A dual-airway ischemic preconditioning training device and intelligent training method

Abstract

This invention belongs to the technical field of ischemic preconditioning training devices, specifically relating to a dual-airflow ischemic preconditioning training device and an intelligent training method. The training device includes modules for data acquisition, pressure setting, dual-airflow control, coordinated adjustment, and main control. It quantifies the physiological coupling degree by calculating the difference in blood pressure between the two arms and the pulse wave velocity, and simultaneously applies this coupling influence coefficient as a feedforward term and the coupling degree coefficient as a feedback parameter during the training period, thereby achieving adaptive and precise coordinated control of the two airflows. The method includes: acquiring basic physiological data and setting an initial target pressure individually; using a sliding mode control algorithm incorporating the physiological coupling term to track and stabilize the pressure in real time; and coordinating the adjustment of the target pressure in both arms based on the quantified coupling degree during the rest period. This invention effectively improves pressure control accuracy and training safety, enhances user experience, and enables personalized ischemic preconditioning training.

Description

Technical Field

[0001] This invention belongs to the technical field of ischemic preconditioning training devices, specifically relating to a dual-airway ischemic preconditioning training device and an intelligent training method. Background Technology

[0002] The ischemic preconditioning training device works by intermittently inducing ischemic activity in the upper arm to help improve the organ's remote ischemic preconditioning ability. By training the limb to adapt to ischemia, it mobilizes the body's own resistance to ischemia and hypoxia, aiming for self-repair and compensation. The treatment mechanism of the ischemic preconditioning training device is to block blood flow to the limb, prompting the body to release anti-myocardial ischemia factors. When these factors return to the heart, combined with neuroregulation, they improve the patient's myocardial tolerance to ischemia and open the channels of coronary microvessels.

[0003] Most of these products use a uniform pressure setting to train both arms simultaneously, ignoring the objective differences in baseline blood pressure and pressure tolerance between the two arms due to physiological structure (different distances between the left and right arms from the heart) or pathological conditions. This may pose safety risks to the arm with higher blood pressure or lower tolerance, while the other arm may suffer from insufficient stimulation, affecting training effectiveness. Although a few devices attempt to set different target pressures for the two arms, their underlying airflow systems often suffer from interference such as air source competition and pressure coupling, making it difficult to achieve truly independent and precise closed-loop control of the two pressure pathways.

[0004] Secondly, the blood pressure in both arms is not a separate system; they are closely connected through the cardiovascular system, exhibiting a significant physiological coupling. Applying or releasing pressure to one arm affects the blood pressure and vascular status of the other arm in real time through nerve reflexes and hemodynamic changes. Existing training devices ignore this dynamic interaction; their control models are open-loop and static. Specifically, during adaptive adjustments, the system adjusts the pressure on only one arm based on its own feedback (such as resting blood pressure), without considering the physiological impact on the other arm after this adjustment, or the chain reaction of this impact feeding back. This can lead to an unstable oscillating state during training, failing to converge to a steady state that is safest and most effective for the user overall, and potentially even triggering unpredictable risks. Summary of the Invention

[0005] To address the aforementioned problems in the existing technology, this invention provides a dual-airway ischemic preconditioning training device and an intelligent training method, which solves the problem that the existing technology cannot simultaneously address bilateral pressure interference and the coupling effect caused by the physiology of both arms, thus resulting in inaccurate training.

[0006] The objective of this invention can be achieved through the following technical solution: a dual-airway ischemia preconditioning training device, comprising:

[0007] The data acquisition module is used to obtain the user's baseline systolic blood pressure in both arms and the initial air pressure reference values ​​of the left and right air pumps. It also collects real-time blood pressure data, pulse wave conduction velocity data, and real-time air pressure data of the left and right air pumps in both arms during training and rest periods.

[0008] The pressure setting module is electrically connected to the data acquisition module and is used to set different preset pressure increments based on the acquired baseline systolic pressure of the left and right arms according to individual differences. The initial target training pressure of the left and right arms is the baseline systolic pressure of the corresponding side plus the preset pressure increment.

[0009] The dual-air-path control module includes a left air-path control unit and a right air-path control unit that are independent of each other but have the same structure, which are used to drive the left air pump and the right air pump respectively. The dual-air-path control module has a built-in sliding mode control sub-module, which is used to synchronously start the left and right air pumps to pressurize according to the initial target training pressure, and to track the initial target training pressure in real time based on the sliding mode control algorithm in order to maintain the stability of the pressure of the two arms during the training period.

[0010] The main control module is electrically connected to the data acquisition module, pressure setting module and dual-air path control module respectively. It is used to coordinate the timing and collaborative work of each module, control data transmission and command issuance, and realize closed-loop control of the entire training process.

[0011] The collaborative adjustment module, electrically connected to the main control module, is used to calculate the coupling coefficient by combining the blood pressure difference between the two arms with the pulse wave conduction velocity during the rest period of training, and to use the blood pressure feedback data of one side of the airway as a correlation parameter to collaboratively adjust the current target training pressure of the left and right arms to eliminate the interference of unilateral pressure adjustment on the other side. The coupling coefficient is used to quantify the cardiovascular physiological coupling relationship between the user's two arms.

