Control method and system for balloon counterpulsation system in aorta

Abstract

The invention provides a control method and system of an in-aorta balloon counterpulsation system, and relates to the technical field of medical system control. According to the method provided by the invention, through the pre-trained neural network model, the target cardiac cycle of the target user can be quickly and accurately determined according to the historical electrocardiogram monitoring data of the target user under the condition that the target user has arrhythmia or the electrocardiogram signal quality is poor; according to the embodiment of the invention, the target cardiac cycle of the IABP system is obtained, and then inflation and deflation control of the air bag of the IABP system is realized according to the target cardiac cycle, compared with a mode of controlling the IABP system through electrocardio triggering in related technologies, the method provided by the embodiment of the invention can effectively improve the safety and effectiveness of the IABP system, and the use risk of a user is reduced.

Description

Technical Field

[0001] The present invention relates to the field of medical system control technology, and in particular to a control method and system for an intra-aortic balloon counterpulsation system. Background Art

[0002] The Intra-Aortic Balloon Pump (IABP) system is a key component used in interventional treatment of cardiovascular diseases. The IABP system controls the inflation and deflation of the balloon, thereby controlling the blood flow through the user's aorta to assist the heart through counterpulsation.

[0003] At present, the IABP system assists the heart through ECG triggering, which can be understood as synchronizing the inflation and deflation process of the airbag based on the electrophysiological activity of the user's heart. In each cardiac cycle of the heart, a series of ECG signals are generated. By monitoring the ECG signals, the IABP system can accurately identify the contraction and relaxation phases of the heart, and control the inflation and deflation of the airbag of the IABP system accordingly. However, this ECG triggering method has certain defects and shortcomings. In the case of arrhythmia or poor ECG signal quality, the ECG triggering method may not accurately identify the user's cardiac cycle, resulting in the inflation and deflation process of the airbag being out of sync with the user's heart working process. In addition, since the ECG signal may be affected by factors such as electromagnetic interference, resulting in signal distortion or false triggering, which in turn leads to failure of the IABP system control, the IABP system has problems of poor safety and effectiveness.

[0004] Therefore, there is an urgent need for an inflation control method and system for the aortic balloon counterpulsation bag, which can quickly and accurately determine the user's cardiac cycle when the user has arrhythmia or poor ECG signal quality, and then control the inflation and deflation of the IABP system's bag based on the cardiac cycle, thereby improving the safety and effectiveness of the IABP system. Summary of the invention

[0005] The embodiments of the present invention provide an inflation control method and system for an aortic balloon counterpulsation balloon, which can quickly and accurately determine the user's cardiac cycle when the user has arrhythmia or poor electrocardiogram signal quality, and then control the inflation and deflation of the balloon of the IABP system based on the cardiac cycle, thereby improving the safety and effectiveness of the IABP system.

[0006] To achieve the above object, the embodiments of the present invention adopt the following technical solutions:

[0007] In a first aspect, a control method for an intra-aortic balloon counterpulsation system is provided, which is applied to the intra-aortic balloon counterpulsation system. The intra-aortic balloon counterpulsation system includes a gas cylinder, a voltage regulator, an electromagnetic valve, a processor, a motor, a cylinder, an airbag and an electrocardiogram monitoring module; the gas cylinder, the cylinder and the airbag are connected by an air path; the electromagnetic valve is used to control the gas in the gas cylinder to enter the airbag and the cylinder, the voltage regulator is arranged between the gas cylinder and the electromagnetic valve, and the voltage regulator is used to control the gas pressure output by the gas cylinder within a preset gas pressure threshold range; the processor is electrically connected to the motor and the electrocardiogram monitoring module respectively, and the motor is connected to the cylinder; the above method includes: the electrocardiogram monitoring module obtains the electrocardiogram signal of the target user, and sends the target user to the processor the processor obtains historical ECG monitoring data of the target user when the amplitude of the ECG signal of the target user is less than a preset amplitude or the waveform quality coefficient corresponding to the ECG signal of the target user is less than a preset waveform quality coefficient, the historical ECG monitoring data including ECG signals corresponding to multiple cardiac cycles of the target user within a target time period; the processor inputs the historical ECG monitoring data into a pre-trained neural network model to predict a target cardiac cycle of the target user; the processor generates a control instruction according to the target cardiac cycle of the target user; the processor controls the operation of the motor according to the control instruction, so that the motor adjusts the internal pressure of the cylinder to complete the contraction and expansion operation of the airbag.

[0008] In a possible implementation of the first aspect, the neural network model includes a classifier, a first recurrent neural network and a second recurrent neural network, and the classifier is connected to the first recurrent neural network and the second recurrent neural network, respectively; the classifier is used to determine the category corresponding to the historical ECG monitoring data, the category including normal heart rate or abnormal heart rate; and is used to send the historical ECG monitoring data classified as normal heart rate to the first recurrent neural network, and is also used to send the historical ECG monitoring data classified as abnormal heart rate to the second recurrent neural network; the first recurrent neural network is used to output the corresponding target cardiac cycle based on the input historical ECG monitoring data classified as normal heart rate; the second recurrent neural network is used to output the corresponding target cardiac cycle based on the input historical ECG monitoring data classified as abnormal heart rate.

[0009] In a possible implementation of the first aspect, before the processor inputs historical ECG monitoring data into a pre-trained neural network model to predict the target cardiac cycle of the target user, the method further includes: the processor acquires training data, the training data including ECG signals corresponding to multiple cardiac cycles of each user among multiple users, and a category corresponding to each user; the processor samples the training data to obtain multiple training samples and a category corresponding to each training sample, each training sample including feature data and label data, the feature data being the ECG signals of the first n-1 cardiac cycles among the ECG signals of n consecutive cardiac cycles of any user, and the label data being the ECG signals of the nth cardiac cycle among the ECG signals of n consecutive cardiac cycles; the processor trains a classifier based on the multiple training samples and the category corresponding to each training sample to obtain a trained classifier; the processor trains a first recurrent neural network based on training samples with a category of normal heart rate to obtain a trained first recurrent neural network; the processor trains a second recurrent neural network based on training samples with a category of abnormal heart rate to obtain a trained second recurrent neural network, wherein the loss functions of the first recurrent neural network and the second recurrent neural network are root mean square error functions.

