Traffic determination method, traffic detection model training method, device, and medium

Abstract

The application relates to a flow determination method, a training method of a flow detection model, equipment and a medium, and belongs to the technical field of medical devices. The method comprises the following steps: acquiring perfusion data corresponding to a trans-catheter ventricular assist device, blood pressure data and motor operation data corresponding to the trans-catheter ventricular assist device; performing flow detection processing according to the perfusion data, the blood pressure data and the motor operation data to obtain flow data corresponding to the trans-catheter ventricular assist device. The technical scheme provided by the application greatly reduces the dependence of flow detection on a flow sensor, effectively solves the technical problem that the pumping flow of the trans-catheter ventricular assist device in the heart is difficult to detect, reduces the complexity of flow detection and improves the efficiency of flow detection.

Description

Technical Field

[0001] This application relates to a method for determining flow rate, a method for training a flow rate detection model, equipment, and media, belonging to the field of medical device technology. Background Technology

[0002] Transcatheter ventricular assist devices (VTs) are used to provide mechanical circulatory support to patients. When a VT works, it pumps blood from the ventricles of the heart into the arteries, thus providing ventricular support. During VT operation, it is typically necessary to measure the current pumped blood flow to determine the device's operational status.

[0003] However, the pump head of the interventional pump in the transcatheter ventricular assist device needs to be percutaneously inserted into the heart through a peripheral blood vessel. Therefore, the pump head is small and there is not enough space to install a flow sensor, making it difficult to measure blood flow. Summary of the Invention

[0004] This application provides a method for determining blood flow, a method for training a blood flow detection model, equipment, and medium, solving the technical problem of difficulty in monitoring the blood flow pumped by a transcatheter ventricular assist device. This application provides the following technical solution:

[0005] On the one hand, a method for determining flow rate is provided, the method being applied to a transcatheter ventricular assist device, the method comprising:

[0006] Acquire perfusion data, blood pressure data, and motor operation data corresponding to the transcatheter ventricular assist device;

[0007] The flow rate data corresponding to the transcatheter ventricular assist device is obtained by performing flow rate detection processing based on the perfusion data, the blood pressure data, and the motor operation data.

[0008] Optionally, the perfusion data includes at least one perfusion parameter data, which includes at least one of perfusion fluid pressure data, perfusion fluid flow rate data, and perfusion pump speed data; the blood pressure data includes at least one blood pressure parameter data, which includes at least one of arterial pressure data, ventricular pressure data, and pump pressure differential data.

[0009] Optionally, the step of performing flow detection processing based on the perfusion data, the blood pressure data, and the motor operation data to obtain the flow data corresponding to the transcatheter ventricular assist device includes:

[0010] The flow rate data is obtained by performing flow rate detection processing based on the perfusion fluid pressure data, the arterial pressure data, and the motor operation data.

[0011] Optionally, the motor operating data includes motor speed data and motor current data, and the step of performing flow detection processing based on the perfusion fluid pressure data, the arterial pressure data, and the motor operating data to obtain the flow data includes:

[0012] The flow rate data is obtained by performing flow rate detection processing based on the perfusion fluid pressure data, the arterial pressure data, the motor speed data, and the motor current data.

[0013] Optionally, the transcatheter ventricular assist device includes:

[0014] An interventional catheter pump, comprising a catheter, a pump head connected to the distal end of the catheter, and a coupling assembly connected to the proximal end of the catheter, the pump head being delivered by the catheter to a desired location on the heart for pumping operations;

[0015] A perfusion channel, the perfusion channel extending at least through the catheter, the inlet of the perfusion channel being located on the coupling assembly, and the outlet of the perfusion channel being located at the pump head; the perfusion fluid in the perfusion channel enters a desired location in the cardiovascular system through the outlet; the desired location in the cardiovascular system includes at least one of the aorta, pulmonary artery, left ventricle, right ventricle, left atrium, and right atrium;

[0016] The infusion fluid pressure data characterizes the infusion fluid pressure generated from the pressure acquisition location to the outlet.

[0017] Optionally, the step of performing flow detection processing based on the perfusion data, the blood pressure data, and the motor operation data to obtain the flow data corresponding to the transcatheter ventricular assist device includes:

[0018] The perfusion data, blood pressure data, and motor operation data are input into a preset flow detection model for flow detection processing, and the flow data is output.

[0019] The flow detection model is a machine learning model trained on multiple sets of sample data. Each set of sample data includes corresponding perfusion sample data, blood pressure sample data, motor operation sample data, and sample flow data.

[0020] Optionally, the step of inputting the perfusion data, the blood pressure data, and the motor operation data into a preset flow detection model for flow detection processing, and outputting the flow data, includes:

[0021] The perfusion fluid pressure data, arterial pressure data, motor speed data, and motor current data are input into a preset Gaussian model for flow detection processing, and the flow data is output.

[0022] The perfusion data includes the perfusion fluid pressure data, the blood pressure data includes the arterial pressure data, the motor operation data includes the motor speed data and the motor current data, and the preset flow detection model includes the preset Gaussian model, which is a Gaussian model obtained by training multiple sets of sample data.

[0023] Optionally, before inputting the perfusion data, the blood pressure data, and the motor operation data into a preset flow detection model for flow detection processing and outputting the flow data, the method further includes:

[0024] Based on the data type of perfusion data and / or the data type of blood pressure data, a first flow detection model corresponding to the transcatheter ventricular assist device is determined; the data type of perfusion data is used to indicate the types of perfusion parameter data included in the perfusion data; the data type of blood pressure data is used to indicate the types of blood pressure parameter data included in the blood pressure data.

[0025] Accordingly, the step of inputting the perfusion data, the blood pressure data, and the motor operation data into a preset flow detection model for flow detection processing, and outputting the flow data, includes:

[0026] The perfusion data, the blood pressure data, and the motor operation data are input into the first flow detection model, and the flow data is output.

[0027] The preset flow detection model includes flow detection models corresponding to different data types. In each group of sample data corresponding to the first flow detection model, the data type of the perfusion sample data is consistent with the data type of the currently acquired perfusion data, and the data type of the blood pressure sample data is consistent with the data type of the currently acquired perfusion data.

[0028] Optionally, before inputting the perfusion data, the blood pressure data, and the motor operation data into a preset flow detection model for flow detection processing and outputting the flow data, the method further includes:

[0029] Acquire the first acquisition location of the perfusion data and the second acquisition location of the blood pressure data;

[0030] A second flow detection model is determined based on the first and second acquisition locations;

[0031] The preset flow detection model includes flow detection models corresponding to different collection locations. Each set of sample data corresponding to the second flow detection model includes: perfusion sample data collected based on the first collection location, blood pressure sample data collected based on the second collection location, and motor operation sample data.

[0032] Accordingly, the perfusion data, the blood pressure data, and the motor operation data are input into a preset flow detection model for flow detection processing, and the flow data is output, including:

[0033] The perfusion data, blood pressure data, and motor operation data are input into the second flow detection model for flow detection processing, and the flow data is output.

[0034] On the other hand, a method for training a traffic detection model is provided, the method comprising:

[0035] Acquire a sample dataset collected during the operation of the transcatheter ventricular assist device. Each set of sample data in the sample dataset includes perfusion sample data, blood pressure sample data, motor operation sample data, and corresponding flow measurement data corresponding to the transcatheter ventricular assist device.

[0036] The pre-set machine learning model is trained based on the perfusion sample data, the blood pressure sample data, the motor operation sample data, and the flow measurement data to obtain a flow detection model;

[0037] The flow detection model is used to detect the flow data corresponding to the transcatheter ventricular assist device based on the perfusion data, blood pressure data, and motor operation data generated during the application of the transcatheter ventricular assist device.

[0038] Optionally, the sample dataset includes multiple sets of training data and multiple sets of test data. The step of training a preset machine learning model based on the perfusion sample data, the blood pressure sample data, the motor operation sample data, and the flow measurement data to obtain a flow detection model includes:

[0039] Based on the perfusion sample data, blood pressure sample data, motor operation sample data and flow measurement data in the multiple sets of training data, a modeling process is performed to generate the first machine learning model;

[0040] The perfusion sample data, blood pressure sample data, and motor operation sample data from each set of test data are input into the first machine learning model for flow detection processing, and the flow detection data corresponding to each set of test data is output.

[0041] The traffic measurement data in each set of test data is compared with the traffic detection data corresponding to each set of test data to determine the loss information corresponding to the first machine learning model.

[0042] The model parameters of the first machine learning model are adjusted based on the loss information to obtain the traffic detection model.

