Control method and device for a ventricular assist device

Abstract

The embodiment of the application provides a kind of control method and device of ventricular assist device, it is related to medical instrument technical field, above-mentioned method includes: determine the mathematical model that target ventricular assist device is coupled with sample heart environment, as simulation coupling model;Determine the target model based on the simulation coupling model, which represents the mapping path between the heart pressure change value and the control parameter of the target ventricular assist device;Obtain the actual heart pressure change value of the current target heart environment, determine the target control parameter of the target ventricular assist device based on the target model and actual heart pressure change value;According to the target control parameter, the target ventricular assist device is controlled, so that target ventricular assist device maintains the constant heart pressure change of target heart environment.Application of the scheme provided in this embodiment can realize the accurate control of ventricular assist device.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical devices, in particular to a control method and device of a ventricular assist device. BACKGROUND

[0002] The ventricular assist device is a device for providing support or assistance function for patients with heart-related diseases, such as heart failure patients, to assist the heart to pump blood to other parts of the body.

[0003] The control of the ventricular assist device is a major problem. Reasonable control is helpful for ventricular unloading, meeting cardiac output, pulse pressure difference and blood flow pulsatility; improper control may cause abnormal states such as suction, thrombus and hemolysis. Therefore, a control scheme of the ventricular assist device is urgently needed. SUMMARY

[0004] The purpose of the embodiments of the present application is to provide a control method and device of a ventricular assist device to accurately control the ventricular assist device. The specific technical solutions are as follows:

[0005] In a first aspect, the embodiments of the present application provide a control method of a ventricular assist device, the method comprising:

[0006] determining a mathematical model of coupling the target ventricular assist device with a sample heart environment as a simulation coupling model;

[0007] determining a target model representing a mapping path between a heart pressure change value and a control parameter of the target ventricular assist device based on the simulation coupling model;

[0008] obtaining an actual heart pressure change value of the target heart environment at present, and determining a target control parameter of the target ventricular assist device based on the target model and the actual heart pressure change value;

[0009] controlling the target ventricular assist device according to the target control parameter, so that the target ventricular assist device maintains the heart pressure change of the target heart environment constant.

[0010] In one embodiment of the present application, the above determining a target model representing a mapping path between a heart pressure change value and a control parameter of the target ventricular assist device based on the simulation coupling model comprises:

[0011] determining a plurality of first sample heart pressure change values based on the simulation coupling model, and determining a first sample control parameter corresponding to each first sample heart pressure change value, inputting each first sample heart pressure change value and the corresponding first sample control parameter into the simulation coupling model to obtain a second sample heart pressure change value in the adjacent unit time after each first sample heart pressure change value in the unit time.

[0012] The weight coefficients of the initial model are iteratively adjusted using a plurality of target training samples, wherein each target training sample comprises a first sample cardiac pressure change value, a corresponding first sample control parameter, and a corresponding second sample cardiac pressure change value, and the initial model is used to represent an initial mapping path between the cardiac pressure change value and the control parameter of the target ventricular assist device;

[0013] The initial model with the adjusted weight coefficients is determined as a target model.

[0014] In an embodiment of the present application, the above-mentioned iteratively adjusting the weight coefficients of the initial model using the target training samples comprises:

[0015] The target training samples are clustered to obtain a plurality of sample sets;

[0016] The actual fitting degree of the first sample control parameter is predicted based on the first sample cardiac pressure change value, the first sample control parameter, and the initial model in each target training sample included in the current sample set, wherein the actual fitting degree represents the degree of the control effect of the first sample control parameter as the control of the target ventricular assist device under the cardiac pressure change reflected by the first sample cardiac pressure change value;

[0017] The expected fitting degree of the first sample control parameter is calculated based on the second sample cardiac pressure change value and the first sample control parameter included in the current training sample;

[0018] The weight coefficients of the initial model are adjusted based on the difference between the actual fitting degree and the expected fitting degree, and in the case that the convergence condition is not met, the current sample set is updated, and the step of predicting the actual fitting degree of the first sample control parameter based on the first sample cardiac pressure change value, the first sample control parameter, and the initial model in each target training sample included in the current sample set is executed again based on the updated sample set until the convergence condition is met.

[0019] In an embodiment of the present application, the above-mentioned determining a plurality of first sample cardiac pressure change values based on the simulation coupling model and determining the first sample control parameter corresponding to each first sample cardiac pressure change value comprises:

[0020] A plurality of cardiac pressure change values are determined as the first sample cardiac pressure change values based on the simulation coupling model;

[0021] The control parameter corresponding to the first sample cardiac pressure change value is determined based on the simulation coupling model;

[0022] The first sample control parameter is determined based on the control parameter threshold, the random control parameter, and the determined control parameter.

[0023] In one embodiment of the present application, the first sample cardiac pressure change values are determined based on the simulation coupling model, comprising:

[0024] Each first sample cardiac pressure change value is determined according to the following expression :

[0025] ;

[0026] Wherein, is the average of the estimated cardiac pressure change values, is the average of the actually measured cardiac pressure change values, is the current parameter value in the simulation coupling model, represents the current change rate, is the rotational speed parameter value in the simulation coupling model, , , are all preset coefficients.

[0027] In one embodiment of the present application, the actual cardiac pressure change value represents the pressure difference between the aorta and the left ventricle, and the target ventricular assist device is controlled according to the target control parameter to maintain the cardiac pressure change of the target heart environment constant, comprising:

[0028] Based on the current cardiac pressure of the target heart environment, the starting time of the full support state of the target heart environment is predicted;

[0029] At the starting time of the predicted full support state, the target ventricular assist device is controlled according to the target control parameter to maintain the pressure difference between the aorta and the left ventricle in the target heart environment constant in the full support state.

