A blood pump apparatus remote monitoring parameter adjustment control system and method

Abstract

This invention relates to the field of biomedical engineering and discloses a remote monitoring parameter adjustment control system and method for blood pump equipment. The system includes acquiring pump speed and hemodynamic response signals, performing phase calibration by combining the pulsation period and fluid inertial delay, and constructing a non-uniform response segment. For alternating sections of flow rate drops and pressure drops in the flow response expression structure, cross-segment consistency is checked to form an abnormal residual distribution. The source of disturbance is decomposed around the abnormal residual distribution, dividing the disturbance according to pump load changes, vascular compliance feedback, and measurement hysteresis channels, and reconstructing the disturbance propagation path. Under the constraint of the disturbance propagation path, with the goal of maintaining perfusion stability and avoiding negative pressure, amplitude-limited trial adjustments are applied to key nodes to construct a regulatory response feature set. Combining hemodynamic stability criteria and individual physiological constraints, a regulatory strategy is output. This invention has the advantages of improving blood perfusion stability and the controllability of the regulation process.

Description

Technical Field

[0001] This invention relates to the field of biomedical engineering, specifically to a remote monitoring parameter adjustment and control system and method for blood pump devices. Background Technology

[0002] With the widespread application of extracorporeal circulation support technology and ventricular assist devices, blood pump equipment is continuously used in scenarios such as intensive care, postoperative recovery, and long-term circulatory support. To improve operational safety and clinical response efficiency, existing blood pump systems typically incorporate remote monitoring and parameter adjustment mechanisms. These mechanisms involve real-time acquisition of operating parameters such as blood flow, pump speed, and pressure, and remote intervention control when necessary. Current remote monitoring systems mostly employ a fixed-time data upload method, transmitting blood pump operating parameters to a remote monitoring terminal according to a preset sampling cycle. Medical personnel or the control system then use this discrete data to determine the status and adjust parameters. However, this approach is insufficient when the patient's position changes significantly. When there are transient fluctuations in time or hemodynamic state, the blood pump load and flow parameters will undergo continuous disturbance evolution within a short period of time. However, the existing sampling method based on fixed period is difficult to fully characterize the continuous change process of such transient disturbances, so the propagation correlation of disturbances in the time dimension cannot be accurately reflected. This leads to the following consequences: First, key transient changes may be missed within the sampling interval, resulting in information lag on the remote side; Second, short-term disturbances may have recovered by the time of feedback, but may still be misjudged as abnormal states; Third, when adjusting based on discrete data, it is difficult to reflect the true evolution of the disturbance, resulting in deviations between control commands and actual operating conditions.

[0003] Therefore, it is essential to design a remote monitoring parameter adjustment control system and method for blood pump equipment to improve the stability of blood perfusion and the controllability of the regulation process. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a remote monitoring parameter adjustment and control system and method for blood pump equipment, which has the advantages of improving blood perfusion stability and controllability of the regulation process, and solves the problems mentioned in the background technology.

[0005] To achieve the aforementioned objectives of improving blood perfusion stability and controllability, this invention provides the following technical solution: a method for remote monitoring parameter adjustment and control of a blood pump device, comprising the following steps: The pump speed and hemodynamic response signals were collected, and phase calibration was performed by combining the pulsation period and fluid inertial delay. The segment boundaries were determined based on the changes in physiological rhythm perturbation, and adaptive segmentation processing was performed to construct non-uniform response segments. For the alternating segments of flow drop and pressure fall in the flow response expression structure, cross-segment consistency is checked, and the convergence offset of the homologous response is calculated by multi-path reconstruction. Abnormal segments that do not meet the requirements of blood flow continuity and perfusion stability are extracted to form abnormal residual distribution. The sources of disturbance are decomposed based on the abnormal residual distribution. The disturbances are divided into pump load changes, vascular compliance feedback and measurement hysteresis channels. The transmission relationship of the disturbances in each channel is determined by combining the time decay relationship and mutual influence of each disturbance component, and the disturbance propagation path is reconstructed. Under the constraint of the disturbance propagation path, with the goal of maintaining injection stability and avoiding negative pressure, a trial adjustment with limited amplitude is applied to key nodes, and the change of disturbance in the propagation path before and after adjustment is tracked to construct an adjustment response feature set. Based on the regulatory response feature set, combined with hemodynamic stability criteria and individual physiological constraints, the convergence, oscillation amplitude and perfusion deviation of candidate control parameters are evaluated, and a regulatory strategy that satisfies hemodynamic stability and physiological constraints is output.

[0006] Preferably, the process of constructing a non-uniform response segment is as follows: The pump speed signal and hemodynamic response signal are synchronized in time, and a phase consistency mapping relationship is established by combining the pulsation cycle characteristics and fluid inertial delay. Periodic phase calibration of hemodynamic response is performed based on phase consistency mapping relationship, and physiological rhythm perturbation change characteristics are extracted; Based on the characteristics of physiological rhythm perturbation changes, the boundaries of local fluctuation intervals on the time axis are identified, and segmented boundary candidates are constructed. Stability screening and error constraint optimization are performed on the candidate segment boundaries, and adaptive non-uniform segmentation is carried out to construct non-uniform response segments.

[0007] Preferably, the process for checking cross-segment consistency is as follows: For the alternating flow drop and pressure fall sections in non-uniform response segments, a cross-segment hemodynamic continuity alignment relationship is constructed based on the hemodynamic response characteristics of adjacent segments; Based on the cross-segment hemodynamic continuity alignment relationship, the pump output response between adjacent segments is coupled and mapped with the hemodynamic feedback, and consistency matching analysis is performed on the changes in flow rate, pressure and rotation speed. Based on the consistency matching analysis results, the cross-segment response deviation index is calculated, and segments with deviations exceeding the threshold are identified and marked according to the blood perfusion stability constraints, forming cross-segment inconsistency abnormal segments.

