A method, system, device and storage medium for maintaining pulmonary hypertension exercise intervention regulation in maintenance dialysis

Abstract

The present application belongs to the technical field of medical rehabilitation training and intelligent monitoring and control, and particularly relates to a maintenance dialysis pulmonary hypertension exercise intervention closed-loop regulation method, system, device and storage medium. The method collects dialysis recovery data, pulmonary circulation tolerance data, exercise tolerance data and historical training response data, constructs a dialysis recovery safety boundary, a pulmonary circulation tolerance boundary and an exercise prescription boundary, and fuses to form a current training allowable domain; during training, a comprehensive deviation index is calculated based on a rolling time window, and hierarchical fallback regulation is performed; after training, the next training cycle boundary is updated according to the training completion degree, recovery standard reaching degree and abnormal event penalty value. The scheme can improve the safety, completion degree and individual adaptability of exercise intervention.

Description

Technical Field

[0001] This invention belongs to the field of medical rehabilitation training and intelligent monitoring and control technology, specifically relating to a method, system, device and storage medium for exercise intervention and regulation of pulmonary hypertension during maintenance dialysis. Background Technology

[0002] Patients undergoing maintenance hemodialysis for chronic renal failure are more prone to problems such as elevated pulmonary artery pressure, limited right ventricular function, decreased exercise tolerance, and reduced quality of life due to long-term volume load fluctuations, anemia, arteriovenous fistula shunting, decreased cardiopulmonary reserve, and multiple complications. Clinical research and rehabilitation practice have shown that appropriate aerobic exercise intervention can help improve the physical activity level and some cardiopulmonary function indicators of dialysis patients. However, dialysis patients are characterized by large fluctuations in physiological state, narrow training safety boundaries, and significant individual differences. Therefore, exercise intervention cannot simply adopt the fixed exercise prescriptions for ordinary chronic disease populations, but needs to be targeted and adjusted in a way that takes into account the dialysis recovery status, cardiopulmonary tolerance, and the risks of the training process.

[0003] In existing technologies, there has been some research on exercise prescriptions and training monitoring. For example, Chinese patent CN109308940A discloses an exercise guidance method based on cardiopulmonary exercise assessment. Its overall approach includes assessing the user's cardiopulmonary function, risk stratification, developing personalized exercise prescriptions, monitoring the body's condition during training, and taking appropriate action when abnormalities occur. This technology has certain applicability to general exercise rehabilitation scenarios, enabling basic management of exercise intensity, training duration, and training process risks. Furthermore, in the area of ​​dialysis safety monitoring, Chinese patent CN113289098B discloses a dialysate monitoring method and device based on a dialysis risk assessment model. This technology acquires user dialysis-related information and performs risk assessment. When the dialysis risk exceeds a preset threshold, it outputs an early warning command and prompts medical staff to reduce the dialysis risk by adjusting dialysate temperature, flow rate, or electrolyte composition. This technology primarily focuses on the safety control of the dialysis process itself and the adjustment of dialysate parameters.

[0004] However, the aforementioned existing technologies still have at least the following shortcomings: First, CN109308940A focuses more on exercise assessment, risk stratification, and prescription formulation in the general population or common scenarios. Although it involves training monitoring and abnormality handling, it does not establish specific control logic for maintenance dialysis patients with pulmonary hypertension. In particular, it does not treat post-dialysis recovery status and pulmonary circulation tolerance status as two independent but coupled constraint sources to jointly limit the allowable training range. Therefore, it is difficult to fully adapt to the dynamic training needs of dialysis patients in different dialysis cycles, different volume states, and different cardiopulmonary states. Second, although CN113289098B can assess dialysis risk and adjust dialysate parameters, its control object is still the dialysis treatment process and dialysate configuration parameters. Essentially, it belongs to dialysis process monitoring technology and does not extend to the post-dialysis exercise rehabilitation scenario. It also does not solve the problem of how to dynamically adjust the exercise intervention boundary for the next cycle based on the degree of recovery after dialysis, real-time physiological deviations during training, and post-training recovery feedback. Third, existing research on exercise rehabilitation for dialysis patients mainly focuses on verifying the effectiveness of exercise or the feasibility of a certain type of exercise program. There are few closed-loop technical solutions that combine real-time monitoring, risk classification, dynamic regression adjustment, and adaptive updates across training cycles. This leads to problems in practical applications, such as overly coarse training load settings, delayed abnormality identification, and insufficient prescription continuity between different dialysis cycles. It is difficult to balance safety, specificity, and long-term stable implementation.

[0005] Therefore, it is still necessary to provide a closed-loop regulation technology for exercise intervention for maintenance dialysis patients with pulmonary hypertension, so as to achieve pre-training boundary generation, dynamic regulation during training, and post-training cross-cycle update under the dual constraints of dialysis recovery state and pulmonary circulation tolerance state. Summary of the Invention

[0006] The technical objective of this invention is to address the problems of significant individual differences, substantial post-dialysis status fluctuations, difficulty in accurately determining training safety boundaries, and the lack of dynamic regulation in existing exercise intervention programs for maintenance dialysis patients with pulmonary hypertension. This invention provides a method, system, device, and storage medium for regulating exercise intervention in maintenance dialysis patients with pulmonary hypertension. Under the dual constraints of dialysis recovery and pulmonary circulation tolerance, it enables dynamic generation of exercise prescriptions, real-time monitoring and graded adjustment of the training process, and post-training cross-cycle updates, thereby improving the safety, relevance, and continuous effectiveness of exercise intervention. Firstly, to achieve the above objective, this invention adopts the following technical solution:

[0007] A method for regulating pulmonary hypertension during maintenance dialysis through exercise intervention, comprising the following steps:

[0008] S1. Collect baseline status data of the target patient within a preset recovery time window after a hemodialysis session. The baseline status data includes at least dialysis recovery data, pulmonary circulation tolerance data, exercise tolerance data, and historical training response data.

[0009] S2. Based on the aforementioned basic state data, construct the dialysis recovery safety boundary, the pulmonary circulation tolerance boundary, and the exercise prescription boundary respectively, and fuse them to obtain the allowable domain for the current training session;

[0010] S3. During the exercise intervention for the target patient, real-time training data is continuously collected. Trend analysis is performed on the real-time training data based on a rolling time window, and a comprehensive deviation index is calculated. To determine the current training status;

[0011] S4. Compare the current training state with the allowed range for the current training session. When the current training state is within the compensation range, perform one or more levels of load rollback adjustment. When the current training state continues to reach the termination range, terminate the current training session and generate the next training order reduction constraint.

[0012] S5. After this training session, collect recovery period response data and abnormal event records, and calculate the training completion rate. and recovery correction factor And based on the training completion rate and recovery correction factor Update the initial control boundary for the next training iteration and repeat steps S1 through S5.

[0013] Preferably, in step S1, the dialysis recovery data includes at least the ultrafiltration volume. , weight before dialysis Post-dialysis weight Number of hypotension events during dialysis and the duration of symptom recovery after dialysis The change in body weight before and after dialysis is calculated using the following formula:

[0014] ;

[0015] In the formula, This represents the change in body weight before and after dialysis. Weight before dialysis; Weight after dialysis;

[0016] And / or, in step S1, the pulmonary circulation tolerance data includes at least pulmonary artery systolic pressure. Right ventricular function indicators resting heart rate Resting blood oxygen saturation and resting systolic blood pressure ;

[0017] And / or, in step S1, the exercise tolerance data includes at least peak oxygen uptake. 6-minute walk and subjective fatigue rating ;

[0018] And / or, in step S1, the historical training response data includes at least the previous training completion rate. Recovery time after previous training and the level of previous abnormal events ;

[0019] And / or, in step S1, the preset recovery window is the [number]th [time] after dialysis ends. Hours to the The time interval between hours, To allow the earliest possible start time for training, The latest allowed start time for training, and ;

[0020] And / or, in step S1, after constructing the basic status data, a training suitability screening is performed to exclude target patients who have active cardiovascular instability, acute respiratory failure, severe hypotension, or who are clinically deemed unsuitable for exercise intervention.