[0043] Optionally, the first machine learning model includes a Gaussian model, and the step of comparing the traffic measurement data in each set of test data with the traffic detection data corresponding to each set of test data to determine the loss information corresponding to the first machine learning model includes:

[0044] The average flow rate is determined based on the flow rate measurement data in each group of sample data;

[0045] The average flow data, the flow detection data corresponding to each set of test data, and the flow measurement data in each set of test data are compared to obtain the variance data and root mean square error data corresponding to the Gaussian model. The loss information includes the variance data and the root mean square error data.

[0046] Optionally, the flow measurement data is determined by the change in liquid weight per unit time.

[0047] Optionally, the perfusion sample data in the sample dataset includes perfusion fluid pressure sample data corresponding to the transcatheter ventricular assist device, the blood pressure sample data in the sample dataset includes arterial pressure sample data corresponding to the transcatheter ventricular assist device, and the motor operation sample data in the sample dataset includes motor speed sample data and motor current sample data corresponding to the transcatheter ventricular assist device.

[0048] Accordingly, the flow detection model is specifically used to detect the flow data based on the perfusion fluid pressure data, arterial pressure data, motor speed data, and motor current data generated during the application of the transcatheter ventricular assist device.

[0049] On the other hand, a flow rate determining device is provided, the flow rate determining device being applied to a transcatheter ventricular assist device, the device comprising:

[0050] The data acquisition module is used to acquire perfusion data, blood pressure data, and motor operation data corresponding to the transcatheter ventricular assist device.

[0051] The flow rate determination module is used to perform flow rate detection processing based on the perfusion data, the blood pressure data, and the motor operation data to obtain the flow rate data corresponding to the transcatheter ventricular assist device.

[0052] On the other hand, a training apparatus for a traffic detection model is provided, the apparatus comprising:

[0053] The data acquisition module is used to acquire sample datasets collected during the operation of the transcatheter ventricular assist device. Each set of sample data in the sample dataset includes perfusion sample data, blood pressure sample data, motor operation sample data, and corresponding flow measurement data corresponding to the transcatheter ventricular assist device.

[0054] The model training module is used to train a preset machine learning model based on the perfusion sample data, the blood pressure sample data, the motor operation sample data, and the flow measurement data to obtain a flow detection model.

[0055] The flow detection model is used to detect the flow data corresponding to the transcatheter ventricular assist device based on the perfusion data, blood pressure data, and motor operation data generated during the application of the transcatheter ventricular assist device.

[0056] On the other hand, a transcatheter ventricular assist device is provided, the transcatheter ventricular assist device comprising: a computer device including a processor and a memory; the memory storing a program, the program being loaded and executed by the processor to implement the flow determination method provided above; or, implementing the flow detection model training method provided above.

[0057] On the other hand, a computer device is provided, the device including a processor and a memory; the memory stores a program, the program being loaded and executed by the processor to implement the traffic determination method provided in the above aspects; or, to implement the training method for the traffic detection model provided in the above aspects.

[0058] On the other hand, a computer-readable storage medium is provided, wherein a program is stored in the storage medium, and when executed by a processor, the program is used to implement the traffic determination method provided in the above aspects; or, to implement the training method for the traffic detection model provided in the above aspects.

[0059] On the other hand, a computer program product includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform actions to implement the traffic determination method provided in the above aspects; or, to implement the training method for the traffic detection model provided in the above aspects.

[0060] One embodiment of this application provides a flow determination method that can detect the pumping flow of a transcatheter ventricular assist device (VTAVD) based on perfusion data, blood pressure data, and motor operation data. This greatly reduces the dependence of flow detection on flow sensors, effectively solves the technical problem of difficulty in detecting the pumping flow of a VTAVD inside the heart, reduces the complexity of flow detection, and improves the efficiency of flow detection.

[0061] In addition, since the perfusion state of the infusion fluid, the pressure state of the blood, and the operating state of the motor are all related to the flow rate pumped by the catheter-guided ventricular assist device, the accuracy of flow rate detection can be ensured by determining the flow rate data based on the perfusion fluid pressure data characterizing the perfusion state, the arterial pressure data characterizing the pressure state of the blood, and the motor operating data characterizing the operating state of the motor.

[0062] In addition, by training a machine learning model as a traffic detection model, the traffic detection model can learn the data correlation between various parameters from the distribution of multiple sets of sample data. By detecting traffic data through the traffic detection model, the accuracy of traffic detection results can be guaranteed.

[0063] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, the preferred embodiments of this application are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0064] Figure 1 This is a block diagram of a transcatheter ventricular assist device provided in one embodiment of this application;

[0065] Figure 2 This is a schematic diagram of the connection between the infusion assembly and the working assembly according to an embodiment of this application;

[0066] Figure 3 This is a flowchart of a flow determination method provided in one embodiment of this application;

[0067] Figure 4 This is a flowchart of a training method for a traffic detection model provided in one embodiment of this application;

[0068] Figure 5 This is a block diagram of a flow determination device provided in one embodiment of this application;

[0069] Figure 6 This is a block diagram of a training apparatus for a traffic detection model provided in one embodiment of this application;

[0070] Figure 7 This is a block diagram of a computer device provided in one embodiment of this application. Detailed Implementation

[0071] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0072] Traditional methods for determining blood flow typically involve installing a flow sensor within the blood pump to detect flow data. However, for transcatheter ventricular assist devices (TAAVDs), the blood pump is an interventional catheter pump, and the pump head of such pumps needs to be implanted in the body during use; therefore, the pump head is usually quite small. In this case, it is difficult to install a flow sensor within the pump head to detect flow data.

[0073] Based on this, this application provides a flow rate determination method, which can perform flow rate detection processing based on the perfusion data, blood pressure data and motor operation data corresponding to the transcatheter ventricular assist device, thereby detecting the flow rate data pumped by the transcatheter ventricular assist device, without the need to install a flow sensor in the transcatheter ventricular assist device to detect the flow rate.

[0074] Before introducing the method for determining traffic flow in this application, we will first introduce the application scenarios of this method.

[0075] The flow rate determination method provided in this application is applied to a transcatheter ventricular assist device (VTAD), which is used to provide mechanical circulatory support to patients. In one illustrative scenario, the VTAD can be used as a left ventricular assist device, pumping blood from the left ventricle into the ascending aorta during operation. In other feasible and not explicitly excluded scenarios, the VTAD can also be used as a right ventricular assist device, pumping venous blood into the right ventricle during operation. Alternatively, the VTAD can also be applied to scenarios such as pumping blood from the vena cava and / or the right atrium into the right ventricle, pumping blood from the vena cava and / or the right atrium into the pulmonary artery and / or pumping blood from the renal vein into the vena cava, etc. This application does not limit the application scenarios of the VTAD. In the following embodiments, the use of a VTAD to pump blood from the ventricle into the aorta is illustrated as an example.

[0076] refer to Figure 1 The block diagram shown is of a transcatheter ventricular assist device 100 provided in one embodiment of this application, which includes at least an interventional catheter pump 110.

[0077] The pump head of the interventional catheter pump 110 can be percutaneously inserted into the heart through peripheral blood vessels. For example, the pump head can be placed between the left ventricle and the ascending aorta. The blood inlet of the pump head is placed into the left ventricle, and the blood outlet of the pump head is placed into the ascending aorta, thereby pumping blood from the left ventricle into the ascending aorta to achieve ventricular assist function.

[0078] Optionally, the interventional duct pump 110 includes a pump head 111, a coupler, and a driver. The driver is detachably connected to the coupler. After the driver is connected to the coupler, the impeller in the pump head 111 can be rotated via the coupler and the duct.

[0079] The driver includes a drive motor, and an impeller is provided in the pump head 111. The impeller is connected to the drive motor for transmission, so that the impeller is driven to rotate by controlling the rotation of the motor, thereby driving the pump head 111 to work. The pump head 111 may also include other components required for operation, such as a pump casing to house the impeller. In this embodiment, the components included in the pump head 111 are not listed one by one.

[0080] In the interventional tubing pump 110, the pump head 111 is connected to the drive motor in the driver via a tubing.

[0081] The transcatheter ventricular assist device 100 also includes a perfusion assembly 120 for providing perfusion fluid to the interventional catheter pump 110. The perfusion fluid may be a physiological fluid required to maintain human function, such as saline, glucose solution, anticoagulant, or any combination thereof.

[0082] Optionally, the perfusion fluid is used to perfuse the interventional catheter pump through the catheter to remove air from the catheter space, preventing air from entering the blood vessels and heart through the catheter and causing adverse physiological reactions. The perfusion fluid flows through the perfusion channel and through the opening at the pump head into the desired location in the cardiovascular system. Optionally, the desired location in the cardiovascular system includes at least one of the aorta, pulmonary artery, left ventricle, right ventricle, left atrium, and right atrium.

[0083] In one example, such as Figure 2 As shown, the perfusion assembly 120 includes a perfusion fluid injection port 121, through which perfusion fluid is injected into the interventional catheter pump and flows through the catheter 210 into the human body. The perfusion fluid injection port 121 can be the inlet of the perfusion channel.