[0008] Preferably, the process of forming an abnormal residual distribution is as follows: Based on cross-segment inconsistency abnormal segments, multi-pathway retrospective reconstruction of homologous hemodynamic response is performed to generate a corresponding set of multi-pathway response trajectories. Based on the set of multi-path response trajectories, deviation analysis is performed on the consistency of hemodynamic response under different pathways, and the convergence offset between trajectories is calculated to characterize the trajectories. Based on the convergence offset characterization results, blood flow continuity and perfusion stability are jointly determined, and abnormal segments exceeding the stability constraint threshold are extracted and summarized to form an abnormal residual distribution.

[0009] Preferably, the process of classifying the disturbance according to pump load changes, vascular compliance feedback, and measurement hysteresis channels is as follows: Based on the abnormal residual distribution, the system disturbance is structurally analyzed and divided into pump load change disturbance component, vascular compliance feedback disturbance component and measurement lag observation disturbance component. By combining the attenuation characteristics of each disturbance component in the time dimension and their mutual coupling influence, the interaction strength between disturbance components is quantitatively characterized, forming a disturbance correlation characterization result; Based on the perturbation correlation characterization results, the propagation path of the perturbation in the pump drive channel, vascular response channel and observation delay channel is inferred and analyzed, and the perturbation propagation path results are obtained.

[0010] Preferably, the process of reconstructing the disturbance propagation path is as follows: Based on the perturbation propagation path results, a transitive mapping relationship with time-dependent constraints is constructed between the pump-driven channel, the vascular response channel, and the observation delay channel; Based on the transfer mapping relationship, the consistency conditions of time decay, phase offset and propagation direction of the disturbances in each channel are introduced to perform constraint-driven consistency evolution screening and path convergence processing on the cross-channel disturbance propagation process. The cross-channel consistent propagation trajectory that satisfies the constraints is used as the constraint convergence result to form the perturbation propagation path.

[0011] Preferably, the process of applying amplitude-limited trial adjustments to key nodes is as follows: Based on the disturbance propagation path structure, the influence weight of each node in the hemodynamic response is evaluated, and key control nodes that contribute significantly to blood flow stability are selected. Under the constraints of maintaining stable blood perfusion and avoiding the risk of negative pressure, exploratory regulatory inputs with limited amplitude are applied to key control nodes; The evolution of the disturbance propagation path under the trial regulation is monitored in real time to obtain the path structure offset characteristics and the redistribution of disturbance energy in each channel; By combining blood flow fluctuations, pressure change trends, and negative pressure risk indicators, the impact of trial-and-error input on local blood flow stability is comprehensively assessed, generating assessment results to characterize changes in the perturbation propagation response.

[0012] Preferably, the process of constructing the regulation response feature set is as follows: Comparative analysis of the disturbance propagation paths before and after the trial adjustment was conducted to extract information on changes in path structure; Based on the evaluation results, the credibility of the path structure change is weighted and corrected, and the difference in the distribution change of disturbance energy in each propagation channel is calculated to form a disturbance redistribution characteristic index. Combining the temporal variation characteristics of hemodynamic response and based on the flow and pressure assessment results in the evaluation results, the response difference vectors of flow and pressure before and after regulation are extracted; The path structure change information, disturbance redistribution characteristic indicators and response difference vector are fused and mapped to form a multidimensional regulation response feature. The multidimensional regulatory response features are sequentially organized according to their temporal evolution to construct a regulatory response feature set.

[0013] The preferred process for outputting a regulatory strategy that satisfies both hemodynamic stability and physiological constraints is as follows: A candidate control parameter space is constructed based on the regulation response feature set, and simulation mapping is performed on different parameter combinations to generate candidate control strategies. Based on hemodynamic stability criteria, dynamic convergence analysis is performed on the parameter combinations in the candidate control strategies to generate stability scores. Based on the stability score results, the oscillation amplitude and periodic fluctuation characteristics of blood flow response under different control strategies are calculated to form an oscillation assessment index. By combining individual physiological constraints, blood perfusion deviations are corrected and corresponding risk assessment results are generated. Based on stability scores, oscillation assessment indicators, and risk assessment results, a multidimensional fusion decision is made to output a regulatory strategy that satisfies hemodynamic stability and physiological constraints.

[0014] A remote monitoring and parameter adjustment control system for a blood pump device, comprising: Phase segmentation module: Performs phase calibration on pump speed and blood flow response, and determines segmentation boundaries by combining pulsation period and inertial delay to complete non-uniform segmentation; Residual detection module: performs cross-segment consistency checks on the flow drop and pressure recovery sections, and calculates the convergence offset through multi-path reconstruction to extract abnormal residual distribution; Disturbance modeling module: Based on the distribution of abnormal residual disturbances, the source of disturbances is identified and classified according to pump load, vascular compliance and measurement hysteresis, and the disturbance propagation path is reconstructed; Trial adjustment module: Apply amplitude-limited adjustment to key nodes under propagation path constraints, track changes in disturbance propagation, and generate a set of adjustment response features; Strategy optimization module: Based on the characteristics of the regulatory response combined with blood flow stability and physiological constraints, control parameters are screened and regulatory strategies are output.