[0021] Preferably, in step S2, the dialysis recovery safety boundary is used to limit the maximum allowable training load, maximum continuous exercise duration, and minimum recovery period duration for the current training session; the pulmonary circulation tolerance boundary is used to limit the allowable heart rate deviation range, blood pressure fluctuation range, lower limit of blood oxygenation, and upper limit of symptom score for the current training session; the exercise prescription boundary is used to limit the target intensity, target duration, and target rhythm for the current training session; and the allowable domain for the current training session is a common control domain that simultaneously satisfies the dialysis recovery safety boundary, the pulmonary circulation tolerance boundary, and the exercise prescription boundary.

[0022] And / or, in step S2, the target load for the current training session is calculated using the following formula:

[0023] ;

[0024] In the formula, The target load for this training session; This is the prescription conversion factor; Peak oxygen uptake; This is a safety correction factor determined based on dialysis recovery status and pulmonary circulation tolerance status;

[0025] And / or, in step S2, the dialysis recovery safety boundary is generated based on the dialysis recovery risk level, which is divided into at least three levels: low risk, medium risk, and high risk; the higher the risk level, the lower the upper limit of the intensity allowed for the current training, the shorter the duration of the continuous exercise segment, and the longer the duration of the recovery segment;

[0026] And / or, in step S2, the pulmonary circulation tolerance boundary is determined by... , , , and It was jointly determined that, The systolic blood pressure of the pulmonary artery. This is an indicator of right ventricular function. Resting heart rate Resting blood oxygen saturation This is the resting systolic blood pressure.

[0027] Preferably, in step S3, the real-time training data includes at least real-time heart rate. Real-time systolic blood pressure Real-time blood oxygen saturation Real-time symptom score Real-time exercise load and real-time cadence ;

[0028] The comprehensive deviation index is calculated using the following formula:

[0029] In the formula, For a moment The overall deviation index; , , , , Let be the weight coefficients of each deviation term, and satisfy .

[0030] ;

[0031] in, This refers to the deviation from the real-time heart rate. This refers to the real-time blood oxygen deviation term; This refers to the deviation of real-time blood pressure. This refers to items that deviate from the symptom score; This is the load deviation item;

[0032] And / or, in step S3, the scrolling time window has a length of The continuously updated analysis interval, where, The preset time length is used; the comprehensive deviation index is updated once after each sampling step. The current training state is determined based on the direction of change, degree of deviation, and duration of multiple consecutive sampling points within the rolling time window.

[0033] Preferably, in step S3, the real-time heart rate deviation term is calculated using the following formula:

[0034] In the formula, For a moment Real-time heart rate deviation; For a moment Real-time heart rate; Target heart rate; This represents the upper limit of the allowable deviation in heart rate.

[0035] And / or, in step S3, the real-time blood oxygen deviation term is calculated using the following formula:

[0036] ;

[0037] In the formula, For a moment Real-time blood oxygen deviation; This is the lower limit of blood oxygenation. For a moment Real-time blood oxygen saturation;

[0038] And / or, in step S3, the real-time blood pressure deviation term is calculated using the following formula:

[0039] ;

[0040] In the formula, For a moment Real-time blood pressure deviation; For a moment Real-time systolic blood pressure; The target systolic blood pressure reference value; This is the upper limit of the allowable deviation of systolic pressure;

[0041] And / or, in step S3, the symptom score deviation item is calculated using the following formula:

[0042] ;

[0043] In the formula, For a moment Symptom score deviation items; For a moment Subjective symptom scores; This represents the upper limit of the symptom score.

[0044] And / or, in step S3, the load deviation term is calculated using the following formula:

[0045] ;

[0046] In the formula, For a moment The load deviation term; For a moment Real-time motion load; The target motion load; This represents the upper limit of the allowable load offset.

[0047] Preferably, in step S4, the one or more levels of load reduction adjustment include at least one or more of the following: reducing resistance, reducing target cadence, shortening the duration of the continuous exercise segment, extending the duration of the recovery segment, and switching to a recovery training mode; the training termination determination condition satisfies:

[0048] ;

[0049] In the formula, The duration of the unsafe state; The threshold for the duration of training termination;

[0050] And / or, in step S4, the backoff adjustment corresponding to the compensation interval is a three-level graded adjustment: when When entering the first-level compensation zone, reduce motion resistance; when When entering the secondary compensation zone, reduce the target cadence and shorten the duration of the continuous movement segment; when When entering the third-level compensation range, switch to the recovery segment;

[0051] And / or, in step S4, the next training reduction constraint includes at least one or more of the following: reducing the upper limit of the starting intensity of the next training, reducing the upper limit of the continuous exercise duration, and increasing the proportion of the recovery segment.

[0052] Preferably, in step S5, the training completion rate is calculated using the following formula:

[0053] ;

[0054] In the formula, Training completion rate; This represents the actual amount of training completed in this session; This is the planned training amount; the recovery correction coefficient is calculated using the following formula:

[0055] ;

[0056] In the formula, To restore the correction factor; , , To adjust the weighting coefficients; To restore compliance; This is the penalty value for abnormal events;

[0057] In step S5, the recovery achievement rate The degree of recovery is determined by the extent of heart rate recovery, blood oxygen recovery, and blood pressure recovery during the recovery period; when When, perform a downgraded update for the next training session; when When, keep the training boundaries unchanged for the next time; when At that time, a graded update is performed for the next training session; where, This is the order reduction threshold; This is the order-up threshold, and .

[0058] In step S5, the abnormal event record includes at least one of the following: chest tightness event, shortness of breath event, hypotension event, abnormal heart rate event, and training termination event.

[0059] Secondly, the present invention also provides a closed-loop control system for maintenance dialysis pulmonary hypertension exercise intervention. The system is used to implement the method and includes: a basic state acquisition module for executing step S1; a dual-boundary construction and prescription generation module for executing step S2; a rolling monitoring and comprehensive deviation assessment module for executing step S3; a hierarchical closed-loop control module for executing step S4; and a cross-cycle write-back update module for executing step S5.

[0060] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the method described above.

[0061] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described thereon.

[0062] The technical advantages of this invention are as follows: By collecting dialysis recovery data, pulmonary circulation tolerance data, exercise tolerance data, and historical training response data before training, a safety boundary for dialysis recovery, a pulmonary circulation tolerance boundary, and an exercise prescription boundary are constructed and fused to form the allowable domain for the current training session. This allows the initial training load to be adaptively determined based on the patient's actual condition after the current dialysis session, rather than relying on fixed experience settings. Furthermore, by continuously collecting real-time training data such as heart rate, blood pressure, blood oxygen, symptom scores, and exercise load during training, and using a rolling time window and a comprehensive deviation index to determine the trend of the training status, misjudgments or omissions caused by single-point anomalies can be avoided. This invention improves the timeliness and stability of abnormality identification. By setting compensation intervals, termination intervals, and multi-level load reduction adjustment chains, it prioritizes gradual adjustment measures such as reducing resistance, reducing cadence, shortening exercise segments, extending recovery segments, or switching recovery modes when risks gradually increase, rather than directly terminating training, thus balancing training safety and training completion. Furthermore, by introducing training completion, recovery achievement, and abnormal event penalty values ​​after training, the initial control boundary for the next training cycle is rewritten and corrected, forming a closed-loop learning and adaptive optimization mechanism across training cycles, making subsequent prescriptions more aligned with the patient's staged recovery level. Therefore, this invention can effectively improve the safety, controllability, individual adaptability, and long-term stability of exercise intervention for maintenance dialysis patients with pulmonary hypertension. Attached Figure Description

[0063] Figure 1 This is a schematic diagram of the overall process of a closed-loop regulation method for exercise intervention in maintenance dialysis pulmonary hypertension according to the present invention.

[0064] Figure 2 This is a schematic diagram of the structure of a closed-loop regulation system for exercise intervention in pulmonary hypertension during maintenance dialysis, according to the present invention.

[0065] Figure 3 This is a schematic diagram illustrating the construction of dual boundaries and the formation of the allowed domain during the current training in this invention.

[0066] Figure 4 This is a schematic diagram of the rolling monitoring and hierarchical closed-loop control during the training process in this invention.

[0067] Figure 5 This is a schematic diagram of post-training recovery evaluation and cross-cycle write-back update in this invention.

[0068] Figure 6 This is a schematic diagram of the hardware structure of an electronic device according to the present invention.