[0084] Optionally, the transcatheter ventricular assist device 100 also includes a peristaltic pump for pumping perfusion fluid. Indicatively, the perfusion fluid tubing may be clamped onto the peristaltic pump; this embodiment does not limit the location of the peristaltic pump.

[0085] In this embodiment of the application, in order to detect the flow data of the interventional catheter pump, the catheter-based ventricular assist device may further include a first detection unit 130, a second detection unit 140, a third detection unit 150, and a control unit 160 respectively connected to the first detection unit 130, the second detection unit 140 and the third detection unit 150.

[0086] The first detection unit 130 is used to detect perfusion data corresponding to the transcatheter ventricular assist device;

[0087] The second detection unit 140 is used to detect blood pressure data corresponding to the transcatheter ventricular assist device;

[0088] The third detection unit 150 is used to detect the motor operation data corresponding to the transcatheter ventricular assist device;

[0089] The control unit 160 is used to acquire perfusion data, blood pressure data and motor operation data corresponding to the transcatheter ventricular assist device; and to perform flow detection processing based on the perfusion data, blood pressure data and motor operation data to obtain the flow data corresponding to the transcatheter ventricular assist device.

[0090] The technical solution provided in this application embodiment can detect the pumping flow of the transcatheter ventricular assist device (VTAVD) based on the perfusion data, blood pressure data, and motor operation data corresponding to the VTAVD. This greatly reduces the dependence of flow detection on flow sensors, effectively solves the technical problem of difficulty in detecting the pumping flow of the VTAVD inside the heart, reduces the complexity of flow detection, and improves the efficiency of flow detection.

[0091] Optionally, the control unit 160 performs flow detection processing based on perfusion data, blood pressure data, and motor operation data to obtain flow data corresponding to the transcatheter ventricular assist device, including: inputting perfusion data, blood pressure data, and motor operation data into a preset flow detection model for flow detection processing, and outputting flow data.

[0092] The flow detection model is a machine learning model trained on multiple sets of sample data. Each set of sample data includes corresponding perfusion sample data, blood pressure sample data, motor operation sample data, and sample flow data.

[0093] The flow detection model in the control unit 160 can be trained based on the control unit 160, or it can be trained in other devices and then configured into the control unit 160. This embodiment does not limit the training environment of the flow detection model.

[0094] Taking the flow detection model trained based on the control unit 160 as an example, the control unit 160 is also used to: acquire sample datasets collected during the operation of the transcatheter ventricular assist device, each set of sample data in the sample dataset including perfusion sample data, blood pressure sample data, motor operation sample data and corresponding flow measurement data corresponding to the transcatheter ventricular assist device; and train a preset machine learning model based on the perfusion sample data, blood pressure sample data, motor operation sample data and flow measurement data to obtain the flow detection model.

[0095] Among them, the flow detection model is used to detect the flow data corresponding to the transcatheter ventricular assist device based on the perfusion data, blood pressure data and motor operation data generated during the application of the transcatheter ventricular assist device.

[0096] Indicatively, the sample dataset may be collected by the first detection unit, the second detection unit, and the third detection unit, or it may be sent by other devices. This embodiment does not limit the method of obtaining the sample dataset.

[0097] In actual implementation, if the training process of the traffic detection model is implemented in other devices, the steps are the same as the training steps described above, and this embodiment will not elaborate further.

[0098] Alternatively, the control unit 160 can also be implemented as a device independent of the transcatheter ventricular assist device. This embodiment does not limit the implementation of the control unit 160.

[0099] Schematic, the control unit 160 includes at least a processor and a memory, the memory storing a program that is loaded and executed by the processor to implement the flow determination method of this application; or, to implement the training method of the flow detection model of this application.

[0100] In actual implementation, the transcatheter ventricular assist device may also include other components required for operation, such as power supply components and display screens. This embodiment will not describe each of the components included in the transcatheter ventricular assist device.

[0101] The following describes a traffic determination method provided by an embodiment of this application, which is applied to... Figure 1 The transcatheter ventricular assist device shown can be executed by a control unit in the transcatheter ventricular assist device or by other devices that are communicatively connected to the transcatheter ventricular assist device, such as a computer or tablet computer. This embodiment does not limit the type of other devices. Figure 3 This is a flowchart of a traffic determination method provided in one embodiment of this application, which includes at least the following steps:

[0102] Step 301: Obtain perfusion data, blood pressure data, and motor operation data corresponding to the transcatheter ventricular assist device.

[0103] Perfusion data is used to characterize the perfusion status of the perfusion fluid in the perfusion assembly. In transcatheter ventricular assist devices, the perfusion fluid enters the heart through the internal gaps of the catheter. The perfusion pathway (formed by the perfusion tubing and catheter) connects to the blood vessels or the interior of the heart, i.e., to the blood flow pathway. In other words, there is a correlation between the flow of the perfusion fluid and the flow of blood. Therefore, perfusion data can be acquired to determine flow rate data.

[0104] Blood pressure data is used to characterize the pressure state of blood in the body. There is a strong correlation between blood flow and blood pressure. Therefore, blood pressure data at a preset location can be obtained when determining blood flow.

[0105] Motor operating data is used to characterize the operating efficiency of the motor in a transcatheter ventricular assist device (VAVD). The type of fluid pumped by the VAVD (e.g., viscosity) affects the motor's operating efficiency at a set speed. At the set speed, the amount of electricity consumed by the motor pumping different types of fluids also varies. Therefore, there is a correlation between motor operating data and the type of fluid pumped by the blood pump, and a correlation also exists between flow rate and fluid type. Furthermore, the motor is the power source driving the impeller rotation; therefore, the aforementioned motor operating data can be used to determine flow rate data.

[0106] Optionally, the above-mentioned perfusion data includes at least one perfusion parameter data. Optionally, the above-mentioned at least one perfusion parameter data includes, but is not limited to, at least one of the following:

[0107] 1. Infusion fluid pressure data. Optionally, the first detection unit 130 includes an infusion fluid pressure sensor. Optionally, the infusion fluid pressure sensor can be set at a designated position on the infusion channel; this application does not limit the setting position of the infusion fluid pressure sensor.

[0108] In an exemplary embodiment, the transcatheter ventricular assist device includes:

[0109] An interventional catheter pump includes a catheter, a pump head connected to the distal end of the catheter, and a coupling assembly connected to the proximal end of the catheter. The pump head can be delivered by the catheter to the desired location on the heart to perform a pumping operation.

[0110] The perfusion channel extends through at least the catheter, with the inlet of the perfusion channel located on the coupling assembly and the outlet of the perfusion channel located at the pump head; the perfusion fluid in the perfusion channel enters the desired location in the cardiovascular system through the outlet; the desired location in the cardiovascular system includes at least one of the aorta, pulmonary artery, left ventricle, right ventricle, left atrium, and right atrium.

[0111] The injection fluid pressure data characterizes the injection fluid pressure generated from the pressure acquisition point to the outlet. That is, the injection fluid pressure data is the sum of all pipeline pressures from the pressure sensor location (i.e., the pressure acquisition point) to the outlet and the outlet pressure.

[0112] 2. Injection fluid flow rate data. Optionally, the first detection unit 130 includes an injection fluid flow sensor. Optionally, the injection fluid flow sensor can be installed in the injection pipeline of the injection fluid; this application does not limit the installation location of the flow sensor. Furthermore, the injection fluid flow rate can also be determined based on changes in the volume of the injection fluid.

[0113] 3. Infusion pump speed data. The infusion pump is used to drive the flow of the infusion fluid, and the infusion pump speed data is used to characterize the speed of the infusion pump. The infusion pump can be a peristaltic pump or other types of pumps (such as embolized pumps). Optionally, the first detection unit 130 includes a speed sensor corresponding to the infusion pump, which is used to detect the speed data of the pump (such as a peristaltic pump).

[0114] The blood pressure data mentioned above can be detected by the blood pressure sensor accessory in the transcatheter ventricular assist device, transmitted by other devices, or manually entered.

[0115] Optionally, the second detection unit 140 includes a blood pressure sensor for collecting blood pressure data. Optionally, the blood pressure data includes at least one blood pressure parameter, which includes, but is not limited to, at least one of the following:

[0116] 1. Arterial pressure data. Optionally, arterial pressure data includes, but is not limited to, aortic pressure data and pulmonary artery pressure data. Since the interventional catheter pump in a transcatheter ventricular assist device pumps blood from the ventricles to the arteries—for example, from the left ventricle to the aorta and from the right ventricle to the pulmonary artery—aortic pressure data or pulmonary artery pressure data is correlated with the pumping flow rate, and the aforementioned arterial pressure data can be selected to determine the flow rate.

