the above method embodiment, and no further details will be given here.

[0303] The number of devices and processing scales described herein are intended to simplify the description of the present invention. Applications, modifications, and variations of the present invention will be readily apparent to those skilled in the art.

[0304] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the description and implementation methods. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

[0305] The apparatus, computer device, non-volatile computer storage medium, and method provided in the embodiments of this specification correspond to each other. Therefore, the apparatus, computer device, and non-volatile computer storage medium also have similar beneficial technical effects as the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding apparatus, computer device, and non-volatile computer storage medium will not be repeated here.

[0306] Those skilled in the art will also appreciate that, in addition to implementing the controller in pure computer-readable program code, it is entirely possible to implement the same functionality by programming the method steps logically, such as through logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be considered structures within the hardware component. Alternatively, the devices for implementing various functions can be considered both software units implementing the method and structures within the hardware component.

[0307] The systems, devices, or units described in the above embodiments can be implemented by computer chips or physical devices, or by products with certain functions. For ease of description, the above devices are described separately by function, with each unit described separately. Of course, when implementing one or more embodiments of this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware components.

[0308] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Thus, the embodiments of this specification may take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware. Furthermore, the embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0309] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0310] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0311] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0312] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0313] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program units. Generally, program units include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program units may be located in local and remote computer storage media, including storage devices.

[0314] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

[0315] The foregoing is merely an example of the present invention and is not intended to limit the present invention to one or more embodiments. It will be apparent to those skilled in the art that various modifications and variations may be made to the present invention to one or more embodiments. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention to one or more embodiments shall be included within the scope of the claims of the present invention to one or more embodiments.

Claims

1. A motion assessment method based on a closed-loop cardiopulmonary function digital twin model, characterized in that: The following steps are involved: Acquire multi-dimensional physiological signals collected synchronously based on a single contact point; Use long-term continuous monitoring data to build a temporal dynamic feature library; Combined with deep learning algorithms to analyze individual physiological pattern characteristics; Establish a progressively optimized closed-loop cardiopulmonary function digital twin model; Based on the closed-loop cardiopulmonary function digital twin model, the maximum amount of exercise, the range of heart rate changes, and the changes in cardiac output are achieved in the virtual space, and a warning signal is issued before the patient's cardiopulmonary function limit is reached; The multi-dimensional physiological signal is configured to be obtained by synchronously collecting signals from an ECG sensor, a PPG sensor, and a pressure sensor; The steps of establishing a progressively optimized closed-loop cardiopulmonary function digital twin model include: During the cold start phase, the steady-state parameters of the model are gradually optimized through long-term continuous monitoring data; Real-time update of dynamic parameters based on autoregressive plus exogenous variable model; Through the multi-time scale parameter coupling algorithm, accurate simulation of physiological responses from seconds to long-term can be achieved; The step of updating dynamic parameters in real time based on the autoregressive plus exogenous variable model includes: Through the dynamic changes of the measured physiological parameters, the CMA-ES method is used to fit the physiological parameters and dynamically identify the model parameters that have the greatest impact on the system response and the model parameter values that minimize the difference between the model prediction and the experimental data at a specific time point; Evaluate dynamic parameters; Implement a deterministic and parameterized multi-input single-output model for each dynamic parameter so that the model can capture the trend of the parameter over time; The autoregressive plus exogenous variable model uses the trend of parameter changes over time to predict future states; Evaluate model performance by calculating prediction error, which is the ratio of the difference between the experimental variable value and the simulated predicted value. The closed-loop cardiopulmonary function digital twin model includes a respiratory control system, an autonomic nervous system, a respiratory system, a gas transport and exchange system, and a circulatory system; wherein, The respiratory control system monitors the oxygen and carbon dioxide levels in the blood through peripheral chemoreceptors and the degree of lung expansion through lung expansion receptors; The autonomic nervous system regulates heart rate and vasoconstriction through the sympathetic nerves, regulates heart rate and vasodilation through the parasympathetic nerves, and regulates peripheral blood flow resistance through the perception of metabolites; The respiratory system is neurally regulated, and restricted breathing in the lungs also causes changes in the levels of carbon dioxide and oxygen in the blood, which in turn affects various chemical receptors; The gas transport exchange pairs exchange oxygen and carbon dioxide in tissues and alveoli; The heart in the circulatory system is responsible for pumping blood, and the flow of blood throughout the body is regulated by vascular tension and blood flow resistance; The arteries in the circulatory system are represented by RC circuits, the veins in the circulatory system are described by RC circuits and unidirectional diodes, and the capillaries in the circulatory system are represented by RC circuits connected in series with resistors or by L and T topological structures; The overall blood flow pattern in the autonomic nervous system is: ， in, represents the venous oxygen concentration in the ith circulation region, is the reference venous oxygen concentration value, is a static function, indicating the state of metabolic control; is the slope coefficient; This static function is used in a first-order dynamic control block to control the peripheral arterial and venous resistance in each circulatory region: ， in, is the change in resistance caused by metabolic control, is the time constant of metabolic control, To control the gain; The venous resistance of the i-th vascular segment , and its final control formula is: ， in, is the set point value of the venous resistance of the i-th vascular segment, It is a change in venous resistance caused by metabolic control; The control formula of peripheral arterial resistance is: ， in, is the time-varying peripheral arterial resistance, is the set arterial peripheral resistance, is a constant parameter that represents the arterial resistance when the sympathetic vasoconstrictor effect is completely eliminated. is the change caused by sympathetic control, is the effect of metabolic control on resistance, is the amount of change caused by metabolic control; The relationship between metabolism and respiration is mediated by oxygen consumption: ， ， ， ， ， in, is the resting oxygen consumption, and are the oxygen consumption of the left and right legs respectively. is the respiratory quotient, and the subscript HF represents the corresponding changes in heart failure. The power used to do work for movement; The exchange process of carbon dioxide between the alveoli and veins during the breathing process of the respiratory control system is: ， in, is the oxygen concentration in the arterial blood flow, is the oxygen concentration in the venous blood flow, is the blood volume of the ith region, is the arterial blood flow through the ith region, is the venous blood flow through the ith region, is the oxygen consumption of the ith region; By introducing arterial and venous blood flow, as well as oxygen consumption, the change of carbon dioxide over time during breathing is simulated: ， in, is the concentration of carbon dioxide in the arterial blood stream, is the concentration of carbon dioxide in the venous bloodstream; Starting from the mass balance equation, the oxygen concentration in arterial blood is derived by combining the oxygen exchange process between tissues and blood. It is described by the following mass balance equation: ， The rate of change of oxygen concentration in arterial blood is approximated using the mass balance equation under steady-state conditions: ， ， in, represents the ventilation volume calculated by the model, is the partial pressure of oxygen in the arterial blood of the upper body, is the partial pressure of carbon dioxide in the arterial blood of the upper body, Indicates that when the partial pressure of carbon dioxide in arterial blood exceeds this value, a new ventilation cycle is triggered. represent the ventilation control gains for oxygen and carbon dioxide, respectively, is a constant parameter; ventilation Expressed as frequency and tidal volume : ， ， ， ， in, and is the fitting constant; The effective tidal volume for alveolar ventilation is: ， in, is the dead space ratio.