[0069] Figure 7 This is a cross-cycle update trend diagram of the recovery correction coefficient and the target load for the next cycle in an embodiment of the present invention.

[0070] Figure 8This is a comparison chart of the changes in rehabilitation indicators of the embodiment group and the comparative group of the present invention over 12 weeks.

[0071] Figure 9 The comprehensive deviation index for a single training process in this embodiment of the invention. Diagram showing the trigger point for hierarchical rollback. Detailed Implementation

[0072] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. These specific embodiments include explanations of terminology, system structure, and specific technical routes for implementing the method of the present invention, and will be combined with… Figures 1 to 6 The implementation process of this invention is fully disclosed. It should be noted that the following embodiments are for illustrative purposes only and are not intended to limit the scope of protection of this invention. Any equivalent substitutions, parameter optimizations, module splitting or merging, hardware replacements, or software deployment method adjustments made by those skilled in the art based on this specification without departing from the concept and technical essence of this invention should fall within the scope of protection of this invention.

[0073] This invention is applicable to target patients undergoing maintenance hemodialysis who exhibit signs of pulmonary hypertension. Preferably, the target patients are those who regularly receive hemodialysis treatment, have relatively stable conditions, require exercise rehabilitation, and have no obvious contraindications. Application scenarios for this invention include dialysis center rehabilitation rooms, hospital rehabilitation departments, integrated dialysis rehabilitation platforms, and distributed systems linking bedside monitoring and training terminals.

[0074] I. Terminology Explanation

[0075] To facilitate understanding of this invention, the main terms used in this specification will be explained first.

[0076] 1. Maintenance dialysis refers to a stage in which the target patient is in regular hemodialysis treatment and has formed a relatively stable dialysis cycle.

[0077] 2. Pulmonary hypertension refers to the elevated pulmonary artery pressure in the target patient. This invention does not limit the detection method to a single method, as long as parameters reflecting the level of pulmonary circulation load can be obtained, preferably pulmonary artery systolic pressure and right ventricular function-related parameters obtained by color Doppler ultrasound detection.

[0078] 3. Dialysis recovery data refers to a set of data reflecting changes in volume, circulatory stability, and symptom recovery after a single dialysis session. Preferred data include pre-dialysis weight, post-dialysis weight, ultrafiltration volume, number of hypotensive events during dialysis, and duration of symptom recovery after dialysis.

[0079] 4. Pulmonary circulatory tolerance data refers to a set of data reflecting the patient's current pulmonary circulatory load, right ventricular function, and oxygenation capacity. Preferably, this includes pulmonary artery systolic blood pressure. Right ventricular function indicators resting heart rate Resting systolic blood pressure and resting blood oxygen saturation .

[0080] 5. Exercise tolerance data refers to a set of data reflecting the patient's current tolerable training intensity, duration, and rhythm. Peak oxygen uptake is preferably included. 6-minute walk and subjective fatigue rating .

[0081] 6. Historical training response data refers to a dataset derived from previous training cycles that reflects the patient's training completion status, post-training recovery, and any abnormal events. Preferably, it includes previous training completion rates. Recovery time after previous training and the level of previous abnormal events .

[0082] 7. The dialysis recovery safety boundary refers to the boundary formed by constraining the maximum training load, the upper limit of continuous exercise duration, and the lower limit of the recovery segment based on the dialysis recovery status.

[0083] 8. The pulmonary circulation tolerance boundary refers to the boundary formed by constraining heart rate deviation, blood pressure fluctuation, lower limit of blood oxygen, and upper limit of symptom score during training, based on the pulmonary circulation and right heart tolerance status.

[0084] 9. Exercise prescription boundaries refer to the boundaries of target training intensity, target duration, and target rhythm based on the patient's exercise tolerance.

[0085] 10. The training allowable range refers to the range of feasible training parameters that simultaneously meet the safety boundaries of dialysis recovery, the pulmonary circulation tolerance boundaries, and the exercise prescription boundaries.

[0086] 11. The comprehensive deviation index refers to a unified risk representation formed by normalizing and weighting multiple deviations during the training process.

[0087] 12. Graded regression adjustment refers to a progressive adjustment method in which measures such as reducing resistance, reducing cadence, shortening continuous movement segments, extending recovery segments, or switching recovery modes are taken in sequence as training risks gradually increase.

[0088] 13. The recovery correction coefficient is a correction amount formed by training completion, recovery achievement, and abnormal event penalty value, used to update the control boundary of the next training cycle.

[0089] 14. Cross-training cycle closed loop refers to the result after the end of this training session continuing to participate in the next training boundary reconstruction, thereby achieving continuous optimization between training cycles.

[0090] II. System Structure

[0091] See Figure 2 The present invention discloses a closed-loop control system for maintenance dialysis pulmonary hypertension exercise intervention, which includes at least a baseline state acquisition module, a dual-boundary construction and prescription generation module, a rolling monitoring and comprehensive deviation assessment module, a hierarchical closed-loop control module, a cross-cycle write-back update module, a training execution terminal, a monitoring terminal, a data storage unit, and a communication interface.

[0092] The baseline status acquisition module is used to collect dialysis recovery data, pulmonary circulation tolerance data, exercise tolerance data, and historical training response data of the target patient before training begins. This module can be connected to dialysis equipment, electronic medical record systems, ultrasound detection equipment, cardiopulmonary exercise testing equipment, wearable monitoring terminals, and manual data entry terminals to obtain the multi-source status information required before training.

[0093] The dual-boundary construction and prescription generation module is used to generate dialysis recovery safety boundaries, pulmonary circulation tolerance boundaries, and exercise prescription boundaries based on baseline state data, and then merges these three to form the allowable domain for the current training session. This module is one of the core modules of the intelligent pre-training modeling of this invention.

[0094] The rolling monitoring and comprehensive deviation assessment module continuously collects real-time data such as heart rate, systolic blood pressure, blood oxygen saturation, symptom score, real-time load, and cadence during training. Based on a rolling time window, it performs trend analysis on the real-time data to generate a comprehensive deviation index. .

[0095] The hierarchical closed-loop control module is used to... The relationship between the training current allowed domain and the current training state determines whether to stabilize, compensate, or terminate the current training state; when in the compensation range, it performs one or more levels of load rollback adjustment; when the termination condition is met, it outputs a training termination command and generates a reduced-order constraint for the next training cycle.

[0096] The cross-cycle write-back update module is used to collect recovery period response data and abnormal event records after training, and to calculate the training completion rate. and recovery correction factor And update the starting boundary for the next training cycle.

[0097] The training execution terminal is preferably a power bike training device, but it can also be a walking trainer, elliptical trainer, or other aerobic training equipment suitable for dialysis patients. The monitoring terminal preferably includes a heart rate monitor, a non-invasive blood pressure module, a pulse oximeter, a symptom entry terminal, and an exercise parameter acquisition device. The data storage unit stores raw monitoring data, pre-training parameters, in-training status data, post-training recovery data, and model parameters. The communication interface is used for data transmission between modules and between the system and external devices.

[0098] III. Overall Technical Route for Implementing the Method of the Invention

[0099] See Figure 1 The method of the present invention includes the following steps:

[0100] Basic status data collection and training suitability screening;

[0101] Dual-boundary construction and current prescription generation;

[0102] Rolling monitoring and comprehensive deviation assessment;

[0103] Hierarchical closed-loop regulation and training termination determination;

[0104] Cross-cycle write-back update.

[0105] (a) Steps Specific implementation method

[0106] step The main task is to establish a basic state set before training and to perform training suitability screening.

[0107] In a preferred embodiment, the system initiates a pre-training assessment within a preset recovery time window after the target patient completes a dialysis treatment. The preset recovery time window can be defined as the [number]th [period] after the end of dialysis. Hours to the The time interval between hours, To allow the earliest possible start time for training, The latest allowed start time for training, and .

[0108] 1. Dialysis recovery data collection

[0109] The system preferably obtains predialysis weight from the dialysis device, dialysis record sheet, or electronic medical record system. Post-dialysis weight Ultrafiltration volume Number of hypotension events during dialysis and the duration of symptom recovery after dialysis .