[0010] In a possible implementation of the first aspect, the intra-aortic balloon counterpulsation system also includes a blood pressure monitoring module and a display module, the blood pressure monitoring module is electrically connected to the processor, and the display module is electrically connected to the processor; after the processor controls the operation of the motor according to the control instruction to adjust the internal pressure of the cylinder and completes the contraction and expansion operation of the airbag, the method also includes: the blood pressure monitoring module obtains the arterial blood pressure value of the target user and sends the arterial blood pressure value of the target user to the processor; the processor generates a first alarm message when the arterial blood pressure value of the target user is not within a preset blood pressure threshold range; the display module displays the first alarm message, and the first alarm message is used to prompt that the arterial blood pressure value of the target user is abnormal.

[0011] In a possible implementation of the first aspect, the intra-aortic balloon counterpulsation system also includes an electronic pressure relief valve and multiple pressure sensors, which are arranged on the air path between the cylinder and the airbag, and are electrically connected to the processor; the above method also includes: the pressure sensor obtains the pressure value inside the airbag, and sends the pressure value inside the airbag to the processor; the processor controls the electronic pressure relief valve to open when the pressure value inside the airbag is greater than a preset pressure threshold; the above method also includes: the processor generates a second alarm message when the difference between multiple pressure values ​​between the cylinder and the airbag obtained by multiple sensors exceeds a preset pressure difference; the display module displays the second alarm message, and the second alarm message is used to indicate that the pressure sensor is abnormal.

[0012] In a possible implementation of the first aspect, the intra-aortic balloon counterpulsation system also includes a mechanical pressure relief valve, which is arranged on the air path between the cylinder and the airbag, and is used to open when the pressure value inside the airbag is greater than a preset pressure threshold.

[0013] The beneficial effects of the present invention are as follows: the method provided by the present invention can, through the pre-trained neural network model, quickly and accurately determine the target cardiac cycle of the target user according to the historical ECG monitoring data of the target user when the target user has arrhythmia or poor ECG signal quality, and then realize the inflation and deflation of the airbag of the IABP system according to the target cardiac cycle. Compared with the control method triggered by ECG in the related art, the method provided by the embodiment of the present invention can effectively improve the safety and effectiveness of the IABP system and reduce the user's use risk. In addition, the method provided by the embodiment of the present invention can control the gas output of the gas cylinder within the preset gas pressure threshold range by setting a voltage stabilizer, and then can accurately control the amount of gas output by the gas cylinder, so that the size of the contraction and expansion of the airbag can be controlled more accurately. Finally, the method provided by the embodiment of the present invention can realize accurate early warning of the pressure value inside the airbag in the event of failure of the electronic pressure relief valve by setting a mechanical pressure relief valve, thereby avoiding unnecessary danger to the user.

[0014] In a second aspect, an embodiment of the present invention provides an intra-aortic balloon counterpulsation system, which includes a gas cylinder, a voltage regulator, an electromagnetic valve, a processor, a motor, a cylinder, an airbag and an electrocardiogram monitoring module; the gas cylinder, the cylinder and the airbag are connected by an air circuit; the electromagnetic valve is used to control the gas in the gas cylinder to enter the airbag and the cylinder, the voltage regulator is arranged between the gas cylinder and the electromagnetic valve, and the voltage regulator is used to control the gas pressure output by the gas cylinder within a preset gas pressure threshold range; the processor is electrically connected to the motor and the electrocardiogram monitoring module respectively, and the motor is connected to the cylinder; the electrocardiogram monitoring module is used to: obtain the electrocardiogram signal of the target user, and send the electrocardiogram signal of the target user to the processor; the processor is used to: when the amplitude of the electrocardiogram signal of the target user is less than the preset amplitude or the waveform quality coefficient corresponding to the electrocardiogram signal of the target user is less than the preset waveform quality coefficient, obtain the historical electrocardiogram monitoring data of the target user, the historical electrocardiogram monitoring data including the electrocardiogram signals corresponding to multiple cardiac cycles of the target user within the target time period; input the historical electrocardiogram monitoring data into A pre-trained neural network model predicts a target cardiac cycle of a target user; generates control instructions according to the target cardiac cycle of the target user; controls the operation of a motor according to the control instructions so that the motor adjusts the internal pressure of the cylinder to complete the contraction and expansion of the airbag; wherein the neural network model includes a classifier, a first recurrent neural network and a second recurrent neural network, and the classifier is connected to the first recurrent neural network and the second recurrent neural network, respectively; the classifier is used to determine the category corresponding to the historical ECG monitoring data, the category including normal heart rate or abnormal heart rate; and is used to send the historical ECG monitoring data classified as normal heart rate to the first recurrent neural network, and is also used to send the historical ECG monitoring data classified as abnormal heart rate to the second recurrent neural network; the first recurrent neural network is used to output the corresponding target cardiac cycle according to the input historical ECG monitoring data classified as normal heart rate; the second recurrent neural network is used to output the corresponding target cardiac cycle according to the input historical ECG monitoring data classified as abnormal heart rate.

[0015] In a possible implementation of the second aspect, before the processor is used to input historical ECG monitoring data into a pre-trained neural network model to predict the target cardiac cycle of the target user, the processor is specifically used to: obtain training data, the training data including ECG signals corresponding to multiple cardiac cycles of each user among multiple users, and the category corresponding to each user; sample the training data to obtain multiple training samples and the category corresponding to each training sample, each training sample including feature data and label data, the feature data being the ECG signals of the first n-1 cardiac cycles among the ECG signals of n consecutive cardiac cycles of any user, and the label data being the ECG signals of the nth cardiac cycle among the ECG signals of n consecutive cardiac cycles; train a classifier based on the multiple training samples and the category corresponding to each training sample to obtain a trained classifier; train a first recurrent neural network based on training samples with a category of normal heart rate to obtain a trained first recurrent neural network; the processor trains a second recurrent neural network based on training samples with a category of abnormal heart rate to obtain a trained second recurrent neural network, wherein the loss function of the first recurrent neural network and the second recurrent neural network is a root mean square error function.