[0030] In a second aspect, the embodiments of the present application provide a control device of a ventricular assist device, comprising:

[0031] A first model determination module is configured to determine a mathematical model of the coupling between the target ventricular assist device and a sample heart environment as a simulation coupling model;

[0032] A second model determination module is configured to determine a target model representing the mapping path between the cardiac pressure change value and the control parameter of the target ventricular assist device based on the simulation coupling model;

[0033] A control parameter determination module is configured to obtain the actual cardiac pressure change value of the target heart environment, and determine the target control parameter of the target ventricular assist device based on the target model and the actual cardiac pressure change value.

[0034] a device control module configured to control the target ventricular assist device according to the target control parameter, so that the target ventricular assist device maintains the cardiac pressure variation of the target cardiac environment constant.

[0035] In one embodiment of the present application, the second model determination module comprises:

[0036] a sample determination sub-module configured to determine a plurality of first sample cardiac pressure variation values based on the simulation coupling model, and determine a first sample control parameter corresponding to each first sample cardiac pressure variation value, input each first sample cardiac pressure variation value and the corresponding first sample control parameter into the simulation coupling model to obtain a second sample cardiac pressure variation value in a next unit time after each first sample cardiac pressure variation value in a unit time;

[0037] a model training sub-module configured to iteratively adjust a weight coefficient of an initial model using a plurality of target training samples, wherein each target training sample comprises a first sample cardiac pressure variation value, a corresponding first sample control parameter and a corresponding second sample cardiac pressure variation value, and the initial model is used to represent an initial mapping path between the cardiac pressure variation value and the control parameter of the target ventricular assist device;

[0038] a model determination sub-module configured to determine the initial model with the adjusted weight coefficient as a target model.

[0039] In one embodiment of the present application, the model training sub-module is specifically configured to cluster the target training samples to obtain a plurality of sample sets, predict an actual adaptation degree of the first sample control parameter based on the first sample cardiac pressure variation value, the first sample control parameter and the initial model in each target training sample included in a current sample set, wherein the actual adaptation degree represents a degree of the control effect of the first sample control parameter as the control parameter of the target ventricular assist device under the cardiac pressure variation reflected by the first sample cardiac pressure variation value, calculate an expected adaptation degree of the first sample control parameter based on the second sample cardiac pressure variation value and the first sample control parameter included in the current training sample, adjust the weight coefficient of the initial model based on a difference between the actual adaptation degree and the expected adaptation degree, update the current sample set in a case where a convergence condition is not met, and return to execute the step of predicting the actual adaptation degree of the first sample control parameter based on the first sample cardiac pressure variation value, the first sample control parameter and the initial model in each target training sample included in the current sample set based on the updated sample set until the convergence condition is met.

[0040] In one embodiment of the present application, the sample determining sub-module is specifically configured to determine a plurality of cardiac pressure change values as the first sample cardiac pressure change values based on the simulation coupling model; determine the control parameters corresponding to the first sample cardiac pressure change values based on the simulation coupling model; and determine the first sample control parameters based on the control parameter threshold, the random control parameters and the determined control parameters.

[0041] In one embodiment of the present application, the sample determining sub-module is specifically configured to determine each first sample cardiac pressure change value according to the following expression :

[0042] ;

[0043] wherein, is the average of the estimated cardiac pressure change values, is the average of the actually measured cardiac pressure change values, is the current parameter value in the simulation coupling model, represents the current change rate, is the rotation speed parameter value in the simulation coupling model, , , are all preset coefficients.

[0044] In one embodiment of the present application, the actual cardiac pressure change value represents the pressure difference between the aorta and the left ventricle, and the device control module is specifically configured to predict the starting time of the full support state of the target heart environment based on the current cardiac pressure of the target heart environment; and control the target ventricular assist device according to the target control parameters at the predicted starting time of the full support state, so that the target ventricular assist device maintains the pressure difference between the aorta and the left ventricle in the target heart environment constant in the full support state.

[0045] In a third aspect, an electronic device is provided, which includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete mutual communication through the communication bus;

[0046] The memory is used to store a computer program.

[0047] The processor is used to execute the program stored on the memory, and implement the method steps of the first aspect.

[0048] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the method steps of the first aspect.

[0049] From the above, it can be seen that, since the target control parameter is determined based on the real-time heart pressure change value and the target model, on the one hand, the target model is a model representing the mapping path between the heart pressure change value and the control parameter, which is determined based on the simulation coupling model of the target ventricular assist device and the sample heart environment in the simulation environment, so that the accuracy of the target control parameter determined based on the target model is high; on the other hand, since the real-time heart pressure change value reflects the real-time heart pressure change in the current heart environment, so that the target control parameter determined based on the real-time heart pressure change value can adapt to the pressure change of the current heart environment. In summary of the above two aspects, according to the scheme provided in the embodiment, the ventricular assist device can be controlled according to the relatively accurate control parameter to maintain the constant heart pressure change and realize the precise control under the condition of adapting to the pressure change of the current heart environment.

[0050] Of course, implementing any product or method of the present application does not necessarily require all the advantages described above to be achieved at the same time. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other embodiments can also be obtained by those skilled in the art based on these drawings.