[0012] Preferably, the sliding mode control submodule includes:

[0013] The model building unit is used to establish a dual-path nonlinear state model containing physiological coupling terms and disturbance terms; the physiological coupling terms are introduced by a coupling influence coefficient quantified by the real-time systolic pressure difference between the two arms, and the disturbance terms are quantified by the difference between the real-time systolic pressure and the baseline systolic pressure.

[0014] The sliding surface design unit is used to design basic sliding surfaces for the left and right arms based on pressure tracking error and error change rate, and to construct a modified sliding surface for dual-arm collaborative control through a collaborative correction coefficient.

[0015] The control law solving unit is used to derive the discrete sliding mode control law using the exponential approach law, and to suppress chattering by replacing the sign function with a saturation function, and outputs control voltage signals to drive the left and right air pumps respectively.

[0016] Preferably, it also includes a mode selection and safety monitoring module electrically connected to the main control module, used for:

[0017] It can receive user input and switch between multiple training modes, including single-arm training mode, dual-arm synchronous training mode, and loop training mode.

[0018] Before executing the initial target training pressure set by the pressure setting module, a double safety check is performed: first, the pressure value is constrained within the adjustment range of the training device hardware, and an additional pressure upper limit is applied to specific groups of people; then, the difference between the initial target training pressures of the two arms is calculated, and if it exceeds the preset threshold, the pressure on the high-pressure side is automatically adjusted until the difference meets the requirements.

[0019] It monitors gas leaks, pressure anomalies, and signal interference in real time, and generates corresponding error codes and processing prompts.

[0020] A smart training method for a dual-airway ischemic preconditioning training device, comprising the following steps:

[0021] S1: Obtain the baseline systolic pressure of the left and right arms and the initial air pressure reference values ​​of the left and right air pumps;

[0022] S2: Based on the obtained baseline systolic pressure of the left and right arms, different preset pressure increments are set according to individual differences to determine the initial target training pressure, wherein the initial target training pressure is the baseline systolic pressure of the corresponding side plus the preset pressure increment;

[0023] S3: Based on the initial target training pressure of the left and right arms respectively, the left and right air pumps are started simultaneously to increase the pressure. The sliding mode control algorithm is used to track the target training pressure of the left and right arms in real time and maintain the pressure stability during the training period.

[0024] S4: During training, the coupling coefficient is calculated by combining the blood pressure difference between the two arms with the pulse wave conduction velocity. This quantifies the cardiovascular physiological coupling relationship between the user's two arms. The blood pressure feedback data of one side of the airway is used as the correlation parameter for adjusting the pressure of the other side of the airway. This coordinates the adjustment of the current target training pressure of the left and right arms, eliminating the interference of unilateral adjustment on the other side.

[0025] Preferably, in S3, the sliding mode control algorithm is used to track the target training pressure of the left and right arms in real time and maintain pressure stability during the training period, including the following sub-steps:

[0026] Construct a dual-air-path nonlinear state equation containing coupling influence coefficients and external disturbance terms;

[0027] Based on the equation of state, the left and right sliding surfaces are constructed using proportional-differential sliding surfaces, respectively.

[0028] The exponential reaching law is used to integrate pressure convergence speed and stability, and the sliding mode reaching law is calculated by setting the reaching speed coefficient and damping coefficient.

[0029] To perform closed-loop and adjustment, at each sampling moment, after the air pump executes the control signal to adjust the speed, the next round of data acquisition will synchronously acquire the new data of air pressure and blood pressure after adjustment, and repeat the above sub-steps.

[0030] The coupling effect coefficient is used to reflect the degree of interference of unilateral pressure on the condition of the blood vessels on the other side of the patient.

[0031] Preferably, the coupling influence coefficient includes the coupling influence coefficient of the left arm on the right arm. Coupling effect coefficient of the right arm on the left arm The calculation formulas are as follows:

[0032] ;

[0033] ;

[0034] in, and These are the real-time systolic blood pressure of the left arm and the real-time systolic blood pressure of the right arm, respectively.

[0035] Preferably, both the left and right sliding surfaces are based on pressure tracking error and error change rate, and the error convergence speed is adjusted by setting a convergence coefficient.

[0036] Preferably, in S1, determining the baseline systolic blood pressure includes: using the Grubbs criterion to remove outliers from the multiple sets of bilateral blood pressure data, and taking the mean of the remaining data as the baseline systolic blood pressure of the left and right arms.

[0037] Preferably, in S2, after the initial target training pressure, a security check is performed on the calculated initial target training pressure.

[0038] Preferably, in S4, the formula for calculating the coupling coefficient is:

[0039] ;

[0040] in, and These are the average systolic blood pressure of the left and right arms during the training rest period, respectively. The pulse wave conduction velocity during training rest periods.