[0015] Compared with the prior art, the present invention provides a remote monitoring parameter adjustment and control system and method for blood pump equipment, which has the following beneficial effects: This invention achieves refined temporal analysis of hemodynamic processes by performing phase calibration on pump speed and hemodynamic response signals and adaptive segmentation based on physiological rhythm perturbations. This effectively improves the temporal resolution capability for identifying abnormal fluctuations. Through cross-segment consistency checks and multi-path reconstruction, the convergence offset characteristics of homologous responses are quantified, thereby accurately extracting abnormal residual distributions reflecting blood flow continuity disruption and perfusion instability, improving the reliability and robustness of abnormal state detection. By performing multi-channel decomposition of perturbation sources and reconstructing perturbation propagation paths based on time decay and coupling relationships, the system can characterize pump load changes and vascular compliance at the structural level. The comprehensive impact mechanism of feedback and measurement lag on blood flow response enhances the ability to trace the source and explain the mechanism of disturbances. At the same time, under the constraint of disturbance propagation path, a trial adjustment mechanism with limited amplitude at key nodes is introduced, and a set of adjustment response features is constructed by comparing the evolution of disturbances before and after adjustment. This realizes the observable feedback and adaptive correction capability of the control strategy to the dynamic response of the system. Finally, by combining hemodynamic stability criteria and individual physiological constraints, a multi-dimensional comprehensive evaluation of the convergence, oscillation characteristics and perfusion deviation of control parameters is carried out. This enables safe, accurate and individualized optimization adjustment of blood pump operating parameters under remote monitoring conditions while ensuring blood perfusion stability and reducing the risk of negative pressure. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the method of the present invention; Figure 2 This is a schematic diagram of the structure of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1: Please refer to Figure 1 A method for remote monitoring parameter adjustment and control of a blood pump device according to an embodiment of the present invention includes the following steps: S1: Collect pump speed and hemodynamic response signals, perform phase calibration by combining pulsation period and fluid inertial delay, determine segment boundaries based on physiological rhythm perturbation changes, perform adaptive segmentation processing, and construct non-uniform response segments.

[0019] The process of constructing a non-uniform response segment in S1 is as follows: The pump speed signal and hemodynamic response signal are synchronized in time, and a phase consistency mapping relationship is established by combining the pulsation cycle characteristics and fluid inertial delay. When synchronizing the pump speed signal and hemodynamic response signal in time, the timestamps of the two types of signals are recalibrated based on a unified sampling clock to eliminate the time drift caused by the acquisition time offset. Using the pump speed cycle signal as a reference, the hemodynamic response signal is slidably aligned on the time axis to achieve a corresponding relationship within the same pulsation cycle window. Combining the periodic characteristics of the pump mechanical drive and the fluid inertial delay characteristics generated by blood transmission in the pipeline, the lag time of the blood flow response relative to the pump speed signal is estimated, and this lag relationship is mapped into a phase offset, thereby establishing a consistent phase correspondence between the pump speed and the hemodynamic response, providing a unified phase reference for periodic analysis. Based on the phase consistency mapping relationship, periodic phase calibration of hemodynamic response is performed, and physiological rhythm disturbance change characteristics are extracted. After obtaining the phase consistency mapping relationship, the hemodynamic response signal is subjected to periodic phase calibration processing, that is, the signal is divided into periods according to the pump pulsation cycle, and the hemodynamic response in each period is mapped to a unified phase coordinate system for normalization and alignment. The phase changes between different periods are compared and analyzed, and the offset trend and amplitude fluctuation characteristics of the hemodynamic response within the period are extracted. Combined with the natural fluctuation law of human physiological rhythm, local abnormal fluctuations that exceed the stable period range are identified, thereby extracting the physiological rhythm disturbance change characteristics that reflect changes in blood flow regulation, providing a basis for the division of abnormal intervals. Based on the characteristics of physiological rhythm perturbation, the boundaries of local fluctuation intervals on the time axis are identified, and candidate segment boundaries are constructed. Based on the extracted physiological rhythm perturbation characteristics, local fluctuation analysis is performed on the continuous signal on the time axis, focusing on identifying abrupt change points, gradual change inflection points, and abnormal amplitude clustering intervals in hemodynamic response. The consistency of change amplitude, change direction, and fluctuation duration between adjacent sampling points are used as the criteria for judgment. Locations that may form structural changes are marked. By clustering and interval constraint screening of multiple marked points, a set of candidate segment boundaries is formed, thereby dividing the continuous time series into several intervals with potential dynamic differences, providing a boundary basis for segmentation processing. The candidate segment boundaries are subjected to stability screening and error constraint optimization, and adaptive non-uniform segmentation is performed to construct non-uniform response segments. After obtaining the candidate set of segment boundaries, stability screening is performed on each candidate boundary, including evaluating the degree of continuity of hemodynamic response before and after the boundary and eliminating unstable boundary points caused by instantaneous noise or sampling errors. At the same time, an error constraint mechanism is introduced to locally fine-tune the boundary position, so that the hemodynamic changes between adjacent segments have higher distinguishability and consistency. Finally, based on the effective boundaries after screening and optimization, the original time series is subjected to non-uniform segmentation, so that the hemodynamic response in different time intervals exhibits structured segmentation characteristics, thereby constructing non-uniform response segments for analysis.

[0020] S2: For the alternating segments of flow drop and pressure drop in the flow response expression structure, cross-segment consistency is checked, and the convergence offset of the same source response is calculated through multi-path reconstruction. Abnormal segments that do not meet the requirements of blood flow continuity and perfusion stability are extracted to form abnormal residual distribution.

[0021] The process of checking cross-fragment consistency in S2 is as follows: For the alternating flow drop and pressure fall sections in non-uniform response segments, a cross-segment hemodynamic continuity alignment relationship is constructed based on the hemodynamic response characteristics of adjacent segments. For these alternating flow drop and pressure fall sections, the pump speed change trend and hemodynamic response change trend within adjacent time segments are synchronously extracted to obtain the flow change sequence, pressure change sequence, and their temporal correspondence for each segment. Using the temporal boundary between adjacent segments as the connection benchmark, an alignment analysis is performed on the continuity of hemodynamic response changes at segment boundaries. By constraining and matching the consistency of flow gradient and pressure gradient changes per unit time before and after the boundary, a hemodynamic continuity correspondence relationship between adjacent segments is established, thus forming the basic structure for describing cross-segment connection characteristics.