[0052] Figure 1 A structural schematic diagram of an axial flow pump provided by an embodiment of the present application;

[0053] Figure 2 A flow schematic diagram of a first ventricular assist device control method provided by an embodiment of the present application;

[0054] Figure 3 A flow schematic diagram of a second ventricular assist device control method provided by an embodiment of the present application;

[0055] Figure 4 A flow schematic diagram of a third ventricular assist device control method provided by an embodiment of the present application;

[0056] Figure 5 A structural schematic diagram of a first ventricular assist device control device provided by an embodiment of the present application;

[0057] Figure 6 A structural schematic diagram of a second ventricular assist device control device provided by an embodiment of the present application;

[0058] Figure 7 A structural schematic diagram of an electronic device provided by an embodiment of the present application. Detailed Implementation

[0059] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0060] The ventricular assist device of this application can be attached to the apex of the left ventricle, right ventricle, or both ventricles of the heart. The ventricular assist device can be an axial flow pump, a centrifugal pump, or a magnetic levitation pump.

[0061] The following combination Figure 1 Taking an axial flow pump as an example, the structure of ventricular assist devices will be explained. Figure 1 A schematic diagram of the axial flow pump is shown, including a pig tail tube 106, a blood inlet 105, a blood channel 104, a blood outlet 103, a motor housing 102, and a conduit 101 connected and fixed in sequence. The motor housing 102 houses a motor, and the motor shaft passes through the motor housing and is fixedly connected to the axial flow impeller inside the blood channel 104.

[0062] The motor drives the axial flow impeller to rotate. Under this driving action, blood from the heart flows in from the blood inlet 105, passes through the blood flow channel 104, and flows out from the blood outlet 103.

[0063] exist Figure 1 In the structure shown, the motor is located inside the heart when the ventricular assist device is placed inside the patient. Alternatively, the motor can be connected to the impeller via a flexible drive shaft. In this case, the motor is located outside the heart when the ventricular assist device is placed inside the patient, thus reducing the size of the ventricular assist device. The motor drives the impeller to rotate via the flexible drive shaft, enabling the ventricular assist device to perform its auxiliary pumping function.

[0064] The execution entity of each embodiment of this application can be a controller of a ventricular assist device. The controller is used to detect relevant parameters of the ventricular assist device / patient and to control the operation of the ventricular assist device.

[0065] See Figure 2 , Figure 2 This is a flowchart illustrating a control method for a first type of ventricular assist device provided in an embodiment of this application. The method includes the following steps S201-S204.

[0066] Step S201: Determine the mathematical model that couples the target ventricular assist device with the sample cardiac environment, and use it as the simulation coupling model.

[0067] The target ventricular assist device refers to a ventricular assist device that needs to be controlled at present. The sample heart environment can be a heart environment used for simulating a heart failure patient.

[0068] The simulation coupling model is used for simulating the coupling of the target ventricular assist device and the sample heart environment in a simulation environment. The simulation coupling model can be pre-constructed or constructed in real time.

[0069] In constructing the simulation coupling model, in an embodiment, parameter values of each component included in the target ventricular assist device can be determined, and modeling is performed based on the determined parameters; parameter values of each part included in the sample heart environment can be determined, and modeling is performed based on the determined parameters. In the simulation digital environment, the established digital model of the target ventricular assist device and the digital model of the sample heart environment are coupled, thereby obtaining the simulation coupling model.

[0070] Step S202: determining a target model representing the mapping path between the heart pressure change value and the control parameter of the target ventricular assist device based on the simulation coupling model.

[0071] The target model is used for representing the mapping path between the heart pressure change value and the control parameter of the target ventricular assist device. The heart pressure change value reflects the amplitude of the change in the heart pressure, and can be the change in the aortic pressure, the change in the left ventricular pressure, the change in the difference between the aortic pressure and the left ventricular pressure, etc. The control parameter can be the rotational speed, the current, the PWM (Pulse Width Modulation) parameter value, etc. of the target ventricular assist device.

[0072] The target model can be a model based on a neural network model or a model based on a non-neural network model.

[0073] Since the target model is determined based on the simulation coupling model, and the simulation coupling model reflects the coupling of the target ventricular assist device and the sample heart environment, the mapping path between the heart pressure change value and the control parameter of the target ventricular assist device can be accurately determined based on the simulation coupling model, that is, the target model can be accurately determined.

[0074] In determining the target model, in an embodiment, a plurality of heart pressure change value setting values can be input into the simulation model, and a plurality of control parameters can be used to control the simulation model, the control parameter with the best control effect can be selected, thereby obtaining the corresponding relationship between the plurality of heart pressure change value and the control parameter, and the corresponding relationship can be fitted to obtain the target model.

[0075] Other embodiments of determining the target model can be referred to in the subsequent Figure 3Corresponding embodiments are not described in detail here.

[0076] Step S203: obtaining an actual cardiac pressure change value of the target cardiac environment at present, determining a target control parameter of the target ventricular assist device based on the target model and the actual cardiac pressure change value.

[0077] The target cardiac environment refers to the cardiac environment of the heart in which the target ventricular assist device is implanted. The actual cardiac pressure change value reflects the amplitude of the change in cardiac pressure of the target cardiac environment at present. The actual cardiac pressure change value can be the cardiac pressure change value in the current cardiac cycle, or the cardiac pressure change value in the current multiple cardiac cycles.

[0078] In obtaining the actual cardiac pressure change value, the cardiac pressure value of the target cardiac environment at each moment at present can be obtained, and the difference between the maximum value and the minimum value is calculated as the actual cardiac pressure change value. The cardiac pressure value can be obtained by using the pressure sensor inherited by the ventricular assist device.

[0079] In determining the control parameter, the actual cardiac pressure change value can be input into the target model. Since the target model represents the mapping path between the cardiac pressure change value and the control parameter, the target model can determine the control parameter corresponding to the actual cardiac pressure change value as the control parameter of the target ventricular assist device based on the mapping path.

[0080] Since the target model represents the mapping path between the cardiac pressure change value and the control parameter of the target ventricular assist device, the control parameter of the target ventricular assist device can be accurately determined by using the actual cardiac pressure change value at present.