[0041] The beneficial effects of this invention are as follows:

[0042] The dual-airway ischemic preconditioning training device and intelligent training method provided by this invention, by quantifying and integrating the cardiovascular physiological coupling relationship between the two arms, and designing a collaborative adaptive control strategy based on this, achieves the following: 1. By using personalized initial pressure settings and highly robust sliding mode control, it overcomes airway interference and physiological coupling, controlling the pressure fluctuations of both arms during training within a threshold range, ensuring that each limb receives accurate and effective ischemic stimulation, and solving the problem of poor training results caused by inaccurate pressure control. 2. Based on population-type differentiated pressure ranges, dual safety checks before execution, and collaborative adjustments under coupled perception, it significantly reduces the potential risks to hypertensive patients and user groups with large tolerance differences. 3. Through personalized pressure settings and stable pressure tracking, and by effectively suppressing jitter through a saturation function to reduce unilateral discomfort and pressure, it achieves dynamic adjustment of control parameters and target pressure based on the user's real-time physiological feedback, enabling the device to intelligently adapt to the differentiated needs of everyone from healthy individuals to patients with cardiovascular and cerebrovascular diseases. Attached Figure Description

[0043] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0044] Figure 1 This is a schematic diagram of the system modules of the dual-airway ischemia preconditioning training device of the present invention;

[0045] Figure 2 This is a schematic diagram of the method steps of the intelligent training method of the training device of the present invention. Detailed Implementation

[0046] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0047] Please see Figure 1 This embodiment provides a dual-airway ischemia preconditioning training device, the training device including:

[0048] The data acquisition module is used to perform a standardized bi-arm time-sharing blood pressure measurement procedure before training. It acquires and processes the baseline systolic blood pressure of the left and right arms through alternating inflation and oscillometric methods. Throughout the training and rest cycles, it synchronously acquires real-time blood pressure data, pulse wave velocity (PWV) data, and real-time air pressure data in the left and right air pump lines at high frequency. The module further includes a unit for performing no-load cycles to calibrate the initial air pressure reference values ​​of each air pump.

[0049] The pressure setting module is electrically connected to the data acquisition module. It is used to match a differentiated preset pressure increment range based on the baseline systolic pressure of the left and right arms and the user's selected population type (healthy or sub-healthy). It also fine-tunes the increment value according to the inherent physiological structural differences between the left and right arms (such as different distances from the heart), thereby calculating and setting the personalized initial target training pressure for the left and right arms respectively. The initial target training pressure is the baseline systolic pressure of the corresponding side plus the preset pressure increment.

[0050] The dual-airflow control module includes a left airflow control unit and a right airflow control unit, which are independent of each other but structurally identical, and are used to drive the left and right air pumps, respectively. The dual-airflow control module incorporates a sliding mode control submodule, which is used to synchronously start the left and right air pumps to pressurize according to the initial target training pressure, and to track the initial target training pressure in real time based on a sliding mode control algorithm to maintain stable pressure in both arms during the training period. The sliding mode control submodule includes:

[0051] The model building unit is used to establish a dual-path nonlinear state model containing physiological coupling terms and disturbance terms; the physiological coupling terms are introduced by a coupling influence coefficient quantified by the real-time systolic pressure difference between the two arms, and the disturbance terms are quantified by the difference between the real-time systolic pressure and the baseline systolic pressure.

[0052] The sliding surface design unit is used to design basic sliding surfaces for the left and right arms based on pressure tracking error and error change rate, and to construct a modified sliding surface for dual-arm collaborative control through a collaborative correction coefficient.

[0053] The control law solving unit is used to derive the discrete sliding mode control law using the exponential approach law, and to suppress chattering by replacing the sign function with a saturation function. Finally, it outputs smooth and accurate left and right air pump control voltage signals to achieve fast and stable tracking of their respective target pressures during the training period.

[0054] The main control module is electrically connected to the data acquisition module, pressure setting module and dual-air path control module respectively. It is used to coordinate the timing and collaborative work of each module, control data transmission and command issuance, and realize closed-loop control of the entire training process.

[0055] The collaborative adjustment module, electrically connected to the main control module, is activated during the rest period of training. Based on the periodically collected physiological data of both arms, it calculates the cardiovascular physiological coupling coefficient of both arms through a specific formula, thereby quantifying the real-time physiological coupling intensity of the user. Subsequently, based on the coupling level (weak, medium, strong) determined by this coefficient, the blood pressure feedback data of one side is intelligently used as the correlation parameter for adjusting the target training pressure of the other side, realizing bidirectional collaborative optimization and fundamentally eliminating the chain physiological interference caused by unilateral adjustment.

[0056] The user interaction module, electrically connected to the main control module, includes a display screen, physical buttons or a touch interface, and a voice prompt unit. It receives user operation commands and displays training status, blood pressure data, countdown time, and error messages. The user interaction module supports voice broadcasting functionality to prompt operation steps, training status, and error codes. The voice prompts can be enabled or disabled via a settings switch.

[0057] The data storage and management module is electrically connected to the main control module. It is used to store user blood pressure measurement data, training history records and system settings, and supports multi-user data separation and automatic data rolling updates.