[0110] The change in body weight before and after dialysis is calculated using the following formula:

[0111] ;

[0112] In the formula, This represents the change in body weight before and after dialysis. Weight before dialysis; Weight after dialysis.

[0113] and Together they can reflect the degree of volume clearance in this dialysis session. This reflects the circulatory vulnerability during dialysis. It reflects the recovery rate after dialysis. When Larger higher or A prolonged duration indicates a weaker recovery status after dialysis, and subsequent training boundaries should be conservative.

[0114] 2. Pulmonary Circulation Tolerance Data Acquisition

[0115] The system collects pulmonary artery systolic blood pressure. Right ventricular function indicators resting heart rate Resting systolic blood pressure and resting blood oxygen saturation Preferably, and Obtained by color Doppler ultrasound detection. , and Data obtained from pre-training resting state monitoring.

[0116] 3. Exercise tolerance data collection

[0117] The system collects peak oxygen uptake. 6-minute walk and subjective fatigue rating Peak oxygen uptake is preferably obtained through cardiopulmonary exercise testing, 6-minute walking distance is preferably obtained through standard 6-minute walking test, and fatigue score can be obtained through scales or manual input.

[0118] 4. Historical training response data collection

[0119] The system retrieves past training completion rates from historical training archives. Recovery time after previous training and previous abnormal event levels This is one of the key features that distinguishes this invention from traditional approaches that treat training cycles as isolated from each other.

[0120] 5. Suitability screening

[0121] After establishing the baseline state set, the system performs a training suitability screening. If the target patient exhibits active cardiovascular instability, acute respiratory failure, severe hypotension, or is clinically deemed unsuitable for exercise training, the system directly outputs an instruction to not initiate training or postpone training, without proceeding to the next step. .

[0122] (II) Steps Specific implementation method

[0123] See Figure 3 This invention does not employ the traditional method... Instead of directly deriving a single path for training intensity, the approach first establishes separate safety boundaries for dialysis recovery and pulmonary circulation tolerance boundaries, then combines these with the exercise prescription boundaries to form the permissible domain for that training session. This step not only determines the quality of pre-training control but also whether the entire closed-loop system truly establishes a differentiated, safe, and sustainable control logic specifically for the particular population of maintenance dialysis patients with pulmonary hypertension.

[0124] Many existing exercise rehabilitation programs are based on the patient's physical fitness, using indicators such as peak oxygen uptake, walking ability, or maximum heart rate as the direct basis for training prescriptions. While these programs may be feasible in common chronic disease or general rehabilitation scenarios, relying solely on exercise tolerance is insufficient for maintenance dialysis patients with pulmonary hypertension. This is because the condition of these patients is not only affected by physical fitness factors but also significantly influenced by the volume clearance during this dialysis session, circulatory fluctuations, right ventricular load, and resting oxygenation status. In other words, even for the same patient, different dialysis cycles, different ultrafiltration volumes, and different pulmonary circulation states, the results will vary. Even without significant changes, the training intensity and rhythm that can be safely executed in a given session may differ significantly. Therefore, this invention introduces a three-boundary fusion mechanism of dual boundaries + prescription boundaries before training, fundamentally changing the traditional approach to generating fixed or single-dimensional prescriptions.

[0125] 1. Analysis of the construction logic for restoring security boundaries

[0126] The dialysis recovery safety boundary is used to limit the maximum allowable training load, maximum continuous exercise duration, and minimum recovery period for a given training session. Essentially, it reflects how much training safety margin the patient has remaining after the current dialysis session, considering their volume status, circulatory stability, and symptom recovery.

[0127] In a preferred embodiment, the system according to , , and The recovery status of dialysis is graded. Here, Reflects the total amount of fluid removed during dialysis; Reflects the magnitude of changes in body fluids before and after dialysis; It reflects the occurrence of hypotension during dialysis; It reflects the subjective and objective recovery time after dialysis. The dialysis recovery status can be preferentially divided into three levels: low risk, medium risk, and high risk, or it can be further refined into five levels or a continuous scoring method according to actual needs.

[0128] when Moderate Not big and When the time is short, the system determines that the patient's dialysis recovery is sufficient, at which point a higher initial training load, a longer continuous exercise period, and a shorter recovery time can be given.

[0129] when Larger Large or When the duration of the exercise is prolonged, the system determines that the patient is in a state of moderate recovery risk. At this time, the initial load should be appropriately reduced, the continuous exercise segment should be shortened, and the proportion of the recovery segment should be increased.

[0130] When multiple episodes of hypotension occur during dialysis, symptom recovery is significantly delayed after dialysis, or significant fatigue, dizziness, or chest tightness persists, the system can determine that the condition is at high risk of recovery. In this case, only low-load short-interval training is allowed, and if necessary, only restorative pedaling is allowed, without entering the target load training phase.

[0131] To incorporate this boundary into subsequent prescription calculations, this invention preferably quantifies the dialysis recovery safety boundary as a dialysis recovery safety factor. . The range of values ​​can be set to arrive Preferably, in a low-risk state Pick to Take when in medium-risk state to Take in high-risk situations to In a more refined implementation, it is also possible to... , , and The method of assigning weights separately directly yields continuous results. Numerical value.

[0132] For example, the following scoring logic can be used: If If the preset threshold is exceeded, the recovery safety score will be lowered; if If it is greater than 0, then adjust it further downwards; if If the preset duration is exceeded, the time will be reduced again. This results in... It can directly participate in prescription generation. This design allows the system to know not only whether training is possible after this dialysis session, but also the maximum level of training allowed after this dialysis session.

[0133] 2. The construction logic of the pulmonary circulation tolerance boundary

[0134] The pulmonary circulatory tolerance boundary is used to define the range of heart rate deviation, blood pressure fluctuation, lower limit of blood oxygenation, and upper limit of symptom score during training. It reflects the patient's current pulmonary circulation and right ventricular system's tolerance to training stimuli.

[0135] Unlike typical exercise rehabilitation patients, one of the main risks of training for maintenance dialysis patients with pulmonary hypertension stems not only from insufficient physical fitness, but also from the inadequate tolerance of the pulmonary circulation and right heart to exercise stimuli. For example, some patients... and It might be okay, but if higher Poor or resting state If the values ​​are too low, problems such as shortness of breath, hypoxia, and increased right ventricular load are more likely to occur during training. Therefore, this invention does not treat these parameters merely as post-training evaluation indicators, but rather moves them forward to be direct inputs to the pre-training control boundaries.

[0136] In a preferred embodiment, the system is based on , , , and Generate pulmonary circulation safety factor The preferred selection rules are as follows:

[0137] when Lower better, When hemodynamics are normal and at rest, Values ​​close to ;

[0138] when rise, decline, When heart rate decreases or resting heart rate fluctuates significantly. Accordingly reduced;

[0139] When a patient already exhibits obvious symptoms of chest tightness, shortness of breath, hypoxia, or right ventricular dysfunction at rest... A lower value should be chosen to significantly tighten the training boundary.

[0140] Preferably, in a low-risk state Pick to Take when in medium-risk state to Take in high-risk situations to This invention does not limit specific scores, as long as they can reflect the relationship between poorer pulmonary circulation and right ventricular tolerance levels and more conservative training boundaries.

[0141] 3. Logic for generating exercise prescription boundaries

[0142] The boundaries of exercise prescriptions are mainly based on , and The term "boundary" describes the target training range that a patient can tolerate from a motor ability perspective. However, this invention does not directly use this boundary as the execution boundary, but rather considers it as a third boundary to be fused.

[0143] In a preferred embodiment, the target load for the current training session Calculate using the following formula:

[0144] ;

[0145] In the formula, The target load for this training session; This is the prescription conversion factor; Peak oxygen uptake; This is a safety correction factor.

[0146] in, It's not a fixed value, but rather a safety factor restored through dialysis. With pulmonary circulation safety factor Obtained through fusion. Preferably, it can be expressed as:

[0147] ;

[0148] In the formula, For safety correction factors; To ensure the safety margin of dialysis recovery; For pulmonary circulation safety factor; This indicates taking the smaller value among the values ​​within the parentheses.

[0149] The reason for adopting this minimum strategy is that, for dialysis patients with pulmonary hypertension, if either the dialysis recovery status or the pulmonary circulation status is insufficient to support a higher training load, the more conservative boundary must be prioritized. This is significantly different from the common prescription approach of allowing training as long as physical fitness permits.