[0081] Step S204: controlling the target ventricular assist device according to the target control parameter, so that the target ventricular assist device maintains the constant change in cardiac pressure of the target cardiac environment.

[0082] Since the target control parameter is determined based on the real-time cardiac pressure change value of the target cardiac environment at present, the target control parameter is more suitable for the current real-time cardiac pressure environment. Therefore, when the target ventricular assist device is controlled according to the target control parameter, it can better adapt to the current cardiac environment, thereby achieving precise control of the target ventricular assist device.

[0083] When the target ventricular assist device is controlled by using the target control parameter, the control target is to maintain the cardiac pressure variation of the target cardiac environment constant. When the target ventricular assist device is controlled, in an embodiment, variation data between the actual cardiac pressure variation value and the preset cardiac pressure variation value can be calculated, the control parameter corresponding to the variation data can be determined based on the target model as a control deviation, and the target control parameter of the target ventricular assist device can be calculated based on the adjusted parameter of the control parameter of the target ventricular assist device and the control deviation, so that the cardiac pressure variation value of the target ventricular assist device is always stable at the preset cardiac pressure variation value, thereby realizing the maintenance of the cardiac pressure variation of the target cardiac environment constant.

[0084] In an embodiment of the present application, when the actual cardiac pressure variation value represents the pressure difference between the aorta and the left ventricle, the starting time of the full support state of the current cardiac cycle of the target cardiac environment can be predicted based on the current cardiac pressure of the target cardiac environment; and at the predicted starting time of the full support state, the target ventricular assist device is controlled according to the target control parameter, so that the target ventricular assist device maintains the pressure difference between the aorta and the left ventricle in the target cardiac environment constant in the full support state.

[0085] The full support state represents the complete state of a cardiac cycle. The starting time of the full support state can be the starting time of the aortic systole.

[0086] When the starting time of the full support state is predicted, the first derivative and the multiple derivatives of the current cardiac pressure of the target cardiac environment can be calculated, and the starting time of the full support state can be determined based on the first derivative and the multiple derivatives. For example, when the starting time is the starting time of the aortic systole, the time when the first derivative reaches the second trough and the value of the multiple derivatives is positive can be determined as the starting time.

[0087] Since the target ventricular assist device maintains the pressure difference between the aorta and the left ventricle in the target cardiac environment constant in the full support state, the blood flow stability and the cardiac assist effect can be improved on the basis of providing full support blood flow support.

[0088] As can be seen from the above, since the target control parameter is determined based on the real-time heart pressure change value and the target model, on the one hand, the target model is a model representing the mapping path between the heart pressure change value and the control parameter, which is determined based on the simulation coupling model of the target ventricular assist device coupled with the sample heart environment in the simulation environment, so that the accuracy of the target control parameter determined based on the target model is high; on the other hand, since the real-time heart pressure change value reflects the real-time heart pressure change in the current heart environment, so that the target control parameter determined based on the real-time heart pressure change value can adapt to the pressure change of the current heart environment. In summary of the above two aspects, according to the scheme provided in the embodiment, the ventricular assist device can be controlled according to the relatively accurate control parameter to maintain the constant heart pressure change and realize the precise control under the condition of adapting to the pressure change of the current heart environment.

[0089] In the foregoing Figure 2 In step S202 of the corresponding embodiment, in addition to determining the target model in the manner mentioned above, the target model can also be determined in the manner mentioned below. Figure 3 Steps S302-S304 of the corresponding embodiment are implemented. Based on this, Figure 3 The flowchart of a second ventricular assist device control method provided by the embodiment of the present application is shown. The above method comprises the following steps S301-S306.

[0090] Step S301: Determine a mathematical model of the target ventricular assist device coupled with the sample heart environment as a simulation coupling model.

[0091] The above step S301 and the foregoing Figure 2 Step S201 of the corresponding embodiment is the same, and will not be described here.

[0092] Step S302: Based on the simulation coupling model, determine a plurality of first sample heart pressure change values, and determine a first sample control parameter corresponding to each first sample heart pressure change value, input each first sample heart pressure change value and the corresponding first sample control parameter into the simulation coupling model, to obtain a second sample heart pressure change value in the adjacent unit time after each first sample heart pressure change value in the unit time.

[0093] In determining the first sample heart pressure change value and the corresponding first sample control parameter, in one embodiment, a plurality of heart pressure change values can be determined as the first sample heart pressure change value in the coupling environment simulated by the simulation coupling model; then, the control parameter of the simulation coupling model running in the case of the heart pressure change value is determined as the sample control parameter, and the plurality of heart pressure change values are determined as the first sample heart pressure change value.

[0094] In one embodiment of the present application, each cardiac pressure variation value can be calculated according to the following expression:

[0095]

[0096] wherein, is the average of the estimated cardiac pressure variation values, is the average of the actually measured cardiac pressure variation values, is the current parameter value in the simulation coupling model, represents the rate of change of the current, is the rotational speed parameter value in the simulation coupling model, are all preset coefficients.

[0097] The first sample cardiac pressure variation value and the corresponding sample control parameter are input into the simulation coupling model to obtain the next unit time cardiac pressure variation value output by the simulation coupling model, wherein the next unit time is the adjacent unit time after the unit time corresponding to the input first sample cardiac pressure variation value.

[0098] In another embodiment, when determining the first sample cardiac pressure variation value and the corresponding first sample control parameter, a plurality of cardiac pressure variation values can be determined as the first sample cardiac pressure variation value based on the simulation coupling model, the control parameter corresponding to the first sample cardiac pressure variation value can be determined based on the simulation coupling model, and the first sample control parameter can be determined based on the control parameter threshold, the random control parameter, and the determined control parameter.