2. The motion assessment method based on a closed-loop cardiopulmonary function digital twin model according to claim 1, characterized in that: The step of evaluating the dynamic parameters comprises: Perform time-specific fitting of the model using sequential applications of a static fitting strategy; Dynamic modeling of model parameters to reflect their time trends; The strategy is validated by considering the future steady-state response of the modeled physiological system.

3. The motion assessment method based on a closed-loop cardiopulmonary function digital twin model according to claim 1, characterized in that: The steps of predicting future states by using the trend of parameters changing over time through the autoregressive plus exogenous variable model include: For each model, the following inputs were used to initialize the model: time difference between recordings, average weekly physical activity, average daily sleep hours, and anaerobic threshold; After the start of continuous measurement, the model was updated at a set frequency, the polynomial coefficients of each model were estimated using the least squares method, and the optimal order of the autoregressive plus exogenous variable model was selected according to the Akaike Information Criterion.

4. The motion assessment method based on a closed-loop cardiopulmonary function digital twin model according to claim 1, characterized in that: The cardiopulmonary function limit is set according to the cardiac function reserve and the pulmonary function reserve.

5. The motion assessment method based on a closed-loop cardiopulmonary function digital twin model according to claim 1, characterized in that: The construction of the closed-loop cardiopulmonary function digital twin model also includes: The initial parameter values are set by using a parameter initial value library based on scientific research data of different disease populations to make the fitting starting point close to the physiological reasonable range; By presetting physiological correlation constraints between parameters, the convergence stability of the model and the reliability of the results are improved.

6. The motion assessment method based on a closed-loop cardiopulmonary function digital twin model according to claim 5, characterized in that: The steps of improving the convergence stability and result reliability of the model by presetting physiological correlation constraints between parameters include: Simulation of patients with cardiopulmonary restriction, including chronic obstructive pulmonary disease simulation, heart failure simulation, and pulmonary hypertension simulation; The peripheral blood flow resistance is regulated by the formula in the autonomic nervous system, and the oxygen content in the blood is monitored in real time, thereby triggering changes in ventilation. By activating the sympathetic nerves, changes in heart rate, blood pressure and blood flow are simultaneously caused to achieve exercise simulation; Gravity simulation is achieved by adding the influence of gravity to the blood force.

7. A motion assessment system based on a closed-loop cardiopulmonary function digital twin model, applying the motion assessment method according to any one of claims 1 to 6, characterized in that: It includes multi-dimensional physiological signal acquisition module and motion assessment module; among them, The multi-dimensional physiological signal acquisition module synchronously acquires multi-dimensional physiological signals based on a single contact point; The motion assessment module is used to obtain multi-dimensional physiological signals collected synchronously based on a single contact point, use long-term continuous monitoring data to build a time dynamic feature library, combine deep learning algorithms to analyze individual physiological pattern characteristics, and establish a progressively optimized closed-loop cardiopulmonary function digital twin model. Based on the closed-loop cardiopulmonary function digital twin model, the maximum amount of exercise, heart rate variation range and cardiac output changes are achieved in the virtual space, and a warning signal is issued before the patient's cardiopulmonary function limit is reached.

Images