[0117] In one possible implementation, an arterial pressure sensor is mounted on a catheter to detect arterial blood pressure. The sensor is inserted into the body via a percutaneous intubation with a pump head, and once it reaches a designated location, it can measure arterial blood pressure, such as aortic pressure or pulmonary artery pressure.

[0118] In another possible implementation, the arterial pressure sensor is placed in an external tubing connected to the artery, also for detecting arterial blood pressure. This application does not limit the placement of the arterial pressure sensor in its embodiments.

[0119] 2. Ventricular pressure data. Optionally, ventricular pressure data includes, but is not limited to, left ventricular pressure data and right ventricular pressure data. For reasons similar to arterial pressure data, left ventricular pressure data or right ventricular pressure data is correlated with pump flow rate; therefore, the aforementioned ventricular pressure data can be selected to determine the flow rate. The ventricular pressure sensor is similar to the arterial pressure sensor, but its placement differs; this embodiment does not limit its location.

[0120] 3. Pump pressure differential data between arterial pressure and ventricular pressure data. The above pump pressure differential data characterizes the pressure difference between arterial pressure and ventricular pressure, i.e., the pressure differential corresponding to the interventional catheter pump.

[0121] Optionally, the motor operating data includes, but is not limited to, motor speed data and motor current data. Motor speed data represents the motor speed of the drive motor in the transcatheter ventricular assist device, and motor current data represents the motor current of the drive motor in the transcatheter ventricular assist device. Optionally, the drive motor is a brushless DC motor. There may be a time-series correspondence between the motor current data and the motor speed data; the motor current data can reflect the motor's power consumption under the corresponding motor speed data, thus reflecting the motor's operating energy efficiency under the current fluid type.

[0122] In this embodiment of the application, the obtained perfusion data, blood pressure data and motor operation data correspond to each other. Based on this, the perfusion data, blood pressure data and motor operation data are collected synchronously, or the maximum collection time interval between each data item in a set of data is less than a threshold.

[0123] Step 302: Perform flow detection processing based on perfusion data, blood pressure data, and motor operation data to obtain the flow data corresponding to the transcatheter ventricular assist device.

[0124] The perfusion data, blood pressure data, and motor operation data mentioned above are all correlated with flow rate. In one possible implementation, the corresponding flow rate data is determined based on the data quantification relationship between the perfusion data, blood pressure data, and motor operation data.

[0125] In another possible implementation, the correlation between the above-mentioned data types can be learned based on machine learning models to determine the traffic.

[0126] Optionally, step 302 above includes: inputting perfusion data, blood pressure data and motor operation data into a preset flow detection model for flow detection processing, and outputting flow data.

[0127] The flow detection model is a machine learning model trained on multiple sets of sample data. Each set of sample data includes corresponding perfusion sample data, blood pressure sample data, motor operation sample data, and sample flow data. The specific training process of the flow detection model is detailed in the following embodiment, which will not be elaborated further here.

[0128] Since the above motor operating data can characterize the motor's operating efficiency, and the motor's operating efficiency is affected by the type of liquid being pumped, such as viscosity, it means that the flow rate is directly related to the motor's operating efficiency. Moreover, the correlation factors are quite complex. In addition to changes in liquid type (such as viscosity) causing changes in flow rate, the operating state of the motor itself can also cause changes in flow rate.

[0129] Specifically, for a motor, at a set speed, the electrical energy consumed by the motor may vary depending on the type of liquid it pumps. For example, a motor pumping a liquid with higher viscosity consumes more electricity than one pumping a liquid with lower viscosity. Therefore, motor operating data can be used to determine flow rate data.

[0130] Furthermore, given a defined blood flow path that is connected to the perfusion fluid flow path, both fluid type and blood pressure influence the pump's flow rate, and the perfusion fluid flow is also affected by changes in blood flow. Therefore, by simultaneously inputting motor operation data, perfusion data, and blood pressure data associated with fluid type into a trained flow detection model, the current flow rate can be determined without using a flow sensor. This is achieved by leveraging the learned relationships between these four data points during training, significantly reducing the reliance on flow sensors, lowering the complexity of flow detection, and improving its efficiency.

[0131] Furthermore, since the flow detection model is trained directly based on perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data, it can directly and effectively learn the correlation between flow rate, motor operation data, perfusion data, and blood pressure data. In other words, the flow detection model does not depend on the determination of the liquid type (such as the determination of liquid viscosity). Compared to the Bernoulli equation method, which requires calculating viscosity before calculating flow rate, the flow detection method provided in this application does not depend on the liquid type, such as liquid viscosity. Specifically, it can be understood that there is no need to input liquid type information into the flow detection model, and there are no intermediate steps in the entire flow detection process to calculate or determine the liquid type, such as intermediate steps to calculate viscosity.

[0132] In addition to the aforementioned motor operation data and blood pressure data, perfusion data unique to transcatheter ventricular assist devices are also introduced to detect flow from more dimensions. This ensures the accuracy of flow detection of the catheter pump without actually determining specific blood parameters, effectively reducing calculation steps, lowering the computational complexity of flow detection data, and effectively improving detection speed. Moreover, it is more suitable for control hosts with limited computing power in transcatheter ventricular assist devices.

[0133] In some implementations, the sample dataset does not impose constraints on liquid type. For example, the sample dataset does not impose viscosity range constraints on liquid viscosity; the liquid viscosity corresponding to each group of sample data in the sample dataset is not limited by a preset viscosity range. Since the sample dataset contains data samples corresponding to various liquid types, the flow detection model trained based on these data is a flow detection model applicable to all liquid types and is an independent and singular model.

[0134] In one possible implementation, the aforementioned flow detection model can be a trained neural network model. During the training of the neural network model, the flow measurement data in the sample data can serve as supervisory information for the neural network model, thereby constraining the accuracy of the flow output of the trained neural network model. This allows the neural network model to learn the feature correlation between the input data (perfusion sample data, blood pressure sample data, motor operation sample data) and the output data (flow measurement data), thus accurately predicting the corresponding flow data when new data is input.

[0135] In another possible implementation, the aforementioned flow detection model can be a trained Gaussian model. During Gaussian model training, the mean vector and covariance matrix corresponding to each set of sample data are determined based on the perfusion sample data, blood pressure sample data, motor operation sample data, and corresponding flow measurement data in each set of sample data. The covariance matrix effectively characterizes the correlation between various variables; therefore, when the model is applied, for new input data, the Gaussian model can effectively predict the flow data corresponding to the new input data based on the historical data distribution of the sample data. Furthermore, compared to neural network models with a large number of parameters, the Gaussian model has a relatively small computational cost.

[0136] To illustrate, in order to improve the accuracy of flow detection, the data type of perfusion sample data is the same as that of perfusion data; the data type of blood pressure sample data is the same as that of blood pressure data; and the data type of motor operation sample data is the same as that of motor operation data.

[0137] In summary, the flow determination method provided in this embodiment can detect the pumping flow of the transcatheter ventricular assist device (VTAVD) based on the perfusion data, blood pressure data, and motor operation data corresponding to the VTAVD. This greatly reduces the dependence of flow detection on flow sensors, effectively solves the technical problem of difficulty in detecting the pumping flow of the VTAVD inside the heart, reduces the complexity of flow detection, and improves the efficiency of flow detection.

[0138] Furthermore, since the perfusion state of the infusion fluid, the pressure state of the blood, and the operating state of the motor can all affect or reflect the pumping flow rate of the interventional catheter pump, determining the flow rate data based on the perfusion data characterizing the liquid state of the infusion fluid, the blood pressure data characterizing the pressure state of the blood, and the motor operating data characterizing the operating state of the motor can ensure the accuracy of flow rate detection.

[0139] In addition, traffic data is detected through a traffic detection model, which is a machine learning model trained on sample data. Machine learning models can learn relatively accurate correlations between data, thus ensuring the accuracy of traffic detection results.

[0140] Optionally, based on the above embodiments, the perfusion data includes at least one perfusion parameter data, and the blood pressure data includes at least one blood pressure parameter data. The following description takes at least one perfusion parameter data including perfusion fluid pressure data and at least one blood pressure parameter data including arterial pressure data as an example.

[0141] In this case, in step 302, flow detection processing is performed based on perfusion data, blood pressure data, and motor operation data to obtain the flow data corresponding to the transcatheter ventricular assist device, including: flow detection processing is performed based on perfusion fluid pressure data, arterial pressure data, and motor operation data to obtain the flow data.

[0142] In one possible implementation, no pressure sensor is installed on the catheter of the transcatheter ventricular assist device, making it difficult to obtain ventricular pressure data. However, arterial pressure data can be obtained through in vitro detection. In this case, flow rate can be detected based solely on easily detectable arterial pressure data, perfusion fluid pressure data, and motor operation data. Furthermore, the flow rate of the interventional catheter pump can be accurately output, reducing the difficulty of flow rate detection.