[0016] In a possible implementation of the second aspect, the intra-aortic balloon counterpulsation system also includes a blood pressure monitoring module and a display module, the blood pressure monitoring module is electrically connected to the processor, and the display module is electrically connected to the processor; after the processor is used to control the operation of the motor according to the control instruction to adjust the internal pressure of the cylinder and complete the contraction and expansion operation of the airbag, the blood pressure monitoring module is used to: obtain the arterial blood pressure value of the target user, and send the arterial blood pressure value of the target user to the processor; the processor is also used to: generate a first alarm message when the arterial blood pressure value of the target user is not within a preset blood pressure threshold range; the display module is used to: display the first alarm message, and the first alarm message is used to prompt that the arterial blood pressure value of the target user is abnormal.

[0017] In a possible implementation of the second aspect, the intra-aortic balloon counterpulsation system also includes an electronic pressure relief valve and multiple pressure sensors, which are arranged on the air path between the cylinder and the airbag, and are electrically connected to the processor; the pressure sensor is used to: obtain the pressure value inside the airbag, and send the pressure value inside the airbag to the processor; the processor is also used to: control the electronic pressure relief valve to open when the pressure value inside the airbag is greater than a preset pressure threshold; generate a second alarm message when the difference between multiple pressure values ​​between the cylinder and the airbag obtained by multiple sensors exceeds a preset pressure difference; the display module is also used to: display the second alarm message, and the second alarm message is used to indicate that the pressure sensor is abnormal; the intra-aortic balloon counterpulsation system also includes a mechanical pressure relief valve, which is arranged on the air path between the cylinder and the airbag, and is used to open when the pressure value inside the airbag is greater than a preset pressure threshold.

[0018] It can be understood that the beneficial effects that can be achieved by the system of the second aspect provided above can refer to the beneficial effects in the first aspect and any possible design thereof, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic structural diagram of an intra-aortic balloon counterpulsation system according to an embodiment of the present invention;

[0020] Figure 2 A schematic diagram of the structure of a neural network model shown in an embodiment of the present invention;

[0021] Figure 3 A schematic structural diagram of another intra-aortic balloon counterpulsation system according to an embodiment of the present invention;

[0022] Figure 4 A schematic structural diagram of another intra-aortic balloon counterpulsation system according to an embodiment of the present invention;

[0023] Figure 5 A schematic structural diagram of another intra-aortic balloon counterpulsation system according to an embodiment of the present invention;

[0024] Figure 6 A schematic flow chart of a control method of an intra-aortic balloon counterpulsation system according to an embodiment of the present invention;

[0025] Figure 7 The figure is a flow chart of another control method of an intra-aortic balloon counterpulsation system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The technical solution in the embodiment of the present invention will be described below in conjunction with the accompanying drawings in the embodiment of the present invention. Among them, in the description of the present invention, unless otherwise specified, " / " indicates that the objects associated before and after are in an "or" relationship, for example, A / B can represent A or B; the "or" in the present invention is only a description of the association relationship of associated objects, indicating that there can be three relationships, for example, A or B can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. And, in the description of the present invention, unless otherwise specified, "multiple" means two or more than two. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items.

[0027] In addition, in order to clearly describe the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, the words "first", "second", etc. are used to distinguish the same items or similar items with substantially the same functions and effects. Those skilled in the art can understand that the words "first", "second", etc. do not limit the quantity and execution order, and the words "first", "second", etc. do not necessarily limit the difference.

[0028] Meanwhile, in the embodiments of the present invention, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way for easy understanding.

[0029] The Intra-Aortic Balloon Pump (IABP) system is a key component used in interventional treatment of cardiovascular diseases. The IABP system controls the inflation and deflation of the balloon, thereby controlling the blood flow through the user's aorta to assist the heart through counterpulsation.

[0030] At present, the IABP system assists the heart through ECG triggering, which can be understood as synchronizing the inflation and deflation process of the airbag based on the electrophysiological activity of the user's heart. In each cardiac cycle of the heart, a series of ECG signals are generated. By monitoring the ECG signals, the IABP system can accurately identify the contraction and relaxation phases of the heart, and control the inflation and deflation of the airbag of the IABP system accordingly. However, this ECG triggering method has certain defects and shortcomings. In the case of arrhythmia or poor ECG signal quality, the ECG triggering method may not accurately identify the user's cardiac cycle, resulting in the inflation and deflation process of the airbag being out of sync with the user's heart working process. In addition, since the ECG signal may be affected by factors such as electromagnetic interference, resulting in signal distortion or false triggering, which in turn leads to failure of the IABP system control, the IABP system has problems of poor safety and effectiveness.

[0031] Therefore, there is an urgent need for an inflation control method and system for the aortic balloon counterpulsation balloon, which can quickly and accurately determine the user's cardiac cycle when the user has arrhythmia or poor ECG signal quality, and then control the inflation and deflation of the IABP system's balloon based on the cardiac cycle, thereby improving the safety and effectiveness of the IABP system.