[0150] In further implementation, the system can also be based on and A second adjustment is made to the duration of the continuous motion segment and the proportion of the recovery segment. For example, when lower or When the level is too high, the upper limit of the duration of the continuous exercise segment can be appropriately shortened, and the proportion of the recovery segment can be appropriately increased.

[0151] 4. Fusion generation of the allowed domain during the current training.

[0152] After obtaining the dialysis recovery safety boundary, pulmonary circulation tolerance boundary, and exercise prescription boundary respectively, the system merges the three to form the allowable domain for the current training session. This allowable domain includes at least: the initial training load range; the upper limit of the target load; the maximum duration of the continuous exercise segment; the minimum duration of the recovery segment; the permissible deviation range of heart rate during training; the permissible fluctuation range of systolic blood pressure during training; the lower limit of blood oxygen during training; and the upper limit of symptom score during training.

[0153] It must be emphasized here that this invention does not provide a fixed target point, but rather an allowable domain. Its technical value lies in the fact that the training device can dynamically adjust within this allowable domain based on real-time risk conditions, without deviating from the safety range calculated before training. Because of this allowable domain for the current training session, subsequent real-time control is no longer a blind stop upon exceeding the limit, but rather a closed-loop adjustment based on individualized boundaries.

[0154] step The pre-training static prescription generation is upgraded to a multi-boundary fusion control domain generation mechanism that integrates dialysis recovery safety boundaries, pulmonary circulation tolerance boundaries, and exercise prescription boundaries. This approach of modeling dual safety boundaries before training and intersecting them with the prescription boundaries is significantly different from existing single exercise prescription schemes.

[0155] (III) Steps Specific implementation method

[0156] See Figure 4 This invention does not employ the method of immediately determining an anomaly once a single monitoring value exceeds the limit during training. Instead, it uses a dynamic identification mechanism that integrates rolling time window analysis and a comprehensive deviation index. This step addresses the challenge of stably identifying training risk profiles from multi-source, fluctuating, heterogeneous, and potentially noise-affected real-time signals during training of dialysis patients with pulmonary hypertension.

[0157] 1. Real-time training data acquisition and update mechanism

[0158] After training begins, the monitoring terminal continuously collects the following data: :time Real-time heart rate; :time Real-time systolic blood pressure; :time Real-time blood oxygen saturation; :time Subjective symptom scores; :time Real-time motion load; :time Real-time cadence.

[0159] Preferably, heart rate and cadence can be measured every [time / time]. to Updated every second, blood oxygen levels can be increased every... to Updated every second; blood pressure can be adjusted according to equipment conditions. Instant Updated every minute, symptom scores can be entered at the end of each continuous exercise segment, or immediately when the patient complains of discomfort. The training execution terminal transmits data in real time. and The monitoring terminal transmits physiological parameters back in real time.

[0160] 2. Rolling Time Window Analysis Mechanism

[0161] The reason this invention does not employ single-point comparison is that the heart rate, blood pressure, and blood oxygen signals of dialysis patients during training are easily affected by human movement, probe contact, instantaneous changes in exertion, and equipment sampling noise. Judging risk based solely on values ​​at a single moment is highly prone to false alarms or missed detections. Therefore, this invention sets the length to [missing information]. A rolling time window is used to jointly determine risk trends through multiple consecutive sampling points.

[0162] Preferably, Can be , or After each sampling step, the system updates the window and analyzes the average offset, extreme offset, offset direction, and offset duration within the window. In this way, the system can not only determine whether it has deviated from the target, but also whether the deviation is accumulating.

[0163] 3. Construction Logic of the Comprehensive Deviation Index

[0164] To uniformly represent various deviations as a controllable training risk, this invention constructs a comprehensive deviation index. :

[0165] ;

[0166] In the formula, For a moment The overall deviation index; , , , , These are the weighting coefficients for each deviation term; This refers to the deviation from the real-time heart rate. This refers to the real-time blood oxygen deviation term; This refers to the deviation of real-time blood pressure. This refers to items that deviate from the symptom score; This is the load deviation term.

[0167] Preferably, the weighting coefficients satisfy:

[0168] ;

[0169] In the formula, , , , , These are the weighting coefficients for each deviation term.

[0170] The reason for normalizing the deviations is that heart rate, blood oxygen, blood pressure, symptom scores, and workload have different dimensions, value ranges, and risk sensitivities. Without normalization, it is impossible to directly perform uniform weighting. Normalized deviations can map deviations from different sources to a comparable risk scale.

[0171] (1) Real-time heart rate deviation

[0172] ;

[0173] In the formula, For a moment Real-time heart rate deviation; For a moment Real-time heart rate; Target heart rate; This represents the upper limit of the allowable deviation in heart rate.

[0174] This formula means that the greater the deviation of the heart rate from the target, and the more it exceeds the allowable range, the larger the deviation term. When the difference between the real-time heart rate and the target heart rate is small, Keep it low; as the cumulative deviation increases, It increases accordingly.

[0175] (2) Real-time blood oxygen deviation

[0176] ;

[0177] In the formula, For a moment Real-time blood oxygen deviation; This is the lower limit of blood oxygenation. For a moment Real-time blood oxygen saturation; This indicates taking the largest value among the values ​​within the parentheses.

[0178] The meaning of this formula is: when real-time blood oxygen is higher than or equal to the lower limit, the deviation term is... When real-time blood oxygen levels are below the lower limit, the deviation increases as the degree of deviation from the lower limit increases. This avoids unnecessarily including risk when blood oxygen levels are above the safe value.

[0179] (3) Real-time blood pressure deviation

[0180] ;

[0181] In the formula, For a moment Real-time blood pressure deviation; For a moment Real-time systolic blood pressure; The target systolic blood pressure reference value; This represents the upper limit of the allowable deviation in systolic pressure.

[0182] (4) Symptom score deviation items

[0183] ;

[0184] In the formula, For a moment Symptom score deviation items; For a moment Subjective symptom scores; This represents the upper limit of the symptom score.

[0185] The introduction of this item demonstrates that the present invention does not rely solely on instrument data, but also incorporates the patient's subjective discomfort into the risk assessment logic. For dialysis patients with pulmonary hypertension, chest tightness, shortness of breath, and fatigue often appear earlier than any single objective indicator, making it essential to include symptom score deviation items.

[0186] (5) Load Deviation Item

[0187] ;

[0188] In the formula, For a moment The load deviation term; For a moment Real-time motion load; The target motion load; This represents the upper limit of the allowable load offset.

[0189] This item reflects the deviation between the current equipment execution state and the training target. Its function is not only to describe whether the current load is reasonable, but also to determine whether training adjustments have taken effect. For example, if the system has issued a drag reduction command... If the load remains above the target for an extended period, the load deviation will remain high, thus prompting further upgrades to the control logic.

[0190] 4. Logic for determining the current training state

[0191] System basis Based on the trend of its changes within the rolling time window, the training states are divided into stable state, first-level compensation state, second-level compensation state, third-level compensation state, and termination state.

[0192] Preferably, the following logic can be adopted:

[0193] when When the value is below the first threshold, it is determined to be in a stable state;

[0194] when When the value is between the first threshold and the second threshold, it is determined to be in a first-level compensation state;

[0195] when When the value is between the second and third thresholds, it is determined to be in a level two compensation state;

[0196] when When the value is between the third and fourth thresholds, it is determined to be a level three compensation state;

[0197] when When the value exceeds the fourth threshold and continues for a preset duration, it is determined to be in a terminated state.

[0198] In a more preferred embodiment, not only looking The current value is also considered in conjunction with its direction of change and duration within the scrolling time window. For example, if Although the termination threshold has not been exceeded, a continuous and rapid increase in risk, coupled with the simultaneous deterioration of multiple deviations, may allow for early entry into a higher-level compensation state. Thus, this invention achieves trend-based risk identification, rather than static identification.

[0199] step The breakthrough lies not only in defining a comprehensive deviation index, but also in proposing a rolling time window, multi-deviation fusion, and risk trend identification mechanism suitable for training scenarios involving dialysis patients with pulmonary hypertension. This mechanism significantly outperforms existing single-point threshold comparison models, and can more stably and accurately reflect the true risk status during the training process.