[0143] The aforementioned arterial pressure data can be aortic pressure (with left ventricular assist), pulmonary artery pressure (with right ventricular assist), or arterial pressure data from other locations. This application does not limit this.

[0144] The motor operating data includes, but is not limited to, motor speed data and motor current data. When the motor operating data includes motor speed data and motor current data, flow rate data is obtained by performing flow rate detection processing based on the perfusion fluid pressure data, arterial pressure data, and motor operating data. This includes: performing flow rate detection processing based on the perfusion fluid pressure data, arterial pressure data, motor speed data, and motor current data to obtain the flow rate data.

[0145] The aforementioned motor current and motor speed data can have a time-series correspondence. The motor current data can reflect the motor's power consumption under the corresponding motor speed data, thus reflecting the motor's operating efficiency under the current liquid type. Therefore, using specific motor current and motor speed data as motor operating data can effectively characterize the current driving motor's operating efficiency, thereby ensuring the accuracy of flow detection.

[0146] Accordingly, if the flow detection processing method is based on a preset flow detection model, then the perfusion data, blood pressure data, and motor operation data are input into the preset flow detection model for flow detection processing, and the flow data is output. This includes: inputting the perfusion fluid pressure data, arterial pressure data, motor speed data, and motor current data into a preset Gaussian model for flow detection processing, and outputting the flow data.

[0147] Optionally, the preset flow detection model includes a preset Gaussian model, which is a Gaussian model obtained by training multiple sets of sample data. Each set of sample data includes perfusion fluid pressure sample data, arterial pressure sample data, motor speed sample data, and motor current sample data.

[0148] For Gaussian models, during training, the mean vector and covariance matrix for each set of sample data are determined based on perfusion pressure, arterial pressure, motor speed, motor current, and flow measurement data. The covariance matrix effectively characterizes the relationships between variables. Therefore, when applying the model, for new input data, the Gaussian model can effectively predict the flow rate corresponding to that new input data based on the historical data distribution of the sample data. Furthermore, compared to neural network models with a large number of parameters, the Gaussian model has relatively low computational cost while maintaining high accuracy. Moreover, the parameters in these training sample sets conform to a Gaussian distribution, effectively improving the accuracy of flow detection.

[0149] In other embodiments, the preset traffic detection model may also include other mathematical models such as neural network models. This embodiment does not limit the type of traffic detection model. Since the sample data of the Gaussian model follows a normal distribution and has statistical significance, this embodiment uses a preset traffic detection model including a preset Gaussian model as an example for explanation.

[0150] In other embodiments, the motor operating data may also include motor current data but not motor speed data; in this case, the sample data also includes motor current sample data but not motor speed sample data. Alternatively, the motor operating data may also include motor speed data but not motor current data; in this case, the sample data includes motor speed sample data but not motor current sample data. This embodiment does not limit the implementation method of the motor operating data.

[0151] Optionally, in addition to the injection fluid pressure data, at least one injection parameter data may also include injection fluid flow rate data and / or injection pump speed data. In this case, the flow rate detection processing also needs to be combined with the injection fluid flow rate data and / or injection pump speed data.

[0152] Optionally, in addition to arterial pressure data, at least one blood pressure parameter may also include ventricular pressure data and / or differential pressure data. In this case, the flow detection processing also needs to be combined with ventricular pressure data and / or differential pressure data.

[0153] Since perfusion fluid pressure data and arterial pressure data have a significant impact on the pumping flow rate of the interventional catheter pump, in this embodiment, by setting at least one perfusion parameter data including perfusion fluid pressure data and at least one blood pressure parameter data including arterial pressure data, the accuracy of flow rate detection can be ensured.

[0154] In other embodiments, at least one perfusion parameter data may include perfusion fluid flow rate data and / or perfusion pump speed data, but not perfusion fluid pressure data; and / or at least one blood pressure parameter data may include ventricular pressure data and / or differential pressure data, but not arterial pressure data. This embodiment does not limit the implementation method of perfusion parameter data and blood pressure parameter data.

[0155] Optionally, based on the above embodiments, since perfusion data may be a combination of at least two perfusion parameter data, and the flow detection model corresponding to different combinations needs to be trained using sample data of the corresponding combination, the flow detection models corresponding to different combinations may be different. Similarly, since the flow detection models corresponding to different combinations of at least two blood pressure parameter data may be different, before step 302, the method further includes:

[0156] Based on the data type of perfusion data and / or the data type of blood pressure data, a first flow detection model corresponding to the transcatheter ventricular assist device is determined; the data type of perfusion data is used to indicate the types of perfusion parameter data included in the perfusion data; the data type of blood pressure data is used to indicate the types of blood pressure parameter data included in the blood pressure data.

[0157] Accordingly, the perfusion data, the blood pressure data, and the motor operation data are input into a preset flow detection model for flow detection processing, and the flow data is output, including:

[0158] The perfusion data, the blood pressure data, and the motor operation data are input into the first flow detection model, and the flow data is output.

[0159] The preset flow detection model includes flow detection models corresponding to different data types. In each group of sample data corresponding to the first flow detection model, the data type of the perfusion sample data is consistent with the data type of the currently acquired perfusion data, and the data type of the blood pressure sample data is consistent with the data type of the currently acquired perfusion data.

[0160] For example, if the data type of the currently acquired perfusion data is: perfusion data includes perfusion fluid pressure data and perfusion fluid flow data, and the data type of the blood pressure data is: blood pressure data includes arterial pressure data and ventricular pressure data, then the perfusion sample data in the sample data corresponding to the first flow detection model includes perfusion fluid pressure sample data and perfusion fluid flow sample data, and the blood pressure sample data in this sample data includes arterial pressure sample data and ventricular pressure sample data.

[0161] For example, if the data type of the currently acquired perfusion data is: perfusion data includes perfusion fluid pressure data, and the data type of the blood pressure data is: blood pressure data includes arterial pressure data, then the perfusion sample data in the sample data corresponding to the first flow detection model includes perfusion fluid pressure sample data, and the blood pressure sample data in this sample data includes arterial pressure sample data.

[0162] Optionally, the data type can be received through a human-computer interaction interface; or, since the data volume of perfusion data and blood pressure data of different data types is different, the data type can also be determined based on the data volume. This embodiment does not limit the method of obtaining the data type.

[0163] In this embodiment, by setting different flow detection models for perfusion data of different data types and / or blood pressure data of different data types, a suitable first flow detection model can be determined based on the currently collected data type for flow detection processing, which can improve the accuracy of flow detection.

[0164] Optionally, based on the above embodiments, since each type of perfusion parameter data may have multiple acquisition locations, and each type of blood pressure parameter data may have multiple acquisition locations, although the data acquired at different acquisition locations can represent the same type of data, the specific values ​​acquired may be different. Therefore, data acquired at different acquisition locations can correspond to different flow detection models.

[0165] At this point, before step 302, the method further includes: acquiring the first acquisition location of the perfusion data and the second acquisition location of the blood pressure data; and determining the second flow detection model based on the first acquisition location and the second acquisition location.

[0166] The preset flow detection model includes flow detection models corresponding to different collection locations. Each set of sample data corresponding to the second flow detection model includes: perfusion sample data collected based on the first collection location, blood pressure sample data collected based on the second collection location, and motor operation sample data.

[0167] Accordingly, the perfusion data, the blood pressure data, and the motor operation data are input into a preset flow detection model for flow detection processing, and the flow data is output, including:

[0168] The perfusion data, blood pressure data, and motor operation data are input into the second flow detection model for flow detection processing, and the flow data is output.

[0169] In this embodiment, by setting flow detection models corresponding to different acquisition locations, a suitable second flow detection model can be determined based on the first acquisition location of the currently acquired perfusion data and the second acquisition location of the blood pressure data for flow detection processing, which can improve the accuracy of flow detection.

[0170] Based on the above embodiments, the preset traffic detection model in this application is obtained by training a preset machine learning model using a sample dataset. The specific training process is described below.

[0171] Figure 4 This is a flowchart of a training method for a traffic detection model provided in one embodiment of this application. The method includes at least the following steps:

[0172] Step 401: Obtain the sample dataset collected during the operation of the transcatheter ventricular assist device.

[0173] Optionally, each set of sample data in the sample dataset includes perfusion sample data, blood pressure sample data, motor operation sample data, and corresponding flow measurement data corresponding to the transcatheter ventricular assist device.