[0022] Based on the cross-segment hemodynamic continuity alignment, the pump output response and hemodynamic feedback between adjacent segments are coupled and mapped, and consistency matching analysis is performed on changes in flow rate, pressure, and rotational speed. On the basis of the cross-segment hemodynamic continuity alignment, the pump output signal and hemodynamic feedback signal are correlated in time sequence, and an input-response mapping relationship is established under a unified time reference. For the change process between adjacent segments, the consistency of response direction, amplitude change ratio, and hysteresis response characteristics of pump output change and hemodynamic feedback change are compared and analyzed. By judging whether the pump end change can generate a consistent direction of dynamic response at the hemodynamic end, cross-segment consistency matching analysis is achieved, thereby identifying the consistent or deviated coupling state between pump drive and hemodynamic response. Based on the consistency matching analysis results, a cross-segment response deviation index is calculated. Segments with deviations exceeding the threshold are identified and marked according to blood perfusion stability constraints, forming cross-segment inconsistency abnormal segments. Based on the consistency matching analysis results, the degree of deviation between pump output changes and hemodynamic response changes between adjacent segments is quantitatively calculated to form a cross-segment response deviation index. This index is uniformly characterized by comprehensively considering differences in flow rate changes, pressure changes, and response time lag differences to reflect the degree of deviation in cross-segment hemodynamic stability. The deviation index is compared with preset blood perfusion stability constraints. When the deviation exceeds the stability constraint threshold, it is determined that there is an abnormal dynamic inconsistency phenomenon in the boundary region of the corresponding segment, and the cross-segment interval is marked, thus forming a cross-segment inconsistency abnormal segment for abnormal residual analysis and disturbance modeling processing.

[0023] The process by which abnormal residual distribution is formed in S2 is as follows: Based on cross-segment inconsistency anomaly segments, multi-pathway retrospective reconstruction of homologous hemodynamic responses is performed to generate a corresponding set of multi-pathway response trajectories. Based on cross-segment inconsistency anomaly segments, the time window of anomaly occurrence and the range of adjacent hemodynamic response data before and after it are determined. Using pump speed changes, flow rate changes, and pressure changes within this time window as basic input information, the possible evolution process of hemodynamic response under different transmission delay conditions is retrospectively analyzed from different time lag paths. By considering the pump-driven instantaneous response path, the blood fluid inertial delay path, and the vascular elastic feedback path respectively, multiple response propagation possibilities corresponding to the same anomaly segment are reconstructed. Finally, the hemodynamic response change processes obtained under different paths are aggregated and organized to form a set of multi-pathway response trajectories used to describe the multi-pathway evolution characteristics of anomaly sources. Based on a multi-pathway response trajectory set, deviation analysis is performed on the consistency of hemodynamic response under different pathways, and the convergence offset between trajectories is calculated. On the basis of the multi-pathway response trajectory set, the changing trends of hemodynamic response under different pathways are compared and analyzed. The focus is on extracting the flow rate change amplitude, pressure change amplitude, and response trend direction of each trajectory at the same time node. By comparing the response consistency between different pathways within the same time window, the degree of deviation between each pathway in the dynamic change process is identified. The cumulative deviation of each pathway response in the time dimension is comprehensively calculated to characterize the difference in convergence between different propagation paths, thereby obtaining the convergence offset characterization result that reflects the intensity of deviation of multi-pathway response consistency. Based on the convergence offset characterization results, blood flow continuity and perfusion stability are jointly determined, and abnormal segments exceeding the stability constraint threshold are extracted and summarized to form an abnormal residual distribution. Based on the convergence offset characterization results, the deviation degree of different abnormal segments in multi-path response is uniformly evaluated, and flow stability, pressure stability, and perfusion continuity are used as joint determination criteria to comprehensively judge the hemodynamic stability state of each abnormal segment. When an abnormal segment shows continuous deviation in multiple pathways and cannot return to the stable range, it is determined that the segment does not meet the blood flow continuity and perfusion stability constraints. Finally, all segments that meet the abnormal determination conditions are summarized and organized to form an abnormal residual distribution reflecting the continuous distribution state of system abnormalities, which is used for disturbance source analysis and propagation path reconstruction.

[0024] S3: Decompose the sources of disturbance based on the abnormal residual distribution. Divide the disturbance into pump load changes, vascular compliance feedback and measurement hysteresis channels. Combine the time decay relationship and mutual influence of each disturbance component to determine the transmission relationship of the disturbance in each channel and reconstruct the disturbance propagation path.

[0025] The process of classifying disturbances in S3 according to pump load changes, vascular compliance feedback, and measurement hysteresis channels is as follows: Based on the abnormal residual distribution, the system disturbances are structurally analyzed and divided into pump load change disturbance components, vascular compliance feedback disturbance components, and measurement lag observation disturbance components. Based on the abnormal residual distribution, the hemodynamic response deviation signals within the corresponding time window are structurally analyzed and processed to extract the synchronous change characteristics of pump speed change, flow change, and pressure fluctuation within the abnormal interval. According to the differences in the influence mechanism of different sources on hemodynamic response, the sources of disturbance are divided into three categories: the first is the driving side disturbance component caused by pump load change, the second is the feedback side disturbance component caused by vascular compliance change, and the third is the measurement lag observation disturbance component caused by sampling delay and signal transmission delay. By classifying the differences in the time location and response pattern of different disturbances, the preliminary structural decomposition of the disturbance sources in the abnormal residual distribution is achieved. By combining the attenuation characteristics and mutual coupling effects of each perturbation component in the time dimension, the interaction strength between perturbation components is quantitatively characterized, forming a perturbation correlation characterization result. Based on the perturbation component division result, the change process of each perturbation component in the time dimension is continuously analyzed, focusing on extracting its amplitude attenuation trend, duration characteristics, and influence intensity on hemodynamic response. The interaction relationship between different perturbation components is analyzed to determine whether there are coupling phenomena such as superposition enhancement, mutual cancellation, or delayed transmission. In this process, the influence strength between each perturbation component is quantitatively expressed and summarized using a unified interaction strength characterization method, thus forming a perturbation correlation characterization result used to describe the interaction characteristics between different perturbation sources. Based on the disturbance correlation characterization results, the propagation path of the disturbance among the pump-driven channel, vascular response channel, and observation delay channel is inferred and analyzed to obtain the disturbance propagation path results. Based on the disturbance correlation characterization results, combined with the differences in time response characteristics of the pump-driven side, vascular response side, and signal observation side, the propagation direction and transmission sequence of the disturbance among different channels are analyzed. By tracing the path of the disturbance from pump input to blood flow response output step by step, the transmission dependency of the disturbance among each channel is identified. The consistency of the disturbance transmission intensity and propagation direction among different channels is comprehensively judged to infer the actual propagation path of the disturbance among the pump-driven channel, vascular response channel, and observation delay channel, thus obtaining the disturbance propagation path results used to describe the disturbance transmission mechanism inside the system.