[0058] A power management module, electrically connected to the main control module, includes a built-in rechargeable battery and an external power adapter, used to power the system and manage the battery charging status. The power management module provides an alert when the voltage is low.

[0059] The mode selection and safety monitoring module, which is integrated into the main control module or electrically connected to the main control module as an independent unit, is used for:

[0060] It can receive user input and switch between multiple training modes, including single-arm training mode, dual-arm synchronous training mode, and loop training mode.

[0061] Before executing the initial target training pressure set by the pressure setting module, a double safety check is performed: first, the pressure value is constrained within the adjustment range of the training device hardware, and an additional pressure upper limit is applied to specific groups of people; then, the difference between the initial target training pressures of the two arms is calculated, and if it exceeds the preset threshold, the pressure on the high-pressure side is automatically adjusted until the difference meets the requirements.

[0062] It monitors gas leaks, pressure anomalies, and signal interference in real time, and generates corresponding error codes and processing prompts.

[0063] This embodiment of a dual-airway ischemia preconditioning training device introduces a dynamic coupling influence coefficient into sliding mode control to quantify the interference effect of applying pressure to one arm on the other, preemptively offsetting predictable mutual interference. During the rest period, a collaborative adjustment module establishes and quantifies the coupling coefficient of the fused blood pressure difference and pulse wave velocity, transforming the abstract physiological correlation into quantitative parameters that can be used for real-time control decisions. Feedback compensation coupling is performed to correct unmeasurable disturbances and model errors. Through personalized pressure settings and dual collaborative control that overcomes airway interference and physiological coupling, the pressure fluctuations of both arms during the training period can be strictly controlled within the threshold range, ensuring that the pressure applied to each arm can accurately reach the preset ischemic stimulation threshold, thereby ensuring the consistency and reliability of the training effect and avoiding unilateral pain or numbness caused by mismatched bilateral pressure. Shaking suppression reduces the discomfort caused by high-frequency pressure fluctuations.

[0064] Please see Figure 2 This embodiment provides an intelligent training method for a dual-airway ischemic preconditioning training device. The intelligent training method includes the following steps:

[0065] S1: Through the data acquisition module of the training device, the user's initial time-sharing blood pressure is measured in both arms to obtain the baseline systolic pressure of the left and right arms, and the initial air pressure reference values ​​of the left and right air pumps are recorded at the same time.

[0066] Before the training begins, the system guides the user into a standardized preparation process through a set human-computer interaction interface: prompting the user to maintain a seated posture, place both arms naturally at the level of the heart, bind the special cuffs to the left and right upper arms respectively and confirm that the airway interface is securely connected, and remain still for 5 minutes to eliminate the effects of exercise and emotional fluctuations on blood pressure.

[0067] After preparation, the data acquisition module initiates the bi-arm time-sharing blood pressure measurement process: First, the blood pressure of the left arm is measured: the air pump automatically pressurizes to exceed the estimated systolic pressure of the left arm by 30 mmHg and then depressurizes at a constant speed. The blood pressure waveform is captured by oscillometric method, and the systolic and diastolic pressures of the left arm are recorded. After a 1-minute interval, the blood pressure of the right arm is measured using the same process to avoid interference between blood flow caused by simultaneous measurement of both arms. This measurement process is repeated 3 times, and a total of 3 sets of bi-arm blood pressure data are obtained. Then, outliers are removed using the Grubbs criterion (with a significance level set to 0.05), and the mean of the remaining data is taken as the baseline systolic pressure of the left and right arms.

[0068] Simultaneously, the system initiates initial air pressure reference value calibration for the air pumps: controlling the left and right air pumps to perform no-load pressurization-depressurization cycles, sequentially pressurizing to three gradients: 160 mmHg, 180 mmHg, and 200 mmHg. After holding each gradient for 3 seconds, the pressure is uniformly depressurized to 0 mmHg, and this cycle is repeated twice. The data acquisition module records the real-time feedback value of the pressure sensor at each gradient, and takes the average of the feedback values ​​at the same gradient in the two cycles as the initial air pressure reference value for the left and right air pumps. This is used to correct errors in subsequent actual air pressure measurements, ensuring the accuracy of the air pressure data.

[0069] S2: Based on the collected baseline blood pressure data of both arms, different preset pressure increments are set according to individual differences, and fine-tuned according to the physiological differences between the left and right arms to determine the initial target training pressure of the left and right arms. The initial target training pressure is the baseline systolic pressure of the corresponding side plus the preset pressure increment.

[0070] Based on the baseline systolic blood pressure collected by S1, the system combines the physiological differences between the left and right arms (such as the left arm being closer to the heart, having a shorter blood vessel path, and slightly higher pressure tolerance than the right arm) to calculate personalized initial target training pressure:

[0071] S21: Based on the user's pre-selected population type (healthy, sub-healthy, or cardiovascular disease patients), match the corresponding preset pressure increment range: 50-60 mmHg for healthy people, 40-50 mmHg for sub-healthy people, and 30-40 mmHg for patients.