[0032] In view of this, an embodiment of the present invention provides a control method for an intra-aortic balloon counterpulsation system, which is applied to the intra-aortic balloon counterpulsation system. The intra-aortic balloon counterpulsation system includes a gas cylinder, a voltage regulator, an electromagnetic valve, a processor, a motor, a cylinder, an airbag and an electrocardiogram monitoring module; the gas cylinder, the cylinder and the airbag are connected by an air path; the electromagnetic valve is used to control the gas in the gas cylinder to enter the airbag and the cylinder, the voltage regulator is arranged between the gas cylinder and the electromagnetic valve, and the voltage regulator is used to control the gas pressure output by the gas cylinder within a preset gas pressure threshold range; the processor is electrically connected to the motor and the electrocardiogram monitoring module respectively, and the motor is connected to the cylinder; the above method includes: the electrocardiogram monitoring module obtains the electrocardiogram signal of the target user, and sends it to the processor the target user's electrocardiogram signal; when the amplitude of the target user's electrocardiogram signal is less than a preset amplitude or the waveform quality coefficient corresponding to the target user's electrocardiogram signal is less than a preset waveform quality coefficient, the processor obtains the target user's historical electrocardiogram monitoring data, the historical electrocardiogram monitoring data including the electrocardiogram signals corresponding to multiple cardiac cycles of the target user within a target time period; the processor inputs the historical electrocardiogram monitoring data into a pre-trained neural network model to predict the target cardiac cycle of the target user; the processor generates a control instruction according to the target cardiac cycle of the target user; the processor controls the motor to run according to the control instruction, so that the motor adjusts the internal pressure of the cylinder to complete the contraction and expansion operation of the airbag.

[0033] The method provided in the embodiment of the present invention can quickly and accurately determine the target cardiac cycle of the target user according to the historical ECG monitoring data of the target user when the target user has arrhythmia or poor ECG signal quality through a pre-trained neural network model, and then realize the inflation and deflation of the airbag of the IABP system according to the target cardiac cycle. Compared with the control method triggered by ECG in the related art, the method provided in the embodiment of the present invention can effectively improve the safety and effectiveness of the IABP system and reduce the user's use risk. In addition, the method provided in the embodiment of the present invention can control the gas output of the gas cylinder within the preset gas pressure threshold range by setting a voltage stabilizer, and then can accurately control the amount of gas output by the gas cylinder, so that the size of the contraction and expansion of the airbag can be controlled more accurately. Finally, the method provided in the embodiment of the present invention can realize accurate early warning of the pressure value inside the airbag in the event of failure of the electronic pressure relief valve by setting a mechanical pressure relief valve, thereby avoiding unnecessary danger to the user.

[0034] In some embodiments, the control method of the intra-aortic balloon counterpulsation system provided by the embodiments of the present invention can be executed by the intra-aortic balloon counterpulsation system 100 (hereinafter referred to as the counterpulsation system 100). Figure 1 , Figure 1 The schematic diagram of the structure of an intra-aortic balloon counterpulsation system shown in an embodiment of the present invention, the counterpulsation system 100 includes a gas cylinder 110, a voltage regulator 120, a solenoid valve 130, a processor 140, a motor 150, a cylinder 160, an airbag 170 and an ECG monitoring module 180; the gas cylinder 110, the cylinder 160 and the airbag 170 are connected by an air circuit; the solenoid valve 130 is used to control the gas in the gas cylinder 110 to enter the airbag 170 and the cylinder 160, the voltage regulator 120 is arranged between the gas cylinder 110 and the solenoid valve 130, and the voltage regulator 120 is used to control the gas pressure output by the gas cylinder 110 within a preset gas pressure threshold range; the processor 140 is electrically connected to the motor 150 and the ECG monitoring module 180 respectively, and the motor 150 is connected to the cylinder 160, wherein, Figure 1 The dotted lines in the figure are used to represent the gas path for conveying gas, and the solid lines are used to represent the electrical connection.

[0035] The ECG monitoring module 180 is used to obtain the ECG signal of the target user and send the ECG signal of the target user to the processor 140; the processor 140 is used to: when the amplitude of the ECG signal of the target user is less than the preset amplitude or the waveform quality coefficient corresponding to the ECG signal of the target user is less than the preset waveform quality coefficient, obtain the historical ECG monitoring data of the target user, the historical ECG monitoring data including the ECG signals corresponding to multiple cardiac cycles of the target user within the target time period; input the historical ECG monitoring data into the pre-trained neural network model to predict the target cardiac cycle of the target user; generate a control instruction according to the target cardiac cycle of the target user; control the motor 150 to operate according to the control instruction, so that the motor 150 adjusts the internal pressure of the cylinder 160 to complete the contraction and expansion operation of the airbag 170;

[0036] The cardiac cycle of the target user is a time series consisting of the time intervals between two adjacent heart beats of the target user.

[0037] Exemplarily, the processor 140 may include one or more processing cores. The processor 140 uses various interfaces and lines to connect various parts within the counterpulsation system 100, and executes various functions of the counterpulsation system 100 and processes data by running or executing instructions, programs, code sets or instruction sets, and calling data. Optionally, the processor 140 can be implemented in at least one hardware form of a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processing (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA).

[0038] For details, see Figure 2 , Figure 2The present invention is a schematic diagram of the structure of a neural network model shown in an embodiment of the present invention, wherein the neural network model 200 includes a classifier 210, a first recurrent neural network 220 and a second recurrent neural network 230, wherein the classifier 210 is connected to the first recurrent neural network 220 and the second recurrent neural network 230, respectively; the classifier 210 is used to determine the category corresponding to the historical ECG monitoring data, wherein the category includes normal heart rate or abnormal heart rate; and is used to send the historical ECG monitoring data with the category of normal heart rate to the first recurrent neural network 220, and is also used to send the historical ECG monitoring data with the category of abnormal heart rate to the second recurrent neural network 230; the first recurrent neural network 220 is used to output the corresponding target cardiac cycle according to the input historical ECG monitoring data with the category of normal heart rate; the second recurrent neural network 230 is used to output the corresponding target cardiac cycle according to the input historical ECG monitoring data with the category of abnormal heart rate.

[0039] Exemplarily, the first recurrent neural network and the second recurrent neural network are long short-term memory networks (LSTM), which are recurrent neural networks used to process time series data. They can capture long-term dependencies and are suitable for processing long-term memory (long-term dependency) problems. At the same time, they avoid the common gradient vanishing and explosion problems of traditional recurrent neural networks during training. The long short-term memory network accurately predicts time series data through memory units and gating mechanisms.