[0026] The process of reconstructing the disturbance propagation path in S3 is as follows: Based on the perturbation propagation path results, a time-dependent transmission mapping relationship is constructed between the pump-driven channel, the vascular response channel, and the observation delay channel. Based on the perturbation propagation path results, the perturbation response data in the pump-driven channel, the vascular response channel, and the observation delay channel are aligned to a unified time reference, making the data from each channel comparable at the same time axis scale. Taking the output change of the pump-driven side as the starting reference, the corresponding hemodynamic response changes and observed signal changes are correlated and matched time-by-time to establish the input-response correspondence between different channels. A time sequence constraint is introduced so that the transmission relationship of the perturbation in different channels can be arranged according to the actual physiological response sequence, thus forming a time-dependent cross-channel transmission mapping relationship, providing a basic structure for subsequent propagation path reconstruction.

[0027] Based on the transfer mapping relationship, the consistency conditions of time decay, phase offset, and propagation direction for disturbances in each channel are introduced to perform constraint-driven consistency evolution screening and path convergence processing for the cross-channel disturbance propagation process. Based on the transfer mapping relationship, the consistency analysis of the dynamic change characteristics of disturbances in each channel during the propagation process is performed, and the characteristics of the decay trend, phase offset, and propagation direction changes of disturbances in the time dimension are extracted respectively. Among them, time decay consistency is used to reflect whether the disturbance has a synchronous decay trend in different channels, phase offset constraint is used to measure the response lag relationship of the disturbance between different channels, and propagation direction consistency is used to determine whether the disturbance propagation conforms to the physical direction of pump drive to blood flow response. Through the joint constraint screening of the above three types of constraints, inconsistent trajectories in the cross-channel disturbance propagation process are eliminated, and only the propagation trajectory that satisfies multiple constraint consistency is retained, realizing constraint-driven screening and convergence optimization of the disturbance propagation path candidate set. Cross-channel consistent propagation trajectories that meet the constraints are used as the constraint convergence results to form a perturbation propagation path. Based on the cross-channel consistent evolution screening results, propagation trajectories that meet the constraints of time decay consistency, phase offset, and propagation direction consistency are integrated and analyzed. Perturbation propagation segments that maintain continuous and consistent evolution characteristics in different channels are spliced ​​and normalized. In this process, cross-channel response continuity is used as the constraint basis, and propagation segments with small local deviations but consistent overall trends are fused and corrected to form a continuous and traceable propagation link structure. Finally, all cross-channel consistent propagation trajectories that meet the constraints are summarized to form a perturbation propagation path that can reflect the dynamic coupling relationship between pump drive, vascular response, and observation delay.

[0028] S4: Under the constraint of the disturbance propagation path, with the goal of maintaining injection stability and avoiding negative pressure, a trial adjustment with limited amplitude is applied to key nodes, and the change of disturbance in the propagation path before and after adjustment is tracked to construct the adjustment response feature set.

[0029] The process of applying amplitude-limited trial adjustments to key nodes in S4 is as follows: Based on the disturbance propagation path structure, the influence weight of each node in the hemodynamic response is evaluated, and key control nodes with high contribution to blood flow stability are selected. Based on the disturbance propagation path structure, the role of each node in the path in the hemodynamic response is analyzed point by point, and the correlation strength of pump speed change, blood flow response sensitivity, and pressure fluctuation response amplitude corresponding to each node are extracted. By comprehensively comparing the position level, response transmission depth, and influence range of each node on downstream blood flow changes in the disturbance propagation link, the degree of contribution of each node to the overall blood flow stability is evaluated, and nodes with significant amplification effect or key regulatory role in the disturbance propagation process are selected as key control nodes. Under the constraints of maintaining stable blood perfusion and avoiding the risk of negative pressure, exploratory adjustment inputs with limited amplitude are applied to key control nodes. Under the constraints of maintaining stable blood perfusion and avoiding the risk of negative pressure, exploratory adjustment inputs are applied to the selected key control nodes. The exploratory adjustment inputs are controlled in an amplitude-limited manner, that is, the upper limit of the adjustment amplitude is set according to the current blood flow baseline level and the historical safe operating range, so that the adjustment changes are kept within the physiological safety threshold range. The adjustment direction is determined according to the deviation trend of the current hemodynamic response, so that it can make directional corrections to the decrease in flow or pressure abnormality, while avoiding the introduction of new unstable fluctuations, thereby achieving low-disturbance exploratory intervention in the system. The evolution of the disturbance propagation path under the influence of trial regulation is monitored in real time to obtain the path structure shift characteristics and the redistribution of disturbance energy in each channel. Under the input of trial regulation, the change process of the disturbance propagation path structure is continuously monitored, focusing on the changes in the disturbance propagation sequence, propagation intensity and propagation path connection relationship in each channel. By comparing the connection changes between nodes in the path structure before and after regulation, the local shift characteristics of the disturbance propagation path are identified. At the same time, the distribution changes of disturbance in the pump-driven channel, vascular response channel and observation delay channel are analyzed to obtain the redistribution of disturbance energy in different channels, thereby reflecting the degree of influence of trial regulation on the overall propagation structure. By combining blood flow fluctuations, pressure change trends, and negative pressure risk indicators, the impact of trial regulation input on local blood flow stability is comprehensively assessed, generating evaluation results to characterize changes in disturbance propagation response. Based on hemodynamic response data before and after trial regulation, continuous comparative analysis is performed on the amplitude of blood flow fluctuations, pressure change trends, and negative pressure risk indicators to evaluate the effect of regulation input on local blood flow stability. Among these, flow fluctuations reflect the degree of perfusion stability, pressure change trends determine the system's dynamic response recovery capability, and negative pressure risk indicators assess the proximity to the safety boundary. By comprehensively considering the changes in the above indicators, the role of trial regulation in improving or disturbing blood flow stability is quantitatively evaluated, forming evaluation results to characterize the changes in disturbance propagation response and providing a basis for optimizing control strategies.