[0200] (iv) Steps Specific implementation method

[0201] step It is used to perform fast-loop control during training. Its core idea is not to simply stop upon an anomaly, but to prioritize maintaining training as much as possible within a safe range through a tiered backoff chain, thereby balancing training safety and training completion.

[0202] In a preferred embodiment, the system has at least the following control levels:

[0203] Level 1 retraction: Reduce the resistance of the training equipment;

[0204] Level 2 pullback: Reduce the target cadence;

[0205] Level 3 regression: shortening the duration of the continuous movement segment and extending the duration of the recovery segment;

[0206] Level 4 rollback: Switch to recovery training mode;

[0207] Training terminated.

[0208] For example, when Upon entering the first-level compensation zone, the system first reduces resistance to rapidly weaken the training stimulus without immediately interrupting the training. If the risk is not alleviated, it enters the second-level compensation zone and further reduces the target cadence. If the risk continues to accumulate, it enters the third-level compensation zone, shortening the continuous exercise segment and lengthening the recovery segment. If recovery is still not possible, it switches to the recovery training mode. If the termination condition is finally met, the training stops.

[0209] The preferred termination criteria are:

[0210] ;

[0211] In the formula, The duration of the unsafe state; The threshold for the duration of training termination.

[0212] This persistent decision-making logic enables the present invention to tolerate short-term fluctuations and avoid frequent training interruptions due to instantaneous artifacts or brief physiological fluctuations.

[0213] When the system terminates training, it automatically generates reduction constraints for the next training cycle. These reduction constraints include at least one or more of the following: lowering the upper limit of the starting intensity of the next training cycle, lowering the upper limit of continuous motion duration, and increasing the proportion of recovery segments. In this way, risks discovered during training are not only addressed in the current cycle but also directly affect the starting boundary of the next cycle.

[0214] step The essence lies in the fact that it is not binary control, but hierarchical backtracking control. This means that the system can take control actions of varying intensities for different levels of risk, rather than treating all anomalies equally. This improves both training safety and training completion rate.

[0215] (V) Steps Specific implementation method

[0216] step This is another step with a highly significant inventive contribution to the present invention. Its essential innovation lies in elevating post-training evaluation from simply recording results to becoming the input for reconstructing the boundary of the next training cycle. In other words, the present invention not only controls a single training session but also controls the continuous evolution process between training cycles.

[0217] 1. Training completion calculation

[0218] Training completion Calculate using the following formula:

[0219] ;

[0220] In the formula, Training completion rate; This represents the actual amount of training completed in this session; This is the training volume for this plan.

[0221] in, and The training volume can be expressed using load-time integral, cumulative completed work, target phase completion ratio, or other equivalent methods. This invention does not limit the only way to express the training volume, as long as it can truly reflect the difference between planned training and actual training.

[0222] Training completion The significance is that if the training is basically completed under safe control, it indicates that the current boundary setting is generally reasonable; if the training is frequently terminated early or the completion rate is low, it indicates that the current boundary may be too aggressive and needs to be automatically tightened in the next cycle.

[0223] 2. Calculation of Restoration Correction Factor

[0224] Recovery correction factor Calculate using the following formula:

[0225] ;

[0226] In the formula, To restore the correction factor; , , To adjust the weighting coefficients; Training completion rate; To restore compliance; This represents the penalty value for abnormal events.

[0227] in, Used to reflect whether recovery is good after training. Preferably, It is determined by the degree of recovery in heart rate, blood oxygenation, and blood pressure during the recovery period. The more complete the recovery, the better. The higher the level, the worse the recovery. The lower.

[0228] This is used to reflect the severity of adverse events during training. Preferably, events such as chest tightness, shortness of breath, low blood pressure, abnormal heart rate, and training interruption can be assigned different weights and then accumulated to form an adverse event penalty value. The more severe and frequent the adverse events, the higher the penalty value. The larger.

[0229] 3. Boundary update rules for the next training cycle

[0230] The system according to The values ​​are used to update the boundaries for the next training cycle:

[0231] when Then, perform a downgraded update for the next training session;

[0232] when At that time, keep the boundaries unchanged for the next training session;

[0233] when Then, perform a graded update for the next training session.

[0234] In the formula, This is the order reduction threshold; This is the order-up threshold, and .

[0235] The updated objects include at least: the starting load for the next training session, the upper limit of the target load, the upper limit of the duration of continuous exercise segments, the proportion of recovery segments, the upper limit of symptom scores, and the alarm sensitivity.

[0236] The essence of this update rule is as follows: if the completion rate of this training is high, the recovery is good, and there are few anomalies, it means that there may still be room for improvement in the current training boundary. In the next training session, the load limit can be appropriately increased or the proportion of the recovery phase can be reduced. If the completion rate of this training is low, the recovery is poor, or there are many anomalies, it means that the current boundary setting is too aggressive. In the next training session, it should be automatically adjusted to a more conservative direction. If the performance of this training is in an intermediate state, the boundary should be kept unchanged to maintain the continuity of training.

[0237] 4. The significance of slow loop closure across training cycles

[0238] Many existing training control systems, even after statistical analysis is performed following training, only generate reports and do not automatically feed back to the next training cycle. This invention, however, differs. This invention... and The training results are incorporated into the boundary reconstruction process of the next training cycle, thus forming a slow closed loop between training cycles.

[0239] The advantage of this slow-loop closed-loop technology lies in its continuous learning capability. It can automatically adjust the starting conditions for the next cycle based on the patient's actual training response from the previous cycle, eliminating the need for medical staff to rely entirely on experience to reset the system each time. For patients with poor recovery, the system automatically tends towards a conservative approach; for patients with good recovery, the system automatically advances to a more advanced approach. Thus, this invention upgrades single-cycle training control to multi-cycle adaptive control.

[0240] step By elevating post-training evaluation from a secondary step to a core control input, this invention possesses cyclic memory and continuous optimization capabilities. This feature fundamentally distinguishes it from existing solutions that only offer single-cycle monitoring and adjustment.

[0241] See Figure 6 An electronic device for implementing the method of the present invention includes a processor, a memory, a communication interface, and an input / output unit. The processor is used to execute program instructions stored in the memory to implement the steps. To the steps The entire process; the communication interface is used to connect dialysis equipment, training execution terminal, monitoring terminal and external information system; the input / output unit is used to display training suggestions, status prompts and historical trends, and to receive rating input and confirmation operations from therapists or patients.

[0242] The corresponding computer-readable storage medium stores a computer program, which, when executed by a processor, implements the maintenance dialysis pulmonary hypertension exercise intervention regulation method described in this invention.

[0243] IV. Specific Application Examples and Experimental Data

[0244] (I) Application Example 1: Single Training Closed-Loop Control Example

[0245] This application example combines Figure 1 , Figure 3 , Figure 4 and Figure 5 This describes the specific operation process of the present invention in a single exercise intervention. Figure 1 The overall flow of the method of the present invention is shown. Figure 3 This illustrates the fusion process of the pre-training dialysis recovery safety boundary, pulmonary circulation tolerance boundary, and exercise prescription boundary. Figure 4 This illustrates the rolling monitoring and hierarchical closed-loop control process during training. Figure 5 The cross-cycle write-back update process after training is shown.

[0246] A target patient on maintenance hemodialysis with signs of pulmonary hypertension was selected, and dialysis treatment had been ongoing. The patient's condition stabilized after one month. The patient's baseline condition data before training are as follows:

[0247]

[0248] First, the system according to Figure 3 The logic shown calculates the change in weight before and after dialysis:

[0249] ;

[0250] In the formula, This represents the change in body weight before and after dialysis. Weight before dialysis; Weight after dialysis.

[0251] Substituting the above data, we get Combining , and The system determined that the patient's recovery status after this dialysis session was of medium risk and generated a dialysis recovery safety factor. At the same time, according to , and The system determined the pulmonary circulation tolerance status to be moderately conservative and generated a pulmonary circulation safety factor. .

[0252] Safety correction factor Determine by the following formula:

[0253] ;

[0254] In the formula, For safety correction factors; To ensure the safety margin of dialysis recovery; For pulmonary circulation safety factor; This indicates taking the smaller value among the values ​​within the parentheses.