[0099] The cardiac pressure variation value can be set in the simulation coupling model. In the case of setting the cardiac pressure variation value, the control parameter run by the simulation coupling model can be determined as the control parameter corresponding to the first sample cardiac pressure variation value.

[0100] The control parameter threshold can include the minimum value and / or the maximum value of the control parameter, and the random control parameter refers to a randomly generated control parameter, which can be generated by using a preset noise data generation algorithm.

[0101] Due to the introduction of the random control parameter, the randomness of the selected control parameter is high, and the divergence is good, so that when the weight coefficient is adjusted based on the selected control parameter subsequently, the adjustment effect can better achieve the expected target.

[0102] ​​​In determining the first sample control parameter, a sum of the random control parameter and the determined control parameter can be calculated, a control parameter range formed by the control parameter threshold value and the calculated sum value is determined, and the first sample control parameter is selected from the control parameter range. One or more first sample control parameters can be selected, and the selection manner can be to select a control parameter with a preset sequence number in the control parameter range as the first sample control parameter.

[0103] Step S303: The weight coefficients of the initial model are iteratively adjusted by using the plurality of target training samples.

[0104] Each of the target training samples includes a first sample cardiac pressure change value, a corresponding first sample control parameter, and a corresponding second sample cardiac pressure change value.

[0105] The initial model is used to represent an initial mapping path between the cardiac pressure change value and the control parameter of the target ventricular assist device. Compared with the mapping path represented by the target model, the initial mapping path represented by the initial model has more noise data. Therefore, the initial model needs to be adjusted.

[0106] In adjusting the weight coefficients, each training sample can be iteratively input into the initial model according to the arrangement order of the target training samples. After a preset number of target training samples are input, the weight coefficients of the initial model are adjusted. In a case where the adjusted initial model does not satisfy a convergence condition, the current target training sample is updated, and the updated target training sample is used for training until the convergence condition is satisfied. The convergence condition can be a preset adjustment number, a preset weight coefficient range, etc. The adjustment manner can be referred to the subsequent detailed description. Figure 4 The corresponding embodiments are not described in detail.

[0107] Step S304: The initial model with the adjusted weight coefficients is determined as the target model.

[0108] Since the weight coefficients of the initial model are obtained by iteratively adjusting the training samples, the initial model with the adjusted weight coefficients, that is, the target model, can accurately reflect the mapping path between the cardiac pressure change value and the control parameter.

[0109] Step S305: An actual cardiac pressure change value of the target cardiac environment is obtained, and a target control parameter of the target ventricular assist device is determined based on the target model and the actual cardiac pressure change value.

[0110] Step S306: The target ventricular assist device is controlled according to the target control parameter, so that the target ventricular assist device maintains the cardiac pressure change of the target cardiac environment constant.

[0111] The steps S305-S306 are described above.Figure 2 The steps S203-S204 of the corresponding embodiment are the same, and will not be described here again.

[0112] As can be seen from the above, since the target model is obtained by iteratively adjusting the weight coefficients of the initial model using the training samples, and since the training samples contain the heart pressure changes and control parameters in adjacent time, the initial model can learn the correlation characteristics between the heart pressure changes and the control parameters during the adjustment, so that the initial model after adjusting the weight coefficients, that is, the target model, can more accurately reflect the mapping path between the heart pressure changes and the control parameters, and thus can more accurately control the target ventricular assist device.

[0113] The foregoing Figure 3 In the step S303 of the corresponding embodiment, when adjusting the weight coefficients of the initial model, the following can be performed Figure 4 The steps S403-S405 of the corresponding embodiment are implemented. Based on this, Figure 4 A flowchart of a third ventricular assist device control method provided by the embodiment of the present application is shown. The above method comprises the following steps S401-S408.

[0114] Step S401: Determine a mathematical model of coupling the target ventricular assist device with the sample heart environment as a simulation coupling model.

[0115] Step S402: Based on the simulation coupling model, determine a plurality of first sample heart pressure change values, and determine a first sample control parameter corresponding to each first sample heart pressure change value. Input each first sample heart pressure change value and the corresponding first sample control parameter into the simulation coupling model to obtain a second sample heart pressure change value in the adjacent unit time after each first sample heart pressure change value in the unit time.

[0116] The steps S401-S402 described above are the same as the foregoing Figure 3 The steps S301-S302 of the corresponding embodiment are the same, and will not be described here again.

[0117] Step S403: Cluster the target training samples to obtain a plurality of sample sets.

[0118] Each target training sample contains a first sample heart pressure change value, a corresponding first sample control parameter, and a corresponding second sample heart pressure change value.

[0119] In the clustering, the first sample heart pressure change values included in the target training samples can be clustered based on. Specifically, a pressure change difference threshold can be preset, and the target training samples included in a sample set are target training samples whose first sample heart pressure change values included in the target training samples satisfy the pressure change difference threshold. In this way, the number of sample sets is the same as the number of preset pressure change difference thresholds.

[0120] Of course, clustering can also be performed based on other information, such as the unit time corresponding to the first sample heart pressure change value, the first sample control parameter, the second sample heart pressure change value, and the like. This is not limited.

[0121] Step S404: Based on the first sample heart pressure change value, the first sample control parameter, and the initial model in each target training sample included in the current sample set, the actual adaptation degree of the first sample control parameter is predicted.

[0122] The actual adaptation degree represents the degree of the control effect of the first sample control parameter as the control of the target ventricular assist device under the heart pressure change reflected by the first sample heart pressure change value.