[0072] S22: Calculate the initial target training pressure by adding a preset pressure increment to the corresponding side's baseline systolic pressure, based on the initial target training pressure.

[0073] S23: Perform security verification:

[0074] First, the pressure value is limited to the hardware adjustment range of the training device (140-260 mmHg). Based on the user information obtained before training, the upper limit of the additional constraint for patients is 220 mmHg. Second, the initial pressure difference between the user's two arms is calculated. If the difference exceeds 25 mmHg, the pressure on the high-pressure side is reduced to ≤25 mmHg to avoid excessive initial pressure difference causing limb discomfort. Finally, the initial target training pressure of the left and right arms is determined and stored.

[0075] S3: Based on the initial target training pressure of the left and right arms respectively, the left and right air pumps are started simultaneously to increase the pressure, and the initial target training pressure of the left and right arms is tracked in real time to maintain the pressure stability during the training period.

[0076] The real-time tracking of the initial target training pressure of the left and right arms and the control algorithm for maintaining the pressure-free stability of the left and right arms during the training period can be an improved PID algorithm, an MPC algorithm, and a sliding mode control algorithm (SMC). When using the improved PID algorithm, the input of the improved PID algorithm includes the error between the target training pressure and the actual pressure of this airway and the blood pressure feedback data of the other airway. The main control module performs closed-loop control respectively to achieve tracking and stable maintenance of different target training pressures.

[0077] In a preferred embodiment, sliding mode control (SMC) is used to track the initial target training pressure of the left and right arms in real time, maintaining pressure stability during the training period. The SMC control implementation is based on the independent dual-airflow hardware architecture of the ischemic preconditioning training device, relying on physiological data acquisition to construct a dynamic control closed loop. Through the variable structure characteristics of sliding mode control, the effects of physiological coupling between the two arms and airflow disturbances are offset, ultimately strictly controlling the pressure fluctuations of the left and right arms within the threshold range during the training period. This adapts to both basic training (1 minute pressurization) and standard training (5 minutes pressurization) modes, while simultaneously meeting the differentiated tolerance needs of healthy, sub-healthy, and cardiovascular disease patients. The specific implementation process of the control algorithm includes the following sub-steps:

[0078] S31: Obtain input data and preprocess the data:

[0079] During the initial training, pressurization was applied for 1 minute: blood pressure and airway pressure in both arms were collected synchronously every 12 seconds, with a single sampling rate of 100Hz and continuous collection for 1.5 seconds, obtaining 150 sets of raw data to ensure that no key changes in condition were missed within the short training cycle.

[0080] In standard training, each pressurization lasts 5 minutes, with synchronous data acquisition every 8 seconds. The sampling rate for each acquisition is 100Hz, and the continuous acquisition lasts for 1.5 seconds, which is adapted to the slow changes in vascular compliance during long-term training.

[0081] Input data includes: Control data: Real-time actual air pressure of the left air pump. Real-time actual air pressure of the right air pump , k represents the current data acquisition time, and the training pressure of the left arm target. Training pressure for the right arm target ;

[0082] Physiological feedback data: Real-time systolic blood pressure in the left arm Real-time systolic pressure of the right arm ;

[0083] Equipment status data: Left air pump operating voltage Right air pump operating voltage Left airflow rate and right airflow rate ;

[0084] After the data acquisition is completed, the input data is preprocessed. First, the amplitude limiting filter is used to quickly remove pulse-type abnormal data such as instantaneous false touch of the sensor. Then, the Kalman filter is used to eliminate measurement noise, so as to obtain smooth and reliable air pressure, blood pressure and equipment status data, providing a clean input signal for subsequent calculations.

[0085] S32: Constructing a dual-air-path nonlinear state equation containing coupling influence coefficients and external disturbance terms:

[0086] S321: Define state variables for the left and right arms separately. The state variables for the left arm include real-time actual air pressure and the rate of change of air pressure, and the same applies to the right arm. Considering the physiological coupling relationship between the two arms through the cardiovascular system, a coupling influence coefficient is designed. This coefficient is quantitatively calculated from the real-time systolic pressure difference between the two arms and is used to accurately reflect the degree of interference of unilateral pressure on the state of the blood vessels on the other side. Specifically, it includes:

[0087] Left arm state variables: , ,in, The rate of change of air pressure in the left arm. It is the real-time actual air pressure (unit: mmHg) added to the left air path by the left air pump at time k, that is, the actual pressure value inside the left arm cuff during training. It is the rate of change of air pressure in the left arm airway at time k (unit: mmHg / s), which reflects the speed at which the air pressure in the left arm rises or falls and is used to characterize the dynamic trend of air pressure change.

[0088] Left arm state variables: , ,in, This is the real-time actual air pressure (unit: mmHg) applied by the right air pump to the right air passage at time k, corresponding to the actual pressure value inside the right arm cuff. The pressure change rate of the left arm airway (unit: mmHg / s) reflects the speed at which the pressure in the right arm rises or falls, and is used to characterize the dynamic trend of pressure change.