[0040] In some embodiments, before the processor 140 is used to input historical ECG monitoring data into a pre-trained neural network model to predict the target cardiac cycle of the target user, the processor 140 is specifically used to: obtain training data, the training data including ECG signals corresponding to multiple cardiac cycles of each user among multiple users, and the category corresponding to each user; sample the training data to obtain multiple training samples and the category corresponding to each training sample, each training sample including feature data and label data, the feature data being the ECG signals of the first n-1 cardiac cycles among the ECG signals of n consecutive cardiac cycles of any user, and the label data being the ECG signals of the first n-1 cardiac cycles ... The data is the electrocardiogram signal of the nth cardiac cycle among the electrocardiogram signals of n consecutive cardiac cycles; the classifier 210 is trained based on multiple training samples and the category corresponding to each training sample to obtain a trained classifier; the first recurrent neural network 220 is trained based on the training samples with the category of normal heart rate to obtain the trained first recurrent neural network 220; the second recurrent neural network 230 is trained based on the training samples with the category of abnormal heart rate to obtain the trained second recurrent neural network 230, wherein the loss function of the first recurrent neural network 220 and the second recurrent neural network 230 is a root mean square error function.

[0041] The following takes the first recurrent neural network as an example to explain the training process of the first recurrent neural network and the second recurrent neural network. When the first recurrent neural network is LSTM, the training process of the first recurrent neural network includes a forward propagation stage and a back propagation stage. In the forward propagation stage, LSTM processes the input training samples step by step according to the time step and updates the model parameters (weights and biases). In the back propagation stage, LSTM calculates the gradient of the loss function relative to each weight and bias by the chain rule, and uses these gradients to update the weights and biases included in the model parameters.

[0042] It should be understood that the system provided by the embodiment of the present invention can effectively reduce the complexity of the neural network model by setting a classifier to determine the category corresponding to the historical ECG monitoring data of the target user, and then predicting the target cardiac cycle corresponding to the historical ECG monitoring data of different categories based on different recurrent neural networks. In turn, the computing resources required to deploy the neural network model are reduced. At the same time, since a recurrent neural network only needs to train one category of training samples, the training time of the neural network model can be reduced, thereby improving the model training efficiency.

[0043] In one possible implementation, see Figure 3 The counterpulsation system 100 also includes a blood pressure monitoring module 190 and a display module 191. The blood pressure monitoring module 190 is electrically connected to the processor 140, and the display module 191 is electrically connected to the processor 140. After the processor 140 is used to control the operation of the motor 150 according to the control instruction to adjust the internal pressure of the cylinder 160 and complete the contraction and expansion operation of the airbag 170, the blood pressure monitoring module 190 is used to: obtain the arterial blood pressure value of the target user and send the arterial blood pressure value of the target user to the processor 140; the processor 140 is also used to: generate a first alarm message when the arterial blood pressure value of the target user is not within the preset blood pressure threshold range; the display module 191 is used to: display the first alarm message, and the first alarm message is used to prompt that the arterial blood pressure value of the target user is abnormal.

[0044] The system provided by the embodiment of the present invention monitors the arterial blood pressure value of the target user. When the processor 140 is used to control the operation of the motor 150 according to the control instructions to adjust the internal pressure of the cylinder 160 and complete the contraction and expansion operations of the airbag 170, the life and health of the target user can be monitored in real time to avoid unnecessary danger to the target user and improve the safety of the system.

[0045] In some embodiments, see Figure 4The counterpulsation system 100 further includes an electronic pressure relief valve 192 and a plurality of pressure sensors 193, which are arranged on the air path between the cylinder 160 and the airbag 170, and are electrically connected to the processor 140; the pressure sensor 193 is used to obtain the pressure value inside the airbag 170 and send the pressure value inside the airbag 170 to the processor 140; the processor 140 is also used to control the electronic pressure relief valve 192 to open when the pressure value inside the airbag 170 is greater than a preset pressure threshold;

[0046] The system provided by the embodiment of the present invention can quickly and accurately identify a faulty pressure sensor by comparing the pressure values ​​obtained by multiple pressure sensors. For example, if the data difference between two different pressure sensors is too large, it means that one of the pressure sensors is faulty. The second alarm information prompts the operator to replace the pressure sensor in time, which can effectively ensure the stable detection of the gas path air pressure.

[0047] The processor 140 is also used to: generate a second alarm message when the difference between the multiple pressure values ​​between the cylinder and the airbag obtained by the multiple sensors exceeds a preset pressure difference; the display module is also used to: display the second alarm message, and the second alarm message is used to indicate that the pressure sensor is abnormal;

[0048] The system provided by the embodiment of the present invention can quickly and accurately identify a faulty pressure sensor by comparing the pressure values ​​obtained by multiple pressure sensors. For example, if the data difference between two different pressure sensors is too large, it means that one of the pressure sensors is faulty. The second alarm information prompts the operator to replace the pressure sensor in time, which can effectively ensure the stable detection of the gas path air pressure.

[0049] For further information, see Figure 5 The counterpulsation system 100 also includes a mechanical pressure relief valve 195, which is arranged on the air path between the cylinder 160 and the airbag 170. The mechanical pressure relief valve 195 is used to open when the pressure value inside the airbag 170 is greater than a preset pressure threshold.

[0050] The system provided by the embodiment of the present invention is provided with a mechanical pressure relief valve. When the pressure inside the airbag is too high and uncontrolled due to a system freeze or air line failure, if the airbag pressure exceeds the threshold set by the mechanical safety valve, the mechanical safety valve will automatically open to avoid the risk of the airbag being over-inflated or exploding, thereby fully ensuring the user's safety.

[0051] It is to be understood that the structure illustrated in this embodiment does not constitute a specific limitation on the counterpulsation system 100. In other embodiments of the present invention, the counterpulsation system 100 may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0052] The control method of the intra-aortic balloon counterpulsation system provided by the embodiment of the present invention is described below in conjunction with the accompanying drawings.