[0123] The actual adaptation degree represents the degree of the control effect of the first sample control parameter as the control of the target ventricular assist device under the heart pressure change reflected by the first sample heart pressure change value. The higher the target adaptation degree, the better the control effect, and the lower the target adaptation degree, the worse the control effect.

[0124] In predicting the actual adaptation degree, in one embodiment, the initial control parameter corresponding to the first sample heart pressure change value can be determined based on the initial model, the correlation between the initial control parameter and the first sample control parameter is calculated, and the calculated correlation is determined as the actual adaptation degree.

[0125] In another embodiment, the actual adaptation degree S can be calculated according to the following expression:

[0126] ;

[0127] Wherein N represents the total number of target training samples included in the current sample set, i represents the sequence number of the target training samples included in the current sample set, represents the first sample heart pressure change value included in the i-th target training sample, represents the sample control parameter included in the i-th target training sample, represents the function of the initial model, represents a preset entropy prediction algorithm, represents a preset coefficient, represents a decay factor.

[0128] Step S405: calculating an expected fitness degree of the first sample control parameter based on the second sample cardiac pressure change value and the first sample control parameter contained in the current training sample.

[0129] The expected fitness degree represents the fitness degree reached by the expected first sample control parameter. In an embodiment, the expected fitness degree T can be calculated according to the following expression:

[0130] ;

[0131] wherein N represents the total number of target training samples contained in the current sample set, i represents the serial number of the target training samples contained in the current sample set, represents the first sample cardiac pressure change value contained in the i+1th target training sample, represents the sample control parameter contained in the ith target training sample, represents a preset parameter value calculation function, represents a preset entropy prediction algorithm, represents a preset coefficient, represents a decay factor.

[0132] Step S406: adjusting the weight coefficient of the initial model based on the difference between the actual fitness degree and the expected fitness degree, and updating the current sample set in the case that the convergence condition is not met, returning to execute step S404 based on the updated sample set until the convergence condition is met.

[0133] The difference between the actual fitness degree and the expected fitness degree reflects the deviation of each weight coefficient in the initial model. The greater the difference, the greater the coefficient deviation, and the smaller the difference, the smaller the coefficient deviation.

[0134] In adjusting the weight coefficient, in an embodiment, it can be judged whether the difference between the actual fitness degree and the expected fitness degree is greater than a preset threshold, and if yes, a loss value is calculated based on the difference using a preset loss function, and the weight coefficient is adjusted according to the target of minimizing the loss value. In adjusting the weight coefficient, the weight coefficient can be adjusted according to a preset adjustment step and adjustment direction.

[0135] After adjusting the weight coefficient, it is determined whether the current adjusted initial model meets the convergence condition, and if yes, the training is ended; and if not, the current training sample is updated to train the initial model until the convergence condition is met.

[0136] Step S407: determining the initial model after adjusting the weight coefficient as the target model.

[0137] Step S408: obtaining an actual cardiac pressure change value of the target cardiac environment at present, determining a target control parameter of the target ventricular assist device based on the target model and the actual cardiac pressure change value.

[0138] Step S409: controlling the target ventricular assist device according to the target control parameter, so that the target ventricular assist device maintains the cardiac pressure change of the target cardiac environment constant.

[0139] The steps S407-S409 described above are the same as the steps S304-S306 of the foregoing control method of the ventricular assist device, and will not be described here again. Figure 3 The steps S304-S306 of the corresponding embodiment are the same, and will not be described here again

[0140] As can be seen from the above, since the weight coefficient of the initial model is adjusted based on the difference between the actual fitting degree and the expected fitting degree of the first sample control parameter, the difference between the actual fitting degree and the expected fitting degree reflects the deviation of each weight coefficient in the initial model. Therefore, based on the above difference, the weight coefficient can be accurately adjusted.

[0141] Corresponding to the control method of the ventricular assist device described above, the embodiments of the present application also provide a control device of a ventricular assist device.

[0142] Referring to Figure 5 , Figure 5 The first control device of the ventricular assist device provided by the embodiments of the present application has the structure shown in the schematic diagram, and the device comprises:

[0143] The first model determination module 501 is configured to determine a mathematical model of the target ventricular assist device coupled with the sample cardiac environment as a simulation coupling model;

[0144] The second model determination module 502 is configured to determine a target model representing a mapping path between a cardiac pressure change value and a control parameter of the target ventricular assist device based on the simulation coupling model;

[0145] The control parameter determination module 503 is configured to obtain an actual cardiac pressure change value of the target cardiac environment at present, and determine a target control parameter of the target ventricular assist device based on the target model and the actual cardiac pressure change value;

[0146] The device control module 504 is configured to control the target ventricular assist device according to the target control parameter, so that the target ventricular assist device maintains the cardiac pressure change of the target cardiac environment constant.

[0147] As can be seen from the above, since the target control parameter is determined based on the real-time heart pressure change value and the target model, on the one hand, the target model is a model for representing the mapping path between the heart pressure change value and the control parameter, which is determined based on the simulation coupling model of the target ventricular assist device coupled with the sample heart environment in the simulation environment, so that the accuracy of the target control parameter determined based on the target model is higher; on the other hand, since the real-time heart pressure change value reflects the real-time heart pressure change in the current heart environment, so that the target control parameter determined based on the real-time heart pressure change value can adapt to the pressure change of the current heart environment. In summary of the above two aspects, according to the scheme provided in the embodiment, the ventricular assist device can be controlled according to the relatively accurate control parameter to maintain the constant heart pressure change and realize the precise control under the condition of adapting to the pressure change of the current heart environment.