[0255] Therefore, we obtain The target load for this training session Determine using the following formula:

[0256] ;

[0257] In the formula, The target load for this training session; This is the prescription conversion factor; Peak oxygen uptake; This is a safety correction factor.

[0258] In this example, take ,but The system maps it to a power vehicle The target training load is set as follows, and the target cadence is set as follows. to The upper limit of the continuous motion segment is The lower limit of the recovery segment is Blood oxygen saturation lower limit is .

[0259] During training, the system follows Figure 4 As shown, each Collect heart rate, blood oxygen, cadence, and load data once, every Update the scrolling time window once. Overall deviation index. Calculate using the following formula:

[0260] ;

[0261] In the formula, For a moment The overall deviation index; , , , , These are the weighting coefficients for each deviation term; This refers to the deviation from the real-time heart rate. This refers to the real-time blood oxygen deviation term; This refers to the deviation of real-time blood pressure. This refers to items that deviate from the symptom score; This is the load deviation term.

[0262] The weighting coefficients satisfy the following:

[0263] ;

[0264] In the formula, , , , , These are the weighting coefficients for each deviation term.

[0265] In this example, take , , , , .

[0266] The following are some records from the training process:

[0267]

[0268] As can be seen from the table, in the first... At that time, the patient Down to Symptom score Rise to The system determines that it has entered the first-level compensation range and automatically reduces resistance; hour, As the frequency increases further, the system enters secondary compensation, reducing the target cadence and shortening the continuous motion segment; back, The training resumes after the initial drop. This result demonstrates that the present invention does not rely on immediately stopping training to ensure safety, but rather... Figure 4 The tiered fallback chain shown achieves risk mitigation, thus balancing security and training completion.

[0269] After training is completed, the system calculates the training completion rate. :

[0270] ;

[0271] In the formula, Training completion rate; This represents the actual amount of training completed in this session; This is the training volume for this plan.

[0272] In this example, Recovery period Within a short period, the patient's heart rate, blood pressure, and blood oxygen levels all returned to the preset range, achieving the target recovery rate. Exception event penalty value Restoration correction factor Calculate using the following formula:

[0273] ;

[0274] In the formula, To restore the correction factor; , , To adjust the weighting coefficients; Training completion rate; To restore compliance; This represents the penalty value for abnormal events.

[0275] Pick , , ,get:

[0276] ;

[0277] In the formula, To restore the correction factor; Training completion rate; To restore compliance; This represents the penalty value for abnormal events.

[0278] because While in the hold-up range, the system maintains the target load limit for the next training cycle unchanged, only slightly increasing the proportion of the recovery segment. This process corresponds to... Figure 5 This demonstrates that the present invention can write back the results of a single training cycle to the next training cycle.

[0279] (II) Application Example 2: Multi-period closed-loop update example

[0280] This application example demonstrates the cross-training cycle update effect of the present invention. The same target patient is selected and continuously executed... Each training cycle is repeated, and after each training cycle, based on... Figure 5 The logical calculation shown The starting load, duration of continuous motion segment, and proportion of recovery segment for the next cycle are updated.

[0281]

[0282] As can be seen from the above data, in the... To the In the first training session, the patient's recovery rate was relatively low and the penalty value for abnormal events was high, so the system kept the training boundaries conservative; After that... and improve, The system gradually relaxes the target load and the upper limit of continuous motion segments. This demonstrates that the present invention can achieve multi-cycle adaptive optimization through a post-training write-back mechanism, rather than mechanically executing a fixed-intensity prescription. Figure 7 As described in the embodiments of the present invention The trend chart of cross-cycle updates with target load, where the left vertical axis represents The right vertical axis represents the target load, and the horizontal axis represents the training cycle, which is used to visually demonstrate the effect of slow loop updates across cycles.

[0283] (III) Application Example 3: Group Control Experiment Example

[0284] This application example demonstrates the comprehensive technical advantages of this invention compared to a fixed exercise prescription. The experimental design follows the clinical research framework described in the grant specification: patients undergoing maintenance hemodialysis with pulmonary hypertension were included and randomly assigned to groups using a random number table; the experimental period was [duration missing]. Week, every week Each time The observation indicators include pulmonary artery systolic blood pressure, right ventricular function, The training program includes factors such as walking distance per minute and quality of life. The task description clearly defines the control and experimental groups, training intensity, training frequency, and observation time points.

[0285] This example sets up two groups: the comparative group uses a fixed exercise prescription, i.e., according to... Power vehicle training is performed, but the dual-boundary fusion, rolling time window comprehensive deviation evaluation, and cross-cycle write-back update of the present invention are not executed; the example group adopts the closed-loop control method of the present invention, in nominal... Based on this, the training allowable domain is dynamically adjusted according to the dialysis recovery safety boundary and the pulmonary circulation tolerance boundary, and graded rollback during training and cross-cycle update after training are performed.

[0286] 1. Baseline data

[0287]

[0288] The baseline results above indicate that the two groups were similar in key metrics before training and were comparable.

[0289] 2. Training safety and completion results

[0290]

[0291] As can be seen from the table, the average training completion rate of the example group reached The aerobic training completion rate meets the requirements outlined in the task description. The above objectives were achieved. Meanwhile, the incidence of chest tightness, hypoxia, hypotension, and premature termination events in the example group were significantly lower than those in the control group, demonstrating that the present invention achieves these objectives. Figure 3 and Figure 4 The pre-training dual-boundary control and in-training graded closed-loop regulation shown can effectively improve the safety of exercise intervention.

[0292] 3. Changes in pulmonary circulation and functional indicators

[0293]

[0294] The above experimental data show that the first During the week, the example group Depend on Down to The decrease was Comparative group Depend on Down to The decrease was Right ventricular function indicators in the example group , and The improvement in overall scores was also greater than that of the control group. This result indicates that the present invention does not simply improve training safety, but rather improves training effectiveness while maintaining greater safety. Figure 8 As shown, the embodiments of the present invention and the comparative examples of fixed prescriptions are in... Weekday , , and Comparison chart of changing trends. Figure 9 The comprehensive deviation index for a single training process in this embodiment of the invention. A diagram showing the correspondence between the risk changes during training and the tiered backoff trigger points. This diagram demonstrates that changes in risk during training can be identified by rolling monitoring and the comprehensive deviation index, triggering corresponding first-, second-, or third-level backoff controls. to Inside, Within a stable range; in the first... back, Entering the first-level compensation range and triggering a reduction in resistance; in the... about, Entering the secondary compensation zone triggers a reduction in cadence and a shortening of the continuous motion segment; after intervention... The temperature dropped back to a lower level, thus demonstrating that the closed-loop regulation in the training process of this invention is real-time and effective.

[0295] The specific application examples and experimental data mentioned above demonstrate that this invention, through the fusion generation of pre-training dialysis recovery safety boundaries, pulmonary circulation tolerance boundaries, and exercise prescription boundaries, transforms exercise intervention from a fixed prescription model to a state-adaptive prescription model; and utilizes rolling time windows and a comprehensive deviation index during training. It can promptly identify comprehensive risk changes in heart rate, blood pressure, blood oxygen, symptom scores, and exercise load, and mitigate risks through a graded regression chain; it also assesses training completion rates after training. and recovery correction factor It can write the training results back to the next training cycle, thus forming a cross-cycle closed-loop optimization.

[0296] Compared to a fixed exercise prescription comparison, the training completion rate of the embodiments of the present invention is [missing information]. Increase to During training, the incidence of hypoxia, chest tightness, hypotension, and premature termination events was significantly reduced; During the week, the example group Decrease Increase range Increase and The improvement in scores was significantly greater than that of the control group. This demonstrates that the present invention can simultaneously improve safety, completion rate, and rehabilitation effect in exercise intervention for maintenance dialysis patients with pulmonary hypertension, and has clear engineering feasibility and clinical application value.

[0297] The foregoing description of embodiments of the present invention, through which those skilled in the art are able to implement or use the present invention, will be readily apparent to those skilled in the art. Various modifications to these embodiments will be readily apparent to those skilled in the art. The general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novelty disclosed herein.