[0030] The process of constructing the regulation response feature set in S4 is as follows: A comparative analysis of the perturbation propagation paths before and after the trial adjustment was conducted to extract information on changes in path structure. When comparing the perturbation propagation paths before and after the trial adjustment, the connection relationships and propagation order changes of each channel node in the perturbation propagation path structure before and after adjustment were obtained respectively. By comparing the changes in the perturbation transmission path between the pump-driven channel, the vascular response channel and the observation delay channel within the same time window, the newly added, disappeared or reconstructed connection relationships in the path were identified, and the upstream and downstream correlation changes of key nodes in the path were marked, thereby extracting information on changes in path structure reflecting changes in the perturbation propagation structure under the adjustment effect. Based on the evaluation results, the reliability of the path structure change is weighted and corrected, and the difference in the distribution change of disturbance energy in each propagation channel is calculated to form a disturbance redistribution characteristic index. Based on the path structure change information, combined with the change in disturbance propagation intensity before and after the trial adjustment, the reliability of the path change results is weighted and corrected. The weights are set according to the stability of hemodynamic response and signal fluctuation amplitude. Then, the distribution of disturbance in different propagation channels is compared and analyzed, and the redistribution difference of disturbance energy before and after adjustment in the pump drive channel, vascular response channel and observation delay channel is calculated. By summarizing the disturbance change differences in each channel, a disturbance redistribution characteristic index is formed to characterize the degree of migration and redistribution of disturbance within the system. By combining the temporal variation characteristics of hemodynamic response and based on the flow and pressure assessment results in the evaluation results, the response difference vectors of flow and pressure before and after regulation are extracted. By combining the temporal variation characteristics of hemodynamic response, the changes in flow and pressure within the same time window before and after the trial regulation are compared and analyzed point by point. The flow response difference is used to reflect the degree of perfusion change, and the pressure response difference is used to reflect the changes in circulatory resistance and negative pressure risk. By unifying and organizing the difference in the magnitude of flow change, the difference in the trend of pressure change, and the difference in response lag before and after regulation, a difference description result that can characterize the degree of change in hemodynamic response is constructed, thus forming the response difference vectors of flow and pressure, which are used to characterize the direct impact of regulation on the blood flow system. By fusing and mapping information on path structure changes, perturbation redistribution characteristics, and response difference vectors, a multidimensional regulatory response feature is formed. Based on this information, the three types of information are correlated and fused, and features from different sources are aligned according to the same time window so that they have a corresponding relationship within the same regulatory cycle. The system propagation structure changes reflected by path structure changes, the energy migration changes reflected by perturbation redistribution, and the physiological output changes reflected by blood flow response differences are jointly mapped to form a multidimensional regulatory response feature that can simultaneously characterize structural changes, energy changes, and physiological response changes. Multidimensional regulatory response features are sequentially organized according to their temporal evolution to construct a regulatory response feature set. Based on the multidimensional regulatory response features, the regulatory response features within each time window are continuously arranged according to their temporal evolution. The structural changes, disturbance distribution changes, and blood flow response changes at different time points are sequentially organized. By maintaining the consistency of the temporal order, the system response at different stages before, during, and after regulation can form a continuous evolution record. Finally, a regulatory response feature set is constructed to describe the entire process of exploratory regulation, providing basic data support for hemodynamic stability evaluation and control strategy selection.

[0031] S5: Based on the regulatory response feature set, combined with hemodynamic stability criteria and individual physiological constraints, the convergence, oscillation amplitude and perfusion deviation of candidate control parameters are evaluated, and a regulatory strategy that satisfies the hemodynamic stability and physiological constraints is output.