[0089] S322: Calculate the coupling effect coefficient:

[0090] Coupling effect coefficient of left arm on right arm The calculation formula is: ;

[0091] Coupling effect coefficient of the right arm on the left arm The calculation formula is: ;

[0092] S323: Blood pressure fluctuations are quantified as external disturbances. The state equation is calculated based on the difference between real-time systolic blood pressure and pre-training baseline systolic blood pressure. This quantifies the pressure in both arms' airways and its changing trends. It also incorporates equipment disturbances identified by airflow rate, such as cuff loosening or minor leaks. This includes two types of disturbances: physiological changes and equipment malfunctions, allowing the model to accurately reflect the dynamic changes during training. Specifically:

[0093] Left arm state equation: ;

[0094] ;

[0095] in, yes The derivative of , i.e., the instantaneous rate of change of air pressure in the left arm, is related to Equivalent terms are the dynamic correlation terms in the state equation;

[0096] Right arm state equation: ;

[0097] ;

[0098] in, yes The derivative of , i.e., the instantaneous rate of change of air pressure in the right arm, is related to Equivalent to the dynamic correlation terms of the state equation. and These are the control gains of the left and right air pumps, respectively, which were calibrated to 0.85 through experiments. and This is the air path damping coefficient, calibrated to 0.03. and These are the control voltages of the left and right air pumps output at time k, respectively, used to drive the left and right arm air pumps to adjust their speeds. and Let be the external perturbation terms for the left and right arms at time k, respectively. These terms are used to quantify the perturbation of the airway caused by changes in blood pressure during training. The calculation formula is as follows:

[0099] ;

[0100] ;

[0101] and These are the baseline systolic blood pressures of the left and right arms before training, respectively.

[0102] S33: Based on the state equation, construct the left and right sliding surfaces using proportional-differential sliding surfaces respectively:

[0103] S331: A basic sliding surface is designed for each of the left and right arms. The basic sliding surface is based on the pressure tracking error and the error change rate. The convergence speed of the error is adjusted by setting a convergence coefficient. A larger convergence coefficient is used for healthy people to improve training efficiency, while a smaller value is used for sub-healthy and diseased people to ensure stable pressure application.

[0104] To ensure rapid pressure tracking, the left and right sliding surfaces are constructed using proportional-differential sliding surfaces:

[0105] The sliding surface variables of the left arm at time k:

[0106] ;

[0107] in, The pressure tracking error of the left arm airway at time k (unit: mmHg) is calculated using the following formula: , It is the rate of change of the left arm barometric pressure tracking error at time k (unit: mmHg / s), calculated by the following formula:

[0108] Because the target pressure is a fixed value during the training period, its rate of change is 0.

[0109] The sliding surface variables of the right arm at time k:

[0110] ;

[0111] in, The pressure tracking error of the right arm airway at time k (unit: mmHg) is calculated using the following formula: , It is the rate of change of the left arm barometric pressure tracking error at time k (unit: mmHg / s), calculated by the following formula:

[0112] Because the target stress is a fixed value during the training period, its rate of change is 0. and These are the convergence coefficients for the left and right arms, respectively.

[0113] S332: To avoid mutual interference in the pressure adjustment of the two arms, the system introduces a cooperative correction coefficient ( 0.05) The basic sliding surface is optimized so that the adjustment of the left arm sliding surface is synchronously linked to the state of the right arm, and the same applies to the right arm, ensuring that the pressure of both arms always converges in unison, maintaining the blood pressure difference between the two arms within the normal range of 10-15 mmHg. Simultaneously, the system reserves logic for switching between single-arm and dual-arm training modes. During single-arm training, coupling terms are automatically disabled, while during dual-arm training, collaborative correction is activated to adapt to different training scenario requirements.

[0114] Corrected left arm sliding surface variable at time k and the variable of the right arm sliding surface They are respectively:

[0115] ;

[0116] ;

[0117] S34: Calculate the approach law of the left and right sliding surfaces using exponential approach:

[0118] S341: An exponential reaching law is used to integrate pressure convergence speed and stability, and reaching speed coefficients and damping coefficients are set to allow the system state to quickly approach the preset sliding surface, while suppressing oscillations during the reaching process. The calculation formulas for the reaching laws of the left and right sliding surfaces are designed as follows:

[0119] ;

[0120] To approximate the velocity coefficient, a value of 0.3-0.5 is used. This is the damping coefficient, with a value ranging from 0.1 to 0.2. It is a symbolic function;

[0121] S342: Substitute the modified cooperative sliding surface into the approach law, combine it with the dual air path state equation, and calculate the system control law in reverse, that is, the control voltage (control input) of the left and right arm air pumps. This voltage value directly corresponds to the air pump speed. Adjusting the voltage can realize the real-time increase or decrease of air pressure.

[0122] The formulas for calculating the control voltage of the left and right arm air pumps are as follows:

[0123] ;

[0124] ;

[0125] and The control voltages directly correspond to the left and right air pumps, respectively. By adjusting the voltage, the air pump speed can be changed, thereby achieving real-time adjustment of air pressure.