[0053] Figure 6 The flowchart of a control method of an intra-aortic balloon counterpulsation system provided by an embodiment of the present invention. Optionally, the method can be performed by the above-mentioned counterpulsation system 100. The method can include the following steps:

[0054] S1. The ECG monitoring module obtains the ECG signal of the target user and sends the ECG signal of the target user to the processor;

[0055] S2. The processor obtains historical ECG monitoring data of the target user when the amplitude of the ECG signal of the target user is less than a preset amplitude or the waveform quality coefficient corresponding to the ECG signal of the target user is less than a preset waveform quality coefficient.

[0056] Specifically, the historical ECG monitoring data includes ECG signals corresponding to multiple cardiac cycles of the target user within a target time period.

[0057] S3. The processor inputs the historical ECG monitoring data into a pre-trained neural network model to predict the target cardiac cycle of the target user.

[0058] In one possible implementation, the neural network model includes a classifier, a first recurrent neural network and a second recurrent neural network, and the classifier is connected to the first recurrent neural network and the second recurrent neural network, respectively; the classifier is used to determine the category corresponding to the historical ECG monitoring data, the category including normal heart rate or abnormal heart rate; and is used to send the historical ECG monitoring data classified as normal heart rate to the first recurrent neural network, and is also used to send the historical ECG monitoring data classified as abnormal heart rate to the second recurrent neural network; the first recurrent neural network is used to output the corresponding target cardiac cycle based on the input historical ECG monitoring data classified as normal heart rate; the second recurrent neural network is used to output the corresponding target cardiac cycle based on the input historical ECG monitoring data classified as abnormal heart rate.

[0059] In some embodiments, see Figure 7 Before the above S3, the method provided by the embodiment of the present invention further includes the following steps:

[0060] S31, the processor acquires training data, where the training data includes electrocardiogram signals corresponding to multiple cardiac cycles of each user among multiple users, and a category corresponding to each user;

[0061] S32: The processor performs sampling processing on the training data to obtain multiple training samples and a category corresponding to each training sample.

[0062] Each training sample includes feature data and label data, wherein the feature data is the ECG signal of the first n-1 cardiac cycles among the ECG signals of the consecutive n cardiac cycles of any user, and the label data is the ECG signal of the nth cardiac cycle among the ECG signals of the consecutive n cardiac cycles;

[0063] S33: The processor trains the classifier based on the multiple training samples and the category corresponding to each training sample to obtain a trained classifier.

[0064] S34. The processor trains the first recurrent neural network based on the training samples classified as normal heart rate to obtain a trained first recurrent neural network; the processor trains the second recurrent neural network based on the training samples classified as abnormal heart rate to obtain a trained second recurrent neural network.

[0065] Among them, the loss functions of the first recurrent neural network and the second recurrent neural network are root mean square error functions.

[0066] S4. The processor generates a control instruction according to the target cardiac cycle of the target user.

[0067] In a possible implementation, the above S4 specifically includes the following steps:

[0068] The processor determines an average value of a plurality of cardiac cycles included in the historical ECG monitoring data of the target user;

[0069] The processor generates a control instruction according to the target cardiac cycle of the target user when the difference between the target cardiac cycle of the target user and the average value of the plurality of cardiac cycles is less than a preset threshold.

[0070] When the difference between the target cardiac cycle of the target user and the average value of multiple cardiac cycles is greater than or equal to a preset difference, the processor determines the average value of multiple cardiac cycles as the target cardiac cycle of the target user, and generates a control instruction according to the target cardiac cycle of the target user.

[0071] The method provided in the embodiment of the present invention verifies the target cardiac cycle predicted by the neural network model based on the average value of multiple cardiac cycles included in the historical ECG monitoring data of the target user, and after successful verification, generates control instructions according to the target cardiac cycle of the target user, thereby effectively improving the safety of the counterpulsation system.

[0072] S5. The processor controls the operation of the motor according to the control instruction, so that the motor adjusts the internal pressure of the cylinder to complete the contraction and expansion operation of the airbag.

[0073] It can be seen from the above S1-S5 that the method provided in the embodiment of the present invention can, through a pre-trained neural network model, quickly and accurately determine the target cardiac cycle of the target user based on the historical ECG monitoring data of the target user when the target user has arrhythmia or poor ECG signal quality, and then realize the inflation and deflation of the airbag of the IABP system according to the target cardiac cycle. Compared with the control method through ECG triggering in the related art, the method provided in the embodiment of the present invention can effectively improve the safety and effectiveness of the IABP system and reduce the user's use risk.

[0074] In some embodiments, after the above S5, the method provided by the embodiment of the present invention also includes the following steps: the blood pressure monitoring module obtains the arterial blood pressure value of the target user and sends the arterial blood pressure value of the target user to the processor; the processor generates a first alarm message when the arterial blood pressure value of the target user is not within a preset blood pressure threshold range; the display module displays the first alarm message, and the first alarm message is used to prompt that the arterial blood pressure value of the target user is abnormal.

[0075] In some other embodiments, the method provided by the embodiment of the present invention further includes the following steps: the pressure sensor obtains the pressure value inside the airbag, and sends the pressure value inside the airbag to the processor; when the pressure value inside the airbag is greater than a preset pressure threshold, the processor controls the electronic pressure relief valve to open;

[0076] Optionally, the method provided by the embodiment of the present invention also includes the following steps: the processor generates a second alarm message when the difference between multiple pressure values ​​between the cylinder and the airbag obtained by multiple sensors exceeds a preset pressure difference; the display module displays the second alarm message, and the second alarm message is used to indicate that the pressure sensor is abnormal.

[0077] It should be understood that the specific description of the above optional methods can refer to the above system embodiments, which will not be repeated here. In addition, the explanation of any control method provided above and the description of the beneficial effects can refer to the above corresponding system embodiments, which will not be repeated here.

[0078] The embodiment of the present invention further provides a computer-readable storage medium, in which at least one computer instruction is stored, and the at least one computer instruction is loaded and executed by a processor to implement the methods of the above embodiments. For the explanation of the relevant contents and the description of the beneficial effects in any of the above-mentioned computer-readable storage media, reference can be made to the above-mentioned corresponding embodiments, which will not be repeated here.