[0148] Referring to Figure 6 , Figure 6 The second ventricular assist device control device provided in the embodiment of the present application is shown in the structure diagram, and the device comprises:

[0149] The first model determination module 601 is configured to determine a mathematical model of the target ventricular assist device coupled with the sample heart environment as a simulation coupling model;

[0150] The sample determination sub-module 602 is configured to determine a plurality of first sample heart pressure change values based on the simulation coupling model, and determine a first sample control parameter corresponding to each first sample heart pressure change value, input each first sample heart pressure change value and the corresponding first sample control parameter into the simulation coupling model to obtain a second sample heart pressure change value in the adjacent unit time after each first sample heart pressure change value in the unit time;

[0151] The model training sub-module 603 is configured to iteratively adjust the weight coefficient of the initial model by using a plurality of target training samples, wherein each target training sample comprises a first sample heart pressure change value, a corresponding first sample control parameter and a corresponding second sample heart pressure change value, and the initial model is used to represent the initial mapping path between the heart pressure change value and the control parameter of the target ventricular assist device;

[0152] The model determination sub-module 604 is configured to determine the initial model with the adjusted weight coefficient as the target model.

[0153] The control parameter determination module 605 is configured to obtain an actual heart pressure change value of the target heart environment at present, and determine the target control parameter of the target ventricular assist device based on the target model and the actual heart pressure change value;

[0154] The device control module 606 is configured to control the target ventricular assist device according to the target control parameter, so that the target ventricular assist device maintains the constant cardiac pressure change of the target cardiac environment.

[0155] As can be seen from the above, since the target model is obtained by iteratively adjusting the weight coefficients of the initial model based on the training samples, and since the training samples include the cardiac pressure change and the control parameter in adjacent time, the initial model can learn the correlation characteristics between the cardiac pressure change and the control parameter during the adjustment, so that the initial model after the adjustment of the weight coefficients, that is, the target model can more accurately reflect the mapping path between the cardiac pressure change and the control parameter, and thus can more accurately control the target ventricular assist device.

[0156] In an embodiment of the present application, the model training submodule 603 is specifically configured to cluster the target training samples to obtain a plurality of sample sets; predict an actual adaptation degree of the first sample control parameter based on the first sample cardiac pressure change value, the first sample control parameter, and the initial model in each target training sample included in the current sample set, wherein the actual adaptation degree represents the degree of the control effect of the first sample control parameter as the control of the target ventricular assist device under the cardiac pressure change reflected by the first sample cardiac pressure change value; calculate an expected adaptation degree of the first sample control parameter based on the second sample cardiac pressure change value and the first sample control parameter included in the current training sample; adjust the weight coefficients of the initial model based on the difference between the actual adaptation degree and the expected adaptation degree, and update the current sample set in the case that the convergence condition is not met, and return to execute the step of predicting the actual adaptation degree of the first sample control parameter based on the first sample cardiac pressure change value, the first sample control parameter, and the initial model in each target training sample included in the current sample set based on the updated sample set until the convergence condition is met.

[0157] As can be seen from the above, the weight coefficients of the initial model are adjusted based on the difference between the actual adaptation degree and the expected adaptation degree of the first sample control parameter, and the difference between the actual adaptation degree and the expected adaptation degree reflects the deviation of each weight coefficient in the initial model. Therefore, the weight coefficients can be accurately adjusted based on the above difference.

[0158] In an embodiment of the present application, the sample determination submodule 602 is specifically configured to determine a plurality of cardiac pressure change values as the first sample cardiac pressure change values based on the simulation coupling model; determine the control parameter corresponding to the first sample cardiac pressure change values based on the simulation coupling model; and determine the first sample control parameter based on the control parameter threshold, the random control parameter, and the determined control parameter.

[0159] Due to the introduction of the random control parameter, the randomness degree of the selected control parameter is high, and the divergence degree is good, so that when the weight coefficient is adjusted based on the selected control parameter, the adjustment effect can better achieve the expected target.

[0160] In an embodiment of the present application, the sample determination submodule 602 is specifically configured to determine each cardiac pressure change value according to the following expression :

[0161] ;

[0162] wherein, is the average value of the estimated cardiac pressure change value, is the average value of the actually measured cardiac pressure change value, is the current parameter value in the simulation coupling model, represents the current change rate, is the rotational speed parameter value in the simulation coupling model, , , are all preset coefficients.

[0163] In an embodiment of the present application, the actual cardiac pressure change value represents the pressure difference between the aorta and the left ventricle, and the device control module 606 is specifically configured to predict the starting time of the full support state of the target heart environment based on the current cardiac pressure of the target heart environment; at the predicted starting time of the full support state, the target ventricular assist device is controlled according to the target control parameter, so that the target ventricular assist device maintains the pressure difference between the aorta and the left ventricle in the target heart environment constant in the full support state.

[0164] Since the target ventricular assist device maintains the pressure difference between the aorta and the left ventricle in the target heart environment constant in the full support state, the blood flow stability and the heart assist effect can be improved on the basis of providing full support blood flow support.

[0165] Corresponding to the control method of the ventricular assist device, an electronic device is also provided in an embodiment of the present application.

[0166] Referring to Figure 7 , Figure 7 is a structural schematic diagram of an electronic device provided in an embodiment of the present application, which comprises a processor 701, a communication interface 702, a memory 703 and a communication bus 704, wherein the processor 701, the communication interface 702 and the memory 703 complete mutual communication through the communication bus 704,

[0167] The memory 703 is used for storing a computer program.

[0168] The processor 701 is configured to implement the control method of the ventricular assist device according to the embodiments of the present application when executing the program stored in the memory 703.

[0169] The communication bus mentioned in the electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or only one type of bus.

[0170] The communication interface is configured to communicate between the electronic device and other devices.