[0032] The process of the regulation strategy in S5 that satisfies the hemodynamic stability and physiological constraints is as follows: A candidate control parameter space is constructed based on the regulation response feature set. Simulation mapping is performed on different parameter combinations to generate candidate control strategies. Based on the regulation response feature set, key feature variables related to hemodynamic changes are extracted, including flow rate change trends, pressure response offsets, and disturbance propagation path changes. These are used as the constraint basis for control parameter combinations. According to the blood pump speed adjustment range, pump load allowable range, and target perfusion flow range, a candidate control parameter space is constructed. Different parameter combinations are input into the hemodynamic response mapping process. By simulating the change trends of blood flow rate and pressure under different regulation inputs, corresponding candidate control strategies are generated for stability evaluation and screening. Based on hemodynamic stability criteria, dynamic convergence analysis is performed on the parameter combinations in the candidate control strategies to generate stability scores. Based on hemodynamic stability criteria, dynamic response analysis is performed on the parameter combinations in the candidate control strategies, focusing on whether the blood flow system exhibits continuous divergence, convergence, or oscillation under different regulatory inputs. By comparing the flow recovery rate, pressure recovery trend, and disturbance attenuation after regulation, stability evaluation results are assigned to each candidate control strategy. The evaluation focuses on whether blood flow maintains continuous perfusion and whether there is abnormal negative pressure or amplified fluctuations, thus generating stability scores that reflect the dynamic stability of the system. Based on the stability score results, the oscillation amplitude and periodic fluctuation characteristics of the blood flow response under different control strategies are calculated to form an oscillation assessment index. Based on the stability score results, the time dimension analysis of the hemodynamic response under the candidate control strategies is performed, focusing on extracting the fluctuation characteristics of flow and pressure over time. By observing whether there are periodic fluctuations in the blood flow after regulation, whether the amplitude increases, and whether the fluctuation gradually decays, the blood flow oscillation intensity and fluctuation duration are calculated. Combined with the pulsation period characteristics, non-physiological oscillations are identified and distinguished, thus forming an oscillation assessment index to characterize the degree of change in the stability of the blood flow response. By combining individual physiological constraints, blood perfusion deviations are constrained and corrected to generate corresponding risk assessment results. By combining individual physiological constraints, blood perfusion status under candidate control strategies is constrained and corrected. The physiological constraints include the target perfusion flow range, the allowable pressure range, and the safety boundary to avoid negative pressure operation. The deviation between the actual blood flow response and the target physiological parameters under different control strategies is calculated. The focus is on analyzing whether there are insufficient perfusion, local overperfusion, or abnormal pressure. By uniformly assessing the degree of deviation, a risk assessment result reflecting the individual's physiological safety risk level is generated. Based on stability scores, oscillation assessment indices, and risk assessment results, a multi-dimensional fusion decision is made to output a regulation strategy that satisfies hemodynamic stability and physiological constraints. Candidate control strategies are then comprehensively compared and analyzed from multiple dimensions, using stability to measure system dynamic controllability, oscillation indices to measure response smoothness, and risk assessment to measure compliance with physiological safety boundaries. Consistency judgment and priority ranking of the three evaluation results are used to select a control strategy that simultaneously satisfies hemodynamic stability, low oscillation response, and physiological safety constraints. Finally, the optimal candidate control strategy is output to the blood pump actuator for real-time adjustment of pump speed and operating parameters, achieving closed-loop control optimization.

[0033] Example 2: Please refer to Figure 2 A remote monitoring and parameter adjustment control system for a blood pump device, comprising: Phase segmentation module: Performs phase calibration on pump speed and blood flow response, and determines segmentation boundaries by combining pulsation period and inertial delay to complete non-uniform segmentation; Residual detection module: performs cross-segment consistency checks on the flow drop and pressure recovery sections, and calculates the convergence offset through multi-path reconstruction to extract abnormal residual distribution; Disturbance modeling module: Based on the distribution of abnormal residual disturbances, the source of disturbances is identified and classified according to pump load, vascular compliance and measurement hysteresis, and the disturbance propagation path is reconstructed; Trial adjustment module: Apply amplitude-limited adjustment to key nodes under propagation path constraints, track changes in disturbance propagation, and generate a set of adjustment response features; Strategy optimization module: Based on the characteristics of the regulatory response combined with blood flow stability and physiological constraints, control parameters are screened and regulatory strategies are output.

[0034] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0035] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for remote monitoring parameter adjustment and control of a blood pump device, characterized in that, Includes the following steps: The pump speed and hemodynamic response signals were collected, and phase calibration was performed by combining the pulsation period and fluid inertial delay. The segment boundaries were determined based on the changes in physiological rhythm perturbation, and adaptive segmentation processing was performed to construct non-uniform response segments. For the alternating segments of flow drop and pressure fall in the flow response expression structure, cross-segment consistency is checked, and the convergence offset of the homologous response is calculated by multi-path reconstruction. Abnormal segments that do not meet the requirements of blood flow continuity and perfusion stability are extracted to form abnormal residual distribution. The sources of disturbance are decomposed based on the abnormal residual distribution. The disturbances are divided into pump load changes, vascular compliance feedback and measurement hysteresis channels. The transmission relationship of the disturbances in each channel is determined by combining the time decay relationship and mutual influence of each disturbance component, and the disturbance propagation path is reconstructed. Under the constraint of the disturbance propagation path, with the goal of maintaining injection stability and avoiding negative pressure, a trial adjustment with limited amplitude is applied to key nodes, and the change of disturbance in the propagation path before and after adjustment is tracked to construct an adjustment response feature set. Based on the regulatory response feature set, combined with hemodynamic stability criteria and individual physiological constraints, the convergence, oscillation amplitude and perfusion deviation of candidate control parameters are evaluated, and a regulatory strategy that satisfies hemodynamic stability and physiological constraints is output.

2. The method for remote monitoring parameter adjustment and control of a blood pump device according to claim 1, characterized in that, The process of constructing a non-uniform response fragment is as follows: The pump speed signal and hemodynamic response signal are synchronized in time, and a phase consistency mapping relationship is established by combining the pulsation cycle characteristics and fluid inertial delay. Periodic phase calibration of hemodynamic response is performed based on phase consistency mapping relationship, and physiological rhythm perturbation change characteristics are extracted; Based on the characteristics of physiological rhythm perturbation changes, the boundaries of local fluctuation intervals on the time axis are identified, and segmented boundary candidates are constructed. Stability screening and error constraint optimization are performed on the candidate segment boundaries, and adaptive non-uniform segmentation is carried out to construct non-uniform response segments.

3. The method for remote monitoring parameter adjustment and control of a blood pump device according to claim 2, characterized in that, The process of checking cross-segment consistency is as follows: For the alternating flow drop and pressure fall sections in non-uniform response segments, a cross-segment hemodynamic continuity alignment relationship is constructed based on the hemodynamic response characteristics of adjacent segments; Based on the cross-segment hemodynamic continuity alignment relationship, the pump output response between adjacent segments is coupled and mapped with the hemodynamic feedback, and consistency matching analysis is performed on the changes in flow rate, pressure and rotation speed. Based on the consistency matching analysis results, the cross-segment response deviation index is calculated, and segments with deviations exceeding the threshold are identified and marked according to the blood perfusion stability constraints, forming cross-segment inconsistency abnormal segments.