[0079] The embodiment of the present invention further provides a chip. The chip integrates a control circuit and one or more ports for implementing the functions of the above-mentioned prediction system 100. Optionally, the functions supported by the chip can be referred to above and will not be described in detail here.

[0080] Those skilled in the art will appreciate that all or part of the steps of the above embodiments can be implemented by instructing the relevant hardware through a program, and the program can be stored in a computer-readable storage medium. The above-mentioned storage medium can be a read-only memory, a random access memory, etc. The above-mentioned processing unit or processor can be a central processing unit, a general-purpose processor, a specific circuit structure (application specific integrated circuit, ASIC), a microprocessor (digital signal processor, DSP), a field programmable gate array (field prograllable gatearray, FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof.

[0081] The embodiment of the present invention also provides a computer program product including instructions, when the instructions are run on a computer, the computer executes any one of the methods in the above embodiments. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., an SSD), etc.

[0082] It should be noted that the above-mentioned devices for storing computer instructions or computer programs provided in the embodiments of the present invention, such as but not limited to the above-mentioned memory, computer-readable storage medium and communication chip, etc., all have non-transitory. Those skilled in the art should be aware that in one or more of the above-mentioned examples, the functions described in the embodiments of the present invention can be implemented with hardware, software, firmware or any combination thereof. When implemented using software, these functions can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. Computer-readable storage media include computer storage media and communication media, wherein the communication medium includes any medium that is convenient for transmitting a computer program from one place to another. The storage medium can be any available medium that a general or special-purpose computer can access.

[0083] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A control method for an intra-aortic balloon counterpulsation system, characterized in that: Applicable to an intra-aortic balloon counterpulsation system, the intra-aortic balloon counterpulsation system comprises a gas cylinder, a voltage regulator, a solenoid valve, a processor, a motor, a cylinder, an airbag and an electrocardiogram monitoring module; the gas cylinder, the cylinder and the airbag are connected by a gas circuit; the solenoid valve is used to control the gas in the gas cylinder to enter the airbag and the cylinder, the voltage regulator is arranged between the gas cylinder and the solenoid valve, and the voltage regulator is used to control the gas pressure output by the gas cylinder within a preset gas pressure threshold range; The processor is electrically connected to the motor and the ECG monitoring module respectively, and the motor is connected to the cylinder; The method comprises: The ECG monitoring module acquires the ECG signal of the target user and sends the ECG signal of the target user to the processor; The processor acquires historical ECG monitoring data of the target user when the amplitude of the ECG signal of the target user is less than a preset amplitude or the waveform quality coefficient corresponding to the ECG signal of the target user is less than a preset waveform quality coefficient, wherein the historical ECG monitoring data includes ECG signals corresponding to multiple cardiac cycles of the target user within a target time period; The processor inputs the historical ECG monitoring data into a pre-trained neural network model to predict a target cardiac cycle of the target user; The processor generates a control instruction according to a target cardiac cycle of the target user; The processor controls the operation of the motor according to the control instruction, so that the motor adjusts the internal pressure of the cylinder to complete the contraction and expansion operation of the airbag.

2. The method according to claim 1, characterized in that The neural network model includes a classifier, a first recurrent neural network and a second recurrent neural network, and the classifier is connected to the first recurrent neural network and the second recurrent neural network respectively; the classifier is used to determine the category corresponding to the historical ECG monitoring data, and the category includes normal heart rhythm or abnormal heart rate; And it is used to send historical ECG monitoring data classified as normal heart rate to the first recurrent neural network, and is also used to send historical ECG monitoring data classified as abnormal heart rate to the second recurrent neural network; the first recurrent neural network is used to output the corresponding target cardiac cycle based on the input historical ECG monitoring data classified as normal heart rate; the second recurrent neural network is used to output the corresponding target cardiac cycle based on the input historical ECG monitoring data classified as abnormal heart rate.

3. The method according to claim 2, characterized in that Before the processor inputs the historical ECG monitoring data into a pre-trained neural network model to predict the target cardiac cycle of the target user, the method further includes: The processor acquires training data, the training data including electrocardiogram signals corresponding to multiple cardiac cycles of each user among multiple users, and a category corresponding to each user; The processor performs sampling processing on the training data to obtain a plurality of training samples and a category corresponding to each training sample, wherein each training sample includes feature data and label data, wherein the feature data is the electrocardiogram signal of the first n-1 cardiac cycles among the electrocardiogram signals of the consecutive n cardiac cycles of any user, and the label data is the electrocardiogram signal of the nth cardiac cycle among the electrocardiogram signals of the consecutive n cardiac cycles; The processor trains the classifier based on a plurality of training samples and a category corresponding to each training sample to obtain a trained classifier; The processor trains the first recurrent neural network based on training samples classified as normal heart rate to obtain a trained first recurrent neural network; the processor trains the second recurrent neural network based on training samples classified as abnormal heart rate to obtain a trained second recurrent neural network, wherein the loss functions of the first recurrent neural network and the second recurrent neural network are root mean square error functions.

4. The method according to claim 3, characterized in that The intra-aortic balloon counterpulsation system further includes a blood pressure monitoring module and a display module, wherein the blood pressure monitoring module is electrically connected to the processor, and the display module is electrically connected to the processor; after the processor controls the motor to operate according to the control instruction to adjust the internal pressure of the cylinder and completes the contraction and expansion operation of the airbag, the method further includes: The blood pressure monitoring module obtains the arterial blood pressure value of the target user and sends the arterial blood pressure value of the target user to the processor; The processor generates a first alarm message when the arterial blood pressure value of the target user is not within a preset blood pressure threshold range; The display module displays the first alarm information, where the first alarm information is used to prompt that the arterial blood pressure value of the target user is abnormal.