[0171] The memory can include a Random Access Memory (RAM) and can also include a Non-Volatile Memory (NVM), such as at least one disk memory. Optionally, the memory can also be at least one storage device located away from the aforementioned processor.

[0172] The processor mentioned above can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc. It can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component.

[0173] In another embodiment of the present application, a computer readable storage medium is also provided, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the control method of the ventricular assist device according to the embodiments of the present application.

[0174] In another embodiment of the present application, a computer program product containing instructions is also provided, and when the computer program product is executed on a computer, the computer is caused to implement the control method of the ventricular assist device according to the embodiments of the present application.

[0175] As can be seen from the above, by applying the scheme provided in the embodiment, since the target control parameter is determined based on the real-time heart pressure change value and the target model, on the one hand, the target model is a model representing the mapping path between the heart pressure change value and the control parameter, which is determined based on the simulation coupling model in which the target ventricular assist device is coupled with the sample heart environment in the simulation environment, so that the accuracy of the target control parameter determined based on the target model is higher; on the other hand, since the real-time heart pressure change value reflects the real-time heart pressure change in the current heart environment, so that the target control parameter determined based on the real-time heart pressure change value can adapt to the pressure change in the current heart environment. In combination of the above two aspects, according to the scheme provided in the embodiment, the ventricular assist device can be controlled according to the relatively accurate control parameter to maintain the heart pressure change constant and realize the precise control, while adapting to the pressure change in the current heart environment.

[0176] In the above embodiments, the implementation can be wholly or partially realized by software, hardware, firmware or any combination thereof. When realized by software, the implementation can be wholly or partially realized in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed by a computer, the computer instructions wholly or partially generate the processes or functions described in the embodiments of the present application. The computer can be a general purpose computer, a special purpose computer, a computer network or other programmable apparatus. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through a wired (for example, coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (for example, infrared, wireless, microwave, etc.) manner. The computer readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available medium can be a magnetic medium (for example, floppy disk, hard disk, magnetic tape), an optical medium (for example, DVD), or a semiconductor medium (for example, solid state disk (SSD)) and the like.

[0177] 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 a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0178] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for apparatus, electronic devices, and computer-readable storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0179] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A control device of a ventricular assist device, characterized in that, The device comprises: a first model determining module configured to determine a mathematical model of coupling of a target ventricular assist device with a sample heart environment as a simulation coupling model; a second model determining module configured to determine, based on the simulation coupling model, a target model representing a mapping path between a heart pressure change value and a control parameter of the target ventricular assist device; a control parameter determining module configured to obtain an actual heart pressure change value of the target heart environment at present, and determine a target control parameter of the target ventricular assist device based on the target model and the actual heart pressure change value; a device control module configured to control the target ventricular assist device according to the target control parameter, so that the target ventricular assist device maintains the heart pressure change of the target heart environment constant; the second model determining module comprises: a sample determining sub-module configured to determine a plurality of first sample heart pressure change values based on the simulation coupling model, and determine a first sample control parameter corresponding to each first sample heart pressure change value, input each first sample heart pressure change value and the corresponding first sample control parameter into the simulation coupling model to obtain a second sample heart pressure change value in a next unit time after each first sample heart pressure change value in a unit time; a model training sub-module configured to iteratively adjust a weight coefficient of an initial model by using a plurality of target training samples, wherein each target training sample comprises a first sample heart pressure change value, a corresponding first sample control parameter and a corresponding second sample heart pressure change value, and the initial model is used to represent an initial mapping path between a heart pressure change value and a control parameter of the target ventricular assist device; a model determining sub-module configured to determine the initial model with the adjusted weight coefficient as the target model.

2. The apparatus of claim 1, wherein, The model training sub-module is specifically configured to cluster the target training samples to obtain a plurality of sample sets, predict an actual fitting degree of the first sample control parameter based on the first sample heart pressure change value, the first sample control parameter and the initial model in each target training sample included in a current sample set, wherein the actual fitting degree represents a degree of control effect of the first sample control parameter as the control of the target ventricular assist device under the heart pressure change represented by the first sample heart pressure change value, calculate an expected fitting degree of the first sample control parameter based on the second sample heart pressure change value and the first sample control parameter included in the current training sample, adjust the weight coefficient of the initial model based on a difference between the actual fitting degree and the expected fitting degree, update the current sample set in a case where a convergence condition is not met, and return to execute the step of predicting the actual fitting degree of the first sample control parameter based on the first sample heart pressure change value, the first sample control parameter and the initial model in each target training sample included in the current sample set based on the updated sample set until the convergence condition is met.

3. The apparatus of claim 1 or 2, wherein, The sample determination submodule is specifically configured to determine a plurality of heart pressure change values as first sample heart pressure change values based on the simulation coupling model; and determine a control parameter corresponding to the first sample heart pressure change values based on the simulation coupling model; The first sample control parameter is determined based on the control parameter threshold, the random control parameter, and the determined control parameter.

4. The apparatus of claim 3, wherein, The sample determination submodule is specifically configured to determine each heart pressure change value according to the following expression : ； wherein is an average of the estimated heart pressure variation values, is an average of the actually measured heart pressure variation values, is a current parameter value in the simulation coupling model, represents a current variation rate, is a rotational speed parameter value in the simulation coupling model, , , are all preset coefficients.

5. The apparatus of claim 1 or 2, wherein, The actual heart pressure change value represents a pressure difference between the aorta and the left ventricle, and the device control module is specifically configured to predict a starting time of a full support state of the target heart environment based on a current heart pressure of the target heart environment; and control the target ventricular assist device according to the target control parameter at the predicted starting time of the full support state, so that the target ventricular assist device maintains the pressure difference between the aorta and the left ventricle in the target heart environment constant in the full support state.

Images