4. The method for remote monitoring parameter adjustment and control of a blood pump device according to claim 3, characterized in that, The process of forming abnormal residual distribution is as follows: Based on cross-segment inconsistency abnormal segments, multi-pathway retrospective reconstruction of homologous hemodynamic response is performed to generate a corresponding set of multi-pathway response trajectories. Based on the set of multi-path response trajectories, deviation analysis is performed on the consistency of hemodynamic response under different pathways, and the convergence offset between trajectories is calculated to characterize the trajectories. Based on the convergence offset characterization results, blood flow continuity and perfusion stability are jointly determined, and abnormal segments exceeding the stability constraint threshold are extracted and summarized to form an abnormal residual distribution.

5. The method for remote monitoring parameter adjustment and control of a blood pump device according to claim 4, characterized in that, The process of classifying disturbances according to pump load changes, vascular compliance feedback, and measurement hysteresis channels is as follows: Based on the abnormal residual distribution, the system disturbance is structurally analyzed and divided into pump load change disturbance component, vascular compliance feedback disturbance component and measurement lag observation disturbance component. By combining the attenuation characteristics of each disturbance component in the time dimension and their mutual coupling influence, the interaction strength between disturbance components is quantitatively characterized, forming a disturbance correlation characterization result; Based on the perturbation correlation characterization results, the propagation path of the perturbation in the pump drive channel, vascular response channel and observation delay channel is inferred and analyzed, and the perturbation propagation path results are obtained.

6. The method for remote monitoring parameter adjustment and control of a blood pump device according to claim 5, characterized in that, The process of reconstructing the disturbance propagation path is as follows: Based on the perturbation propagation path results, a transitive mapping relationship with time-dependent constraints is constructed between the pump-driven channel, the vascular response channel, and the observation delay channel; Based on the transfer mapping relationship, the consistency conditions of time decay, phase offset and propagation direction of the disturbances in each channel are introduced to perform constraint-driven consistency evolution screening and path convergence processing on the cross-channel disturbance propagation process. The cross-channel consistent propagation trajectory that satisfies the constraints is used as the constraint convergence result to form the perturbation propagation path.

7. The method for remote monitoring parameter adjustment and control of a blood pump device according to claim 6, characterized in that, The process of applying amplitude-limited trial adjustments to key nodes is as follows: Based on the disturbance propagation path structure, the influence weight of each node in the hemodynamic response is evaluated, and key control nodes that contribute significantly to blood flow stability are selected. Under the constraints of maintaining stable blood perfusion and avoiding the risk of negative pressure, exploratory regulatory inputs with limited amplitude are applied to key control nodes; The evolution of the disturbance propagation path under the trial regulation is monitored in real time to obtain the path structure offset characteristics and the redistribution of disturbance energy in each channel; By combining blood flow fluctuations, pressure change trends, and negative pressure risk indicators, the impact of trial-and-error input on local blood flow stability is comprehensively assessed, generating assessment results to characterize changes in the perturbation propagation response.

8. The method for remote monitoring parameter adjustment and control of a blood pump device according to claim 7, characterized in that, The process of constructing the regulation response feature set is as follows: Comparative analysis of the disturbance propagation paths before and after the trial adjustment was conducted to extract information on changes in path structure; Based on the evaluation results, the credibility of the path structure change is weighted and corrected, and the difference in the distribution change of disturbance energy in each propagation channel is calculated to form a disturbance redistribution characteristic index. Combining the temporal variation characteristics of hemodynamic response and based on the flow and pressure assessment results in the evaluation results, the response difference vectors of flow and pressure before and after regulation are extracted; The path structure change information, disturbance redistribution characteristic indicators and response difference vector are fused and mapped to form a multidimensional regulation response feature. The multidimensional regulatory response features are sequentially organized according to their temporal evolution to construct a regulatory response feature set.

9. A method for remote monitoring parameter adjustment and control of a blood pump device according to claim 8, characterized in that, The process of outputting a regulatory strategy that satisfies both hemodynamic stability and physiological constraints is as follows: A candidate control parameter space is constructed based on the regulation response feature set, and simulation mapping is performed on different parameter combinations to generate candidate control strategies. Based on hemodynamic stability criteria, dynamic convergence analysis is performed on the parameter combinations in the candidate control strategies to generate stability scores. Based on the stability score results, the oscillation amplitude and periodic fluctuation characteristics of blood flow response under different control strategies are calculated to form an oscillation assessment index. By combining individual physiological constraints, blood perfusion deviations are corrected and corresponding risk assessment results are generated. Based on stability scores, oscillation assessment indicators, and risk assessment results, a multidimensional fusion decision is made to output a regulatory strategy that satisfies hemodynamic stability and physiological constraints.

10. A remote monitoring parameter adjustment and control system for a blood pump device, applied to the method described in any one of claims 1-9, characterized in that, include: Phase segmentation module: Performs phase calibration on pump speed and blood flow response, and determines segmentation boundaries by combining pulsation period and inertial delay to complete non-uniform segmentation; Residual detection module: performs cross-segment consistency checks on the flow drop and pressure recovery sections, and calculates the convergence offset through multi-path reconstruction to extract abnormal residual distribution; Disturbance modeling module: Based on the distribution of abnormal residual disturbances, the source of disturbances is identified and classified according to pump load, vascular compliance and measurement hysteresis, and the disturbance propagation path is reconstructed; Trial adjustment module: Apply amplitude-limited adjustment to key nodes under propagation path constraints, track changes in disturbance propagation, and generate a set of adjustment response features; Strategy optimization module: Based on the characteristics of the regulatory response combined with blood flow stability and physiological constraints, control parameters are screened and regulatory strategies are output.

Images