[0174] Perfusion sample data is used to characterize the perfusion state of the perfusion fluid in the perfusion assembly at the time the sample dataset is collected. In transcatheter ventricular assist devices, the perfusion fluid enters the heart through the internal gaps of the catheter. The perfusion pathway (formed by the perfusion tubing and catheter) connects to the blood vessels or the interior of the heart, i.e., to the blood flow pathway. In other words, there is a correlation between the flow of the perfusion fluid and the flow of blood. Therefore, by learning the correlation between perfusion sample data and blood flow, flow rate data can be determined based on perfusion data.

[0175] Optionally, the injection fluid sample data includes, but is not limited to, at least one of the following: injection fluid pressure data, injection fluid flow rate data, and injection pump speed data. For a detailed description of each type of injection fluid sample data, please refer to the description of the injection fluid data. This embodiment will not repeat the details here.

[0176] Blood pressure sample data is used to characterize the blood pressure status in the operating environment of the catheter-guided ventricular assist device at the time the sample dataset was collected. There is a strong correlation between blood flow and blood pressure. Therefore, by learning the strong correlation between blood pressure sample data and blood flow, flow rate data can be determined based on blood pressure sample data.

[0177] Optionally, the blood pressure sample data includes, but is not limited to, at least one of: arterial pressure sample data, ventricular pressure sample data, and pump pressure difference sample data between arterial pressure sample data and ventricular pressure sample data. For a detailed description of each type of blood pressure sample data, please refer to the description of blood pressure data. This embodiment will not repeat it here.

[0178] Motor operation sample data is used to characterize the operational efficiency of the motor in the catheter-guided ventricular assist device (VAD) during sample dataset collection. The type of fluid pumped by the VAD (e.g., viscosity) affects the motor's operational efficiency at a set speed. At the set speed, the power consumption of the motor varies depending on the type of fluid pumped. It is evident that there is a correlation between the motor operation sample data and the type of fluid pumped by the blood pump, and a correlation also exists between flow rate and fluid type. Furthermore, the motor is the power source driving the impeller rotation; therefore, by learning the correlation between motor operation sample data and blood flow rate, flow rate data can be determined based on the motor operation sample data.

[0179] Optionally, the motor operation sample data includes, but is not limited to, at least one of motor speed sample data and motor current sample data. For a detailed description of each type of motor operation sample data, please refer to the description of motor operation data. This embodiment will not repeat the details here.

[0180] The perfusion fluid sample data mentioned above may be collected by the first detection unit, the blood pressure sample data may be collected by the second detection unit, and the motor operation sample data may be collected by the third detection unit; or, at least one of the perfusion fluid sample data, blood pressure sample data, and motor operation sample data may be collected by other devices. This embodiment does not limit the data collection method in the sample data set.

[0181] When collecting sample datasets, the operating environment of the transcatheter ventricular assist device can be a real ventricle or a simulated ventricle. This embodiment does not limit the collection environment of the sample datasets.

[0182] The perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data in each sample data set correspond to each other. Illustratively, the perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data in each sample data set are collected synchronously; or, the maximum time interval between data items in a sample data set is less than a threshold.

[0183] In one possible implementation, the aforementioned flow measurement data can be obtained by detecting a flow sensor.

[0184] In another possible implementation, flow measurement data is determined by the change in liquid weight over a unit of time. For example, flow measurement data can be obtained by measuring the change in liquid weight using a balance. This eliminates the influence of flow sensor measurement error on the accuracy of flow measurement data. The flow detection model trained using this flow measurement data can be free from dependence on the accuracy of the flow sensor, and the model accuracy can exceed the measurement accuracy of the flow sensor, thereby improving the accuracy of determining flow data.

[0185] In one example, the transcatheter ventricular assist device includes a regulating device for adjusting the pumping flow rate or speed of the interventional catheter pump. Accordingly, a sample dataset collected during the operation of the transcatheter ventricular assist device is acquired, including:

[0186] The control and adjustment device operates in a first state, acquiring at least one set of sample data in the first state; the operating state of the adjustment device is adjusted from the first state to a second state, and at least one set of sample data in the second state is acquired; wherein the pumping speed corresponding to the second state is greater than the pumping speed corresponding to the first state, or the pumping speed corresponding to the second state is less than the pumping speed of the first state; it is determined whether the number of adjustment cycles of the operating state has reached a preset number; if the preset number of adjustment cycles has not been reached, the second state is used as the first state, and the step of adjusting the operating state of the adjustment device from the first state to the second state and acquiring at least one set of sample data in the second state is executed again; until the number of adjustment cycles reaches the preset number, or the adjustment reaches the maximum or minimum value, the process stops, and a sample dataset is obtained.

[0187] At this point, by operating the adjustment device to collect sample data at different pumping speeds during sample data collection, the richness of the sample dataset can be ensured, thereby improving the performance of the trained model.

[0188] In other embodiments, the working state of the adjustment device can be randomly set when collecting sample data. This embodiment does not limit the method of collecting sample data.

[0189] Optionally, the sample dataset collected during the operation of the transcatheter ventricular assist device includes: obtaining sample data of different models of transcatheter ventricular assist devices to obtain a sample dataset.

[0190] At this point, when collecting sample data, by collecting sample data from different models of transcatheter ventricular assist devices, the flow detection model for the corresponding model can be trained using the sample data of each type of transcatheter ventricular assist device, which can improve the accuracy of flow detection.

[0191] In some implementations, the sample dataset does not impose constraints on liquid type. For example, the sample dataset does not impose viscosity range constraints on liquid viscosity; the liquid viscosity corresponding to each group of sample data in the sample dataset is not limited by a preset viscosity range. Since the sample dataset contains data samples corresponding to various liquid types, the flow detection model trained based on these data is a general flow detection model for all liquid types and is an independent and single model. Step 402: The preset machine learning model is trained based on the perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data to obtain the flow detection model.

[0192] Among them, the flow detection model is used to detect the flow data corresponding to the transcatheter ventricular assist device based on the perfusion data, blood pressure data and motor operation data generated during the application of the transcatheter ventricular assist device.

[0193] In one example, the sample dataset includes multiple sets of training data and multiple sets of test data. A pre-defined machine learning model is trained using perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data to obtain a flow detection model. This process includes: modeling the perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data from the multiple sets of training data to generate a first machine learning model; inputting the perfusion sample data, blood pressure sample data, and motor operation sample data from each set of test data into the first machine learning model for flow detection processing, and outputting the flow detection data corresponding to each set of test data; comparing the flow measurement data in each set of test data with the corresponding flow detection data to determine the loss information corresponding to the first machine learning model; and adjusting the model parameters of the first machine learning model based on the loss information to obtain the flow detection model.

[0194] Optionally, the first machine learning model includes a Gaussian model. In one possible implementation, the first machine learning model is a Gaussian model modeled based on a first sample dataset; correspondingly, the traffic detection model is a parameter-adjusted Gaussian model. Since the sample data of the Gaussian model follows a normal distribution and has statistical significance, the correlation between the parameters can be learned; therefore, a Gaussian model can be selected for traffic detection.

[0195] Optionally, the flow measurement data in each set of test data and the corresponding flow detection data are compared to determine the loss information corresponding to the first machine learning model. This includes: determining the average flow data based on the flow measurement data in each set of sample data; and comparing the average flow data, the corresponding flow detection data, and the flow measurement data in each set of test data to obtain the variance data and root mean square error data corresponding to the Gaussian model. In this case, the loss information includes the variance data and the root mean square error data.

[0196] Variance data r 2 The calculation formula can be expressed by the following formula:

[0197]

[0198] Where n represents the total number of test data sets; i represents the i-th test data set, where i is an integer from 1 to n. This represents the flow measurement data in the i-th test data set; This represents the average traffic data corresponding to the traffic detection data of n sets of test data; This represents the traffic detection data corresponding to the i-th set of test data.

[0199] The root mean square error (RMSE) can be calculated using the following formula:

[0200]

[0201] Where n represents the total number of test data sets; i represents the i-th test data set, where i is an integer from 1 to n. This represents the flow measurement data in the i-th test data set; This represents the traffic detection data corresponding to the i-th set of test data.

[0202] Optionally, when using both variance and root mean square error to determine loss information, variance and root mean square error also have corresponding weighting parameters. In this case, the weighted sum of variance and root mean square error is determined to obtain the loss information. Alternatively, the sum of variance and root mean square error can also be used to determine the loss information. This embodiment does not limit the calculation method of loss information.

[0203] During the training of the Gaussian model, the Gaussian model corresponds to multiple sets of model parameters. The loss data corresponding to each set of parameters is determined. Based on the loss data corresponding to each set of model parameters, the target model parameters with the highest traffic detection accuracy (i.e., the smallest loss data) are determined. The Gaussian model is configured according to the target model parameters to obtain the above traffic detection model.