5. The method according to claim 4, characterized in that The intra-aortic balloon counterpulsation system further comprises an electronic pressure relief valve and a plurality of pressure sensors, wherein the electronic pressure relief valve and the plurality of pressure sensors are arranged on an air path between the cylinder and the air bag, and the electronic pressure relief valve and the plurality of pressure sensors are electrically connected to the processor; The method further comprises: The pressure sensor obtains the pressure value inside the airbag and sends the pressure value inside the airbag to the processor; The processor controls the electronic pressure relief valve to open when the pressure value inside the airbag is greater than a preset pressure threshold; The method further comprises: The processor generates a second alarm message when a difference between a plurality of pressure values ​​between the cylinder and the airbag acquired by the plurality of sensors exceeds a preset pressure difference; The display module displays the second alarm information, where the second alarm information is used to prompt that the pressure sensor is abnormal.

6. The method according to claim 5, characterized in that The intra-aortic balloon counterpulsation system also includes a mechanical pressure relief valve, which is arranged on the air path between the cylinder and the airbag, and is used to open when the pressure value inside the airbag is greater than a preset pressure threshold.

7. An intra-aortic balloon counterpulsation system, characterized in that: The intra-aortic balloon counterpulsation system comprises a gas cylinder, a voltage regulator, a solenoid valve, a processor, a motor, a cylinder, an airbag and an electrocardiogram monitoring module; the gas cylinder, the cylinder and the airbag are connected via a gas circuit; the solenoid valve is used to control the gas in the gas cylinder to enter the airbag and the cylinder, the voltage regulator is arranged between the gas cylinder and the solenoid valve, and the voltage regulator is used to control the gas pressure output by the gas cylinder within a preset gas pressure threshold range; the processor is electrically connected to the motor and the electrocardiogram monitoring module respectively, and the motor is connected to the cylinder; The ECG monitoring module is used for: Acquire an electrocardiogram signal of a target user, and send the electrocardiogram signal of the target user to the processor; The processor is used to: When the amplitude of the target user's electrocardiogram signal is less than a preset amplitude or the waveform quality coefficient corresponding to the target user's electrocardiogram signal is less than a preset waveform quality coefficient, historical electrocardiogram monitoring data of the target user is obtained, wherein the historical electrocardiogram monitoring data includes electrocardiogram signals corresponding to multiple cardiac cycles of the target user within a target time period; the historical electrocardiogram monitoring data is input into a pre-trained neural network model to predict a target cardiac cycle of the target user; a control instruction is generated according to the target cardiac cycle of the target user; and the motor is controlled to operate according to the control instruction so that the motor adjusts the internal pressure of the cylinder to complete the contraction and expansion operation of the airbag; Among them, the neural network model includes a classifier, a first recurrent neural network and a second recurrent neural network, and the classifier is connected to the first recurrent neural network and the second recurrent neural network respectively; the classifier is used to determine the category corresponding to the historical ECG monitoring data, and the category includes normal heart rhythm or abnormal heart rate; and is used to send the historical ECG monitoring data with a normal heart rate to the first recurrent neural network, and is also used to send the historical ECG monitoring data with an abnormal heart rate to the second recurrent neural network; the first recurrent neural network is used to output the corresponding target cardiac cycle according to the input historical ECG monitoring data with a normal heart rate; the second recurrent neural network is used to output the corresponding target cardiac cycle according to the input historical ECG monitoring data with an abnormal heart rate.

8. The system according to claim 7, characterized in that Before the processor is used to input the historical ECG monitoring data into a pre-trained neural network model to predict the target cardiac cycle of the target user, the processor is specifically used to: Acquire training data, where the training data includes electrocardiogram signals corresponding to multiple cardiac cycles of each user among multiple users, and a category corresponding to each user; Sampling the training data to obtain a plurality of training samples and a category corresponding to each training sample, wherein each training sample includes feature data and label data, wherein the feature data is the electrocardiogram signals of the first n-1 cardiac cycles among the electrocardiogram signals of the consecutive n cardiac cycles of any user, and the label data is the electrocardiogram signal of the nth cardiac cycle among the electrocardiogram signals of the consecutive n cardiac cycles; Training the classifier based on multiple training samples and the category corresponding to each training sample to obtain a trained classifier; Training the first recurrent neural network based on training samples with a category of normal heart rate to obtain a trained first recurrent neural network; The processor trains the second recurrent neural network based on training samples classified as abnormal heart rate to obtain a trained second recurrent neural network, wherein the loss functions of the first recurrent neural network and the second recurrent neural network are root mean square error functions.

9. The system according to claim 8, characterized in that The intra-aortic balloon counterpulsation system further includes a blood pressure monitoring module and a display module, wherein the blood pressure monitoring module is electrically connected to the processor, and the display module is electrically connected to the processor; After the processor is used to control the operation of the motor according to the control instruction to adjust the internal pressure of the cylinder and complete the contraction and expansion operation of the airbag, the blood pressure monitoring module is used to: obtain the arterial blood pressure value of the target user and send the arterial blood pressure value of the target user to the processor; The processor is further configured to: generate a first alarm message when the arterial blood pressure value of the target user is not within a preset blood pressure threshold range; The display module is used to display the first alarm information, where the first alarm information is used to prompt that the arterial blood pressure value of the target user is abnormal.

10. The system according to claim 9, characterized in that The intra-aortic balloon counterpulsation system further comprises an electronic pressure relief valve and a plurality of pressure sensors, wherein the electronic pressure relief valve and the plurality of pressure sensors are arranged on an air path between the cylinder and the air bag, and the electronic pressure relief valve and the plurality of pressure sensors are electrically connected to the processor; The pressure sensor is used to: obtain the pressure value inside the airbag and send the pressure value inside the airbag to the processor; The processor is further configured to: When the pressure value inside the airbag is greater than a preset pressure threshold, controlling the electronic pressure relief valve to open; generating a second alarm message when a difference between a plurality of pressure values ​​between the cylinder and the airbag acquired by the plurality of sensors exceeds a preset pressure difference; The display module is further used to: display the second alarm information, where the second alarm information is used to prompt that the pressure sensor is abnormal; The intra-aortic balloon counterpulsation system also includes a mechanical pressure relief valve, which is arranged on the air path between the cylinder and the airbag, and is used to open when the pressure value inside the airbag is greater than a preset pressure threshold.

Images