[0126] To address the inherent chattering problem in sliding mode control, a saturation function is employed. Instead of the traditional sign function, by setting a saturation boundary, the control signal transitions smoothly, eliminating the numbness and tingling in the arms caused by high-frequency jitter, thus improving user training comfort. The saturation function expression is:

[0127] ;

[0128] in, The saturation boundary is set to 0.02;

[0129] Closed-loop feedback and adjustment are performed: At each sampling moment, after the air pump executes the control signal to adjust the speed, the next round of data acquisition will synchronously acquire the new data of air pressure and blood pressure after adjustment, and repeat the above process of data preprocessing, updating the state model, calculating the sliding surface, calculating the control law, safety verification, and signal transmission.

[0130] Through a closed-loop mechanism of rolling iteration, the system can track the target training pressure of the left and right arms in real time, quickly counteract various sudden disturbances that occur during training, such as changes in vascular compliance, limb micro-movements, and slight airway leakage, and ensure that the pressure of the left and right arms always keeps track of their respective target values ​​and maintains pressure stability throughout the entire training cycle.

[0131] Given the close physiological coupling between the two arms through the cardiovascular system, applying pressure to one side will affect the blood pressure and vascular status of the other side in real time. Existing technical solutions ignore this coupling relationship, leading to unstable training. The sliding mode control algorithm in S3 achieves pressure tracking by directly quantifying the coupling effects such as the blood pressure difference between the two arms and the rate of change of air pressure into system control parameters through coupling term modeling, and integrating them into the state equation. At the same time, through collaborative sliding surface correction, it actively cancels the chain interference of unilateral adjustment on the other side, adapting to the training device in the core scenario that requires simultaneous training of both arms, and solving the nonlinear coupling problem that traditional algorithms cannot handle.

[0132] Furthermore, SMC's robust performance effectively resists external disturbances such as airway leakage and sudden changes in vascular condition. It also features a simple structure, fast response speed, and compatibility with the computing power requirements of the embedded hardware of the training device. By integrating real-time blood pressure measurement data, it can achieve accurate tracking of the target training pressure of both arms and improve the training accuracy of the training device.

[0133] S4: During rest periods during training, based on periodic real-time blood pressure data of both arms and actual air pressure data of the left and right air pumps, combined with the physiological coupling relationship of the cardiovascular system of both arms, the blood pressure feedback data of one side of the airway is used as the correlation parameter for adjusting the pressure of the other side of the airway, and the current target training pressure of the left and right arms is adjusted in a coordinated manner to eliminate the interference of unilateral adjustment on the other side.

[0134] The beginner training (1 minute of pressure) starts at the 30th second of the rest period, and the standard training (5 minutes of pressure) starts once at the 2nd minute and once at the 4th minute of the rest period.

[0135] After startup, the data acquisition module synchronously collects real-time systolic blood pressure, mean arterial pressure, and residual air pressure data of both arms, and takes the average value after 5 consecutive collections as the basis for adjustment.

[0136] The coupling coefficient is calculated by combining the blood pressure difference between the two arms with the pulse wave conduction velocity to quantify the cardiovascular physiological coupling relationship between the user's two arms:

[0137] ;

[0138] Where C is the coupling coefficient. and The figures represent the average systolic blood pressure (in mmHg) of the left and right arms during the training rest period. The average is the result of five consecutive measurements of the left arm's systolic blood pressure during the rest period, reflecting the vascular pressure levels of the left and right arms at rest. The absolute value of the difference in systolic blood pressure between the two arms during the resting period is used to quantify the degree of difference in vascular pressure between the two arms. The larger the difference, the stronger the correlation between the physiological states of the two arms. The pulse wave velocity (unit: m / s) during training rest is used to quantify the user's vascular elasticity. The worse the vascular elasticity (such as arteriosclerosis), the larger the PWV value. This indicator is used to reflect the synergy of the vascular status of the two arms. The baseline value of pulse wave velocity for healthy people is about 10 m / s, and the reasonable empirical threshold for the difference in systolic blood pressure between the two arms for healthy people is about 20 mmHg.

[0139] The final calculated coupling coefficient C ranges from 0 to 1. The closer C is to 1, the stronger the physiological coupling between the cardiovascular systems of both arms, and the more significant the interference of unilateral pressure adjustment on the other side, requiring further synergistic adjustment. Conversely, the closer C is to 0, the weaker the coupling, and the greater the impact of unilateral adjustment. Based on the magnitude of the coupling coefficient, the coupling relationship is categorized into three levels: weak, moderate, and strong, according to three set value ranges. Then, the blood pressure data from one side of the airway is used as the correlation parameter for the other side: if the blood pressure in the left arm is higher than that in the right arm and the coupling is moderate, the difference between the real-time systolic blood pressure in the left arm and the baseline value is used as the basis for lowering the left arm blood pressure. The target pressure for the first arm in the next cycle is 5 mmHg, and the target pressure for the right arm is simultaneously lowered by 2.5 mmHg (to avoid fluctuations in blood pressure on the other side caused by unilateral adjustment). If the coupling is "strong", the pressure on the systolic side is adjusted step by step, and the convergence coefficient of the sliding mode control in S3 is corrected simultaneously to enhance bilateral coordination. After the adjustment is completed, the system predicts the changes in blood pressure in both arms in the next training cycle through the LSTM model. If the predicted pressure difference exceeds 15 mmHg, the adjustment amount is corrected a second time, and finally the new target training pressure is transmitted to the system control terminal to ensure that the pressure in both arms is in a coordinated and stable state and to eliminate the chain interference caused by unilateral adjustment.