[0204] In another possible implementation, the machine learning model is a neural network model. The above-described adjustment of the model parameters of the first machine learning model based on loss information to obtain the flow detection model includes: determining whether the loss value represented by the loss information is greater than a loss threshold, and whether the number of adjustments to the model parameters is less than or equal to a threshold; if the loss value is greater than the loss threshold, or the number of adjustments is less than the threshold, then adjusting the model parameters of the first machine learning model and triggering the execution of inputting pump load sample data and motor operation sample data from each set of test data into the first machine learning model for flow detection processing, and outputting the flow detection data corresponding to each set of test data; comparing the flow measurement data in each set of test data with the flow detection data corresponding to each set of test data to determine the loss information corresponding to the first machine learning model; if the loss value is less than or equal to the loss threshold, or the number of adjustments is greater than or equal to the threshold, then outputting the adjusted model parameters to obtain the flow detection model.

[0205] In this embodiment, a flow detection model is obtained by pre-training a sample dataset. This model is then used to determine the flow data of the transcatheter ventricular assist device (VTAVD), which greatly reduces the dependence of flow detection on flow sensors. This effectively solves the technical problem of difficulty in detecting the pumping flow inside the heart by the VTAVD, reduces the complexity of flow detection, and improves the efficiency of flow detection.

[0206] Furthermore, since the perfusion state of the infusion fluid, the pressure state of the blood, and the operating state of the motor can all affect or reflect the pumping flow rate of the interventional catheter pump, training a flow detection model based on perfusion sample data characterizing the liquid state of the infusion fluid, blood pressure sample data characterizing the pressure state of the blood, and motor operating sample data characterizing the operating state of the motor can ensure the accuracy of the flow detection model in determining the flow rate data.

[0207] In addition, traffic data is detected through a traffic detection model, which is a machine learning model trained on sample data. Machine learning models can learn relatively accurate correspondences between data, thus ensuring the accuracy of traffic detection results.

[0208] In addition, by implementing the machine learning model as a Gaussian model, traffic detection results that conform to a normal distribution can be obtained, improving the accuracy of traffic detection. Furthermore, compared to neural network models with a large number of parameters, Gaussian models have a relatively small computational cost, thus improving the computational efficiency of traffic detection and saving computing resources.

[0209] Based on the above embodiments, if the perfusion data includes at least one perfusion parameter data, the at least one perfusion parameter data including perfusion fluid pressure data; and the blood pressure data includes at least one blood pressure parameter data, the at least one blood pressure parameter data including arterial pressure data, then the perfusion sample data in the sample dataset includes perfusion fluid pressure sample data corresponding to the catheter-guided ventricular assist device, and the blood pressure sample data in the sample dataset includes arterial pressure sample data corresponding to the catheter-guided ventricular assist device.

[0210] Accordingly, based on perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data, a pre-set machine learning model is trained to obtain a flow detection model, including:

[0211] The pre-set machine learning model is trained based on perfusion pressure sample data, arterial pressure sample data, motor operation sample data, and flow measurement data to obtain a flow detection model.

[0212] In one possible implementation, no pressure sensor is installed on the catheter of the transcatheter ventricular assist device, making it difficult to obtain ventricular pressure data. However, arterial pressure sample data can be obtained externally. In this case, the flow detection model can be trained based solely on easily detectable arterial pressure sample data, perfusion fluid pressure sample data, and motor operation sample data. This also allows for accurate output of the flow rate of the interventional catheter pump, reducing the difficulty of flow detection.

[0213] The aforementioned arterial pressure sample data is of the same data type as arterial pressure data. This arterial pressure sample data can be aortic pressure (with left ventricular assist), pulmonary artery pressure data (with right ventricular assist), or arterial pressure sample data from other locations. This application does not limit this.

[0214] In the above embodiments, the motor operating data includes at least one of motor speed data and motor current data. When the motor operating data includes both motor speed and motor current, the motor operating sample data in the sample dataset includes motor speed sample data and motor current sample data corresponding to the catheter-based ventricular assist device.

[0215] In this case, a flow detection model is obtained by training a preset machine learning model based on perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data. This includes training a preset machine learning model based on perfusion fluid pressure sample data, arterial pressure sample data, motor speed sample data, motor current sample data, and flow measurement data to obtain the flow detection model.

[0216] Accordingly, the flow detection model is specifically used to detect flow data based on perfusion pressure sample data, arterial pressure sample data, motor speed data, and motor current data generated during the application of the transcatheter ventricular assist device.

[0217] The aforementioned motor current sample data and motor speed sample data can have a temporal correspondence. The motor current sample data can reflect the motor's power consumption under the corresponding motor speed sample data, thus reflecting the motor's operating efficiency under the current liquid type. Therefore, using specific motor current sample data and motor speed sample data as motor operating data can effectively characterize the current driving motor's operating efficiency, thereby ensuring the accuracy of the flow detection model obtained through training in determining the flow data.

[0218] Since there is a correlation between perfusion fluid pressure data, arterial pressure data and the pumping flow rate of the interventional catheter pump, in this embodiment, the flow detection model is trained based on perfusion fluid pressure sample data and arterial pressure sample data, which can ensure the accuracy of the flow detection model in detecting flow data.

[0219] Optionally, in addition to the injection fluid pressure data, at least one injection parameter data may also include injection fluid flow rate data and / or injection pump speed data, etc. In this case, the injection sample data may also include injection fluid flow rate sample data and / or injection pump speed sample data.

[0220] Optionally, in addition to arterial pressure data, at least one blood pressure parameter may also include ventricular pressure data and / or differential pressure data, in which case the blood pressure sample data may also include ventricular pressure sample data and / or differential pressure sample data.

[0221] In other embodiments, at least one perfusion parameter data may include perfusion fluid flow rate data and / or perfusion pump speed data, but not perfusion fluid pressure data; in this case, the perfusion sample data includes perfusion fluid flow rate sample data and / or perfusion pump speed sample data, but not perfusion fluid pressure sample data. And / or, at least one blood pressure parameter data may include ventricular pressure data and / or differential pressure data, but not arterial pressure data; in this case, the perfusion sample data includes ventricular pressure sample data and / or differential pressure sample data, but not arterial pressure sample data. This embodiment does not limit the implementation method of perfusion sample data and blood pressure sample data.

[0222] Optionally, based on the above embodiments, since perfusion data may be a combination of at least two perfusion parameter data, and blood pressure data may be a combination of at least two blood pressure parameter data, different combinations result in different data types, and the corresponding flow detection models may differ. Therefore, it is necessary to train flow detection models corresponding to different data types.

[0223] At this point, the sample dataset collected during the operation of the transcatheter ventricular assist device is acquired, including: perfusion sample data of different data types and blood pressure sample data of different data types.

[0224] Accordingly, a pre-set machine learning model is trained based on perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data to obtain a flow detection model. This includes: training a pre-set machine learning model based on perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data of the same data type to obtain a flow detection model corresponding to the data type of the perfusion sample data and the data type of the blood pressure sample data.

[0225] In this embodiment, by training a corresponding flow detection model based on perfusion sample data and / or blood pressure sample data of each data type, the transcatheter ventricular assist device can determine the appropriate first flow detection model based on the currently collected data type for flow detection processing, thereby improving the accuracy of flow detection.

[0226] Optionally, based on the above embodiments, data collected from different collection locations can correspond to different traffic detection models. Therefore, it is necessary to train the traffic detection models corresponding to different collection locations.

[0227] At this time, the sample dataset collected during the operation of the transcatheter ventricular assist device is acquired, including: perfusion sample data collected at different first acquisition positions and blood pressure sample data collected at different second acquisition positions.

[0228] Accordingly, based on perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data, a pre-set machine learning model is trained to obtain a flow detection model, including:

[0229] Based on the perfusion sample data collected at the same first acquisition location, the blood pressure sample data collected at the same second acquisition location, the motor operation sample data, and the flow measurement data, a preset machine learning model is trained to obtain the flow detection model corresponding to the first acquisition location and the second acquisition location.

[0230] In this embodiment, a corresponding flow detection model is trained based on the perfusion sample data collected at each first acquisition location and the blood pressure sample data collected at each second acquisition location. This allows the transcatheter ventricular assist device to determine a suitable second flow detection model based on the first acquisition location of the currently acquired perfusion data and the second acquisition location of the blood pressure data for flow detection processing, thereby improving the accuracy of flow detection.

[0231] Figure 5 This is a block diagram of a flow determination device according to an embodiment of this application, which is applied to a transcatheter ventricular assist device. The device includes at least the following modules: a data acquisition module 510 and a flow determination module 520.

[0232] Data acquisition module 510 is used to acquire perfusion data, blood pressure data and motor operation data corresponding to the transcatheter ventricular assist device;

[0233] The flow determination module 520 is used to perform flow detection processing based on the perfusion data, the blood pressure data and the motor operation data to obtain the flow data corresponding to the transcatheter ventricular assist device.