[0298] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0299] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0300] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0301] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0302] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0303] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0304] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

Claims

1. A method for exercise intervention and regulation of pulmonary hypertension in maintenance dialysis, characterized in that, Includes the following steps: S1. Collect baseline status data of the target patient within a preset recovery time window after one hemodialysis session. The baseline status data includes at least dialysis recovery data, pulmonary circulation tolerance data, exercise tolerance data, and historical training response data. S2. Based on the aforementioned basic state data, construct the dialysis recovery safety boundary, the pulmonary circulation tolerance boundary, and the exercise prescription boundary respectively, and fuse them to obtain the allowable domain for the current training session; S3. During the exercise intervention for the target patient, real-time training data is continuously collected. Trend analysis is performed on the real-time training data based on a rolling time window, and a comprehensive deviation index is calculated. To determine the current training status; S4. Compare the current training state with the allowed range for the current training session. When the current training state is within the compensation range, perform one or more levels of load rollback adjustment. When the current training state continues to reach the termination range, terminate the current training session and generate the next training order reduction constraint. S5. After this training session, collect recovery period response data and abnormal event records, and calculate the training completion rate. and recovery correction factor And based on the training completion rate and recovery correction factor Update the initial control boundary for the next training iteration and repeat steps S1 through S5. The comprehensive deviation index is calculated using the following formula: In the formula, For a moment The overall deviation index; , , , , Let be the weight coefficients of each deviation term, and satisfy . ； in, This is the real-time heart rate deviation term; This refers to the real-time blood oxygen deviation term; This refers to the deviation of real-time blood pressure. This refers to items that deviate from the symptom score; This is the load deviation item; The restoration correction factor is calculated using the following formula: ； In the formula, To restore the correction factor; , , To adjust the weighting coefficients; To restore compliance; This represents the penalty value for abnormal events.

2. The method according to claim 1, characterized in that, In step S1, the dialysis recovery data includes at least the ultrafiltration volume. , weight before dialysis Post-dialysis weight Number of hypotension events during dialysis and the duration of symptom recovery after dialysis The change in body weight before and after dialysis is calculated using the following formula: ； In the formula, This represents the change in body weight before and after dialysis. Weight before dialysis; Post-dialysis weight; and / or, in step S1, the pulmonary circulation tolerance data includes at least pulmonary artery systolic pressure. Right ventricular function indicators resting heart rate Resting blood oxygen saturation and resting systolic blood pressure ; and / or, in step S1, the exercise tolerance data includes at least peak oxygen uptake. 6-minute walk and subjective fatigue rating ; and / or, in step S1, the historical training response data includes at least the previous training completion rate. Recovery time after previous training and the level of previous abnormal events ; And / or, in step S1, the preset recovery window is the [number]th [time] after dialysis ends. Hours to the The time interval between hours, To be the earliest allowed training start time, The latest allowed start time for training, and ; And / or, in step S1, after constructing the baseline status data, a fitness screening is performed to exclude target patients who have active cardiovascular instability, acute respiratory failure, severe hypotension, or who are clinically deemed unsuitable for exercise intervention.

3. The method according to claim 2, characterized in that, In step S2, the dialysis recovery safety boundary is used to limit the maximum training load, the maximum continuous exercise duration, and the minimum recovery period duration allowed in the current training session; The pulmonary circulation tolerance boundary is used to define the permissible range of heart rate deviation, blood pressure fluctuation, lower limit of blood oxygen, and upper limit of symptom score for the current training session; the exercise prescription boundary is used to define the target intensity, target duration, and target rhythm for the current training session; the permissible domain for the current training session is a common control domain that simultaneously satisfies the dialysis recovery safety boundary, the pulmonary circulation tolerance boundary, and the exercise prescription boundary; and / or, in step S2, the target load for the current training session is calculated using the following formula: ； In the formula, The target load for this training session; This is the prescription conversion factor; Peak oxygen uptake; This is a safety correction factor determined based on dialysis recovery status and pulmonary circulation tolerance status; And / or, in step S2, the dialysis recovery safety boundary is generated based on the dialysis recovery risk level, which is divided into at least three levels: low risk, medium risk, and high risk; the higher the risk level, the lower the upper limit of the intensity allowed for the current training, the shorter the duration of the continuous exercise segment, and the longer the duration of the recovery segment; And / or, in step S2, the pulmonary circulation tolerance boundary is determined by... , , , and It was jointly determined that, The systolic blood pressure of the pulmonary artery. This is an indicator of right ventricular function. Resting heart rate Resting blood oxygen saturation This is the resting systolic blood pressure.

4. The method according to claim 1, characterized in that, In step S3, the real-time training data includes at least real-time heart rate. Real-time systolic blood pressure Real-time blood oxygen saturation Real-time symptom score Real-time exercise load and real-time cadence ; And / or, in step S3, the scrolling time window has a length of The continuously updated analysis interval, where, The preset time length is used; the comprehensive deviation index is updated once after each sampling step. The current training state is determined based on the direction of change, degree of deviation, and duration of multiple consecutive sampling points within the rolling time window.

5. The method according to claim 4, characterized in that, In step S3, the real-time heart rate deviation term is calculated using the following formula: In the formula, For a moment Real-time heart rate deviation; For a moment Real-time heart rate; Target heart rate; This represents the upper limit of the allowable deviation in heart rate. And / or, in step S3, the real-time blood oxygen deviation term is calculated using the following formula: ； In the formula, For a moment Real-time blood oxygen deviation; This is the lower limit of blood oxygenation. For a moment Real-time blood oxygen saturation; And / or, in step S3, the real-time blood pressure deviation term is calculated using the following formula: ； In the formula, For a moment Real-time blood pressure deviation; For a moment Real-time systolic blood pressure; The target systolic blood pressure reference value; This is the upper limit of the allowable deviation of systolic pressure; And / or, in step S3, the symptom score deviation item is calculated using the following formula: ； In the formula, For a moment Symptom score deviation items; For a moment Subjective symptom scores; This represents the upper limit of the symptom score. And / or, in step S3, the load deviation term is calculated using the following formula: ； In the formula, For a moment The load deviation term; For a moment Real-time motion load; The target motion load; This represents the upper limit of the allowable load offset.

6. The method according to claim 1, characterized in that, In step S4, the one-level or multi-level load reduction adjustment includes at least one or more of the following: reducing resistance, reducing target cadence, shortening the duration of the continuous exercise segment, extending the duration of the recovery segment, and switching to a recovery training mode; the training termination determination condition satisfies: In the formula, The duration of the unsafe state; The threshold for the duration of training termination; And / or, in step S4, the backoff adjustment corresponding to the compensation interval is a three-level graded adjustment: when When entering the first-level compensation zone, reduce motion resistance; when When entering the secondary compensation zone, reduce the target cadence and shorten the duration of the continuous movement segment; when When entering the third-level compensation range, switch to the recovery segment; And / or, in step S4, the next training reduction constraint includes at least one or more of the following: reducing the upper limit of the starting intensity of the next training, reducing the upper limit of the continuous exercise duration, and increasing the proportion of the recovery segment.

7. The method according to claim 1, characterized in that, In step S5, the training completion rate is calculated using the following formula: ； In the formula, For training completion rate; This represents the actual amount of training completed in this session; This is the planned training volume; In step S5, the recovery achievement rate The degree of recovery is determined by the extent of heart rate recovery, blood oxygen recovery, and blood pressure recovery during the recovery period; when When, perform a downgraded update for the next training session; when When, keep the training boundaries unchanged for the next time; when At that time, a graded update is performed for the next training session; where, This is the threshold for order reduction; This is the order-up threshold, and ; In step S5, the abnormal event record includes at least one of the following: chest tightness event, shortness of breath event, low blood pressure event, abnormal heart rate event, and training termination event.

8. A closed-loop regulation system for exercise intervention in maintenance dialysis pulmonary hypertension, characterized in that, The system is used to implement the method described in any one of claims 1-7, comprising: a basic state acquisition module for performing step S1; a dual-boundary construction and prescription generation module for performing step S2; a rolling monitoring and comprehensive deviation assessment module for performing step S3; a hierarchical closed-loop control module for performing step S4; and a cross-cycle write-back update module for performing step S5.

9. An electronic device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 7.

Images