[0140] The intelligent training method presented in this embodiment quantifies the cardiovascular physiological coupling of both arms in real time (by fusing the coupling coefficient C of blood pressure difference and pulse wave velocity, as well as the coupling influence coefficient), and deeply integrates it into two key aspects of intelligent control: First, during the training period, the dynamically calculated coupling influence coefficient is embedded as an internal state term into the feedforward nonlinear state equation of sliding mode control to achieve pre-compensation for disturbances; second, during the rest period, based on the quantified coupling strength level, the blood pressure feedback on one side is used as a correlation parameter for the pressure adjustment on the other side to achieve feedforward collaborative optimization of the target.

[0141] By overcoming physiological coupling and equipment interference, the pressure fluctuations in both arms during training can be controlled within a very narrow range of thresholds, ensuring that each limb receives precise and effective ischemic stimulation. In terms of safety, the differentiated pressure settings based on population type, real-time dual safety verification, and coordinated adjustment under coupled perception jointly construct an active protection system, which significantly reduces the training risks for users (especially cardiovascular and cerebrovascular patients).

[0142] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A dual-airway ischemic preconditioning training device, characterized in that: include: The data acquisition module is used to obtain the user's baseline systolic blood pressure in both arms and the initial air pressure reference values ​​of the left and right air pumps. It also collects real-time blood pressure data, pulse wave conduction velocity data, and real-time air pressure data of the left and right air pumps in both arms during training and rest periods. The pressure setting module is electrically connected to the data acquisition module and is used to set different preset pressure increments based on the acquired baseline systolic pressure of the left and right arms according to individual differences. The initial target training pressure of the left and right arms is the baseline systolic pressure of the corresponding side plus the preset pressure increment. The dual-air-path control module includes a left air-path control unit and a right air-path control unit that are independent of each other but have the same structure, which are used to drive the left air pump and the right air pump respectively. The dual-air-path control module has a built-in sliding mode control sub-module, which is used to synchronously start the left and right air pumps to pressurize according to the initial target training pressure, and to track the initial target training pressure in real time based on the sliding mode control algorithm in order to maintain the stability of the pressure of the two arms during the training period. The main control module is electrically connected to the data acquisition module, pressure setting module and dual-air path control module respectively. It is used to coordinate the timing and collaborative work of each module, control data transmission and command issuance, and realize closed-loop control of the entire training process. The collaborative adjustment module, electrically connected to the main control module, is used to calculate the coupling coefficient by combining the blood pressure difference between the two arms with the pulse wave conduction velocity during the rest period of training, and to use the blood pressure feedback data of one side of the airway as a correlation parameter to collaboratively adjust the current target training pressure of the left and right arms to eliminate the interference of unilateral pressure adjustment on the other side. The coupling coefficient is used to quantify the cardiovascular physiological coupling relationship between the user's two arms.

2. The dual-airway ischemic preconditioning training device according to claim 1, characterized in that: The sliding mode control submodule includes: The model building unit is used to establish a dual-path nonlinear state model containing physiological coupling terms and disturbance terms; the physiological coupling terms are introduced by a coupling influence coefficient quantified by the real-time systolic pressure difference between the two arms, and the disturbance terms are quantified by the difference between the real-time systolic pressure and the baseline systolic pressure. The sliding surface design unit is used to design basic sliding surfaces for the left and right arms based on pressure tracking error and error change rate, and to construct a modified sliding surface for dual-arm collaborative control through a collaborative correction coefficient. The control law solving unit is used to derive the discrete sliding mode control law using the exponential approach law, and to suppress chattering by replacing the sign function with a saturation function, and outputs control voltage signals to drive the left and right air pumps respectively.

3. The dual-airway ischemic preconditioning training device according to claim 1, characterized in that... It also includes a mode selection and safety monitoring module electrically connected to the main control module, used for: It can receive user input and switch between multiple training modes, including single-arm training mode, dual-arm synchronous training mode, and loop training mode. Before executing the initial target training pressure set by the pressure setting module, a double safety check is performed: first, the pressure value is constrained within the adjustment range of the training device hardware, and an additional pressure upper limit is applied to specific groups of people. Then, the difference between the initial target training pressures of the two arms is calculated. If it exceeds the preset threshold, the pressure on the high-pressure side is automatically adjusted until the difference meets the requirements. It monitors gas leaks, pressure anomalies, and signal interference in real time, and generates corresponding error codes and processing prompts.

Images