[0234] Optionally, the perfusion data includes at least one perfusion parameter data, which includes at least one of perfusion fluid pressure data, perfusion fluid flow rate data, and perfusion pump speed data. The blood pressure data includes at least one blood pressure parameter data, which includes at least one of arterial pressure data, ventricular pressure data, and pump pressure differential data.

[0235] Optionally, the flow determination module 520 is used for:

[0236] The flow rate data is obtained by performing flow rate detection processing based on the perfusion fluid pressure data, the arterial pressure data, and the motor operation data.

[0237] Optionally, the motor operating data includes motor speed data and motor current data, and the flow determination module 520 is used for:

[0238] The flow rate data is obtained by performing flow rate detection processing based on the perfusion fluid pressure data, the arterial pressure data, the motor speed data, and the motor current data.

[0239] Optionally, the flow determination module 520 includes a determination submodule 521.

[0240] The determining submodule 521 is used to input the perfusion data, the blood pressure data and the motor operation data into a preset flow detection model for flow detection processing, and output the flow data;

[0241] The flow detection model is a machine learning model trained on multiple sets of sample data. Each set of sample data includes corresponding perfusion sample data, blood pressure sample data, motor operation sample data, and sample flow data.

[0242] Optionally, the determining submodule 521 is configured to:

[0243] The perfusion fluid pressure data, arterial pressure data, motor speed data, and motor current data are input into a preset Gaussian model for flow detection processing, and the flow data is output.

[0244] The perfusion data includes the perfusion fluid pressure data, the blood pressure data includes the arterial pressure data, the motor operation data includes the motor speed data and the motor current data, and the preset flow detection model includes the preset Gaussian model, which is a Gaussian model obtained by training multiple sets of sample data.

[0245] For relevant details, please refer to the above method implementation examples.

[0246] It should be noted that the flow determination device provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the flow determination device can be divided into different functional modules to complete all or part of the functions described above. In addition, the flow determination device and the flow determination method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0247] Figure 6 This is a block diagram of a training apparatus for a traffic detection model provided in one embodiment of this application. The apparatus includes at least the following modules: a data acquisition module 610 and a model training module 620.

[0248] The data acquisition module 610 is used to acquire sample datasets collected during the operation of the transcatheter ventricular assist device. Each set of sample data in the sample dataset includes perfusion sample data, blood pressure sample data, motor operation sample data and corresponding flow measurement data corresponding to the transcatheter ventricular assist device.

[0249] The model training module 620 is used to train a preset machine learning model based on the perfusion sample data, the blood pressure sample data, the motor operation sample data, and the flow measurement data to obtain a flow detection model.

[0250] The flow detection model is used to detect the flow data corresponding to the transcatheter ventricular assist device based on the perfusion data, blood pressure data, and motor operation data generated during the application of the transcatheter ventricular assist device.

[0251] Optionally, the sample dataset includes multiple sets of training data and multiple sets of test data, and the model training module 620 includes: a model generation submodule 621, a traffic detection submodule 622, a loss calculation submodule 623, and a parameter tuning submodule 624.

[0252] The model generation submodule 621 is used to perform modeling processing based on the perfusion sample data, blood pressure sample data, motor operation sample data and flow measurement data in the multiple sets of training data to generate a first machine learning model.

[0253] The flow detection submodule 622 is used to input the perfusion sample data, blood pressure sample data and motor operation sample data in each set of test data into the first machine learning model for flow detection processing, and output the flow detection data corresponding to each set of test data.

[0254] The loss calculation submodule 623 is used to compare the traffic measurement data in each set of test data with the traffic detection data corresponding to each set of test data to determine the loss information corresponding to the first machine learning model.

[0255] The parameter adjustment submodule 624 is used to adjust the model parameters of the first machine learning model based on the loss information to obtain the traffic detection model.

[0256] Optionally, the first machine learning model includes a Gaussian model, and the loss calculation submodule 623 is used for:

[0257] The average flow rate is determined based on the flow rate measurement data in each group of sample data;

[0258] The average flow data, the flow detection data corresponding to each set of test data, and the flow measurement data in each set of test data are compared to obtain the variance data and root mean square error data corresponding to the Gaussian model. The loss information includes the variance data and the root mean square error data.

[0259] Optionally, the flow measurement data is determined by the change in liquid weight per unit time.

[0260] Optionally, the perfusion sample data in the sample dataset includes perfusion fluid pressure sample data corresponding to the transcatheter ventricular assist device, the blood pressure sample data in the sample dataset includes arterial pressure sample data corresponding to the transcatheter ventricular assist device, and the motor operation sample data in the sample dataset includes motor speed sample data and motor current sample data corresponding to the transcatheter ventricular assist device.

[0261] Accordingly, the flow detection model is specifically used to detect the flow data based on the perfusion fluid pressure data, arterial pressure data, motor speed data, and motor current data generated during the application of the transcatheter ventricular assist device.

[0262] For relevant details, please refer to the above method implementation examples.

[0263] It should be noted that the training device for the traffic detection model provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the traffic detection model training device can be divided into different functional modules to complete all or part of the functions described above. In addition, the training device for the traffic detection model provided in the above embodiments and the training method embodiments for the traffic detection model belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0264] Figure 7 This is a block diagram of a computer device provided in one embodiment of this application. The computer device includes at least... Figure 1 The control unit in the illustrated transcatheter ventricular assist device can, in other embodiments, be another device communicatively connected to the transcatheter ventricular assist device, such as a computer, tablet, mobile phone, or other computer device with processing capabilities. This computer device includes at least a processor 701 and a memory 702.

[0265] Processor 701 may include one or more processing cores, such as a quad-core processor or an octa-core processor. Processor 701 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 701 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 701 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 701 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0266] The memory 702 may include one or more computer-readable storage media, which may be non-transitory. The memory 702 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 702 are used to store at least one instruction, which is executed by the processor 701 to implement the traffic determination method or the traffic detection model training method provided in the method embodiments of this application.

[0267] In some embodiments, the computer device may also optionally include: a peripheral device interface and at least one peripheral device. The processor 701, memory 702, and peripheral device interface can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface via a bus, signal line, or circuit board. Indicatively, peripheral devices include, but are not limited to: radio frequency circuitry, a touch display screen, audio circuitry, and a power supply.

[0268] Of course, computer devices may also include fewer or more components, and this embodiment does not limit this.

[0269] Optionally, this application also provides a computer-readable storage medium storing a program that is loaded and executed by a processor to implement the traffic determination method or the traffic detection model training method of the above method embodiments.

[0270] Optionally, this application also provides a computer product including a computer-readable storage medium storing a program that is loaded and executed by a processor to implement the traffic determination method or the traffic detection model training method of the above method embodiments.

[0271] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0272] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A flow rate determination method characterized by, The method is applied to a trans-catheter ventricular assist device, and the method comprises: acquiring perfusion data, blood pressure data corresponding to the trans-catheter ventricular assist device, and motor operation data corresponding to the trans-catheter ventricular assist device; performing flow detection processing according to the perfusion data, the blood pressure data, and the motor operation data to obtain flow data corresponding to the trans-catheter ventricular assist device.

2. The method of claim 1, wherein, The perfusion data comprises at least one perfusion parameter data, and the at least one perfusion parameter data comprises at least one of perfusion liquid pressure data, perfusion liquid flow data, and perfusion pump rotating speed data; the blood pressure data comprises at least one blood pressure parameter data, and the at least one blood pressure parameter data comprises at least one of arterial pressure data, ventricular pressure data, and pumping pressure difference data.

3. The method of claim 2, wherein, The flow detection processing according to the perfusion data, the blood pressure data, and the motor operation data to obtain the flow data corresponding to the trans-catheter ventricular assist device comprises: performing flow detection processing according to the perfusion liquid pressure data, the arterial pressure data, and the motor operation data to obtain the flow data.

4. The method of claim 2, wherein, The motor operation data comprises motor rotating speed data and motor current data, and the flow detection processing according to the perfusion liquid pressure data, the arterial pressure data, and the motor operation data to obtain the flow data comprises: performing flow detection processing according to the perfusion liquid pressure data, the arterial pressure data, the motor rotating speed data, and the motor current data to obtain the flow data.

5. The method according to any one of claims 2 to 4, characterized in that, The trans-catheter ventricular assist device comprises: an interventional catheter pump, the interventional catheter pump comprising a catheter, a pump head connected to a distal end of the catheter, and a coupling assembly connected to a proximal end of the catheter, the pump head being capable of being delivered by the catheter to a desired position of a heart for blood pumping operation; a perfusion channel, the perfusion channel at least extending through the catheter, an inlet of the perfusion channel being arranged on the coupling assembly, and an outlet of the perfusion channel being arranged at the pump head; perfusion liquid in the perfusion channel enters a desired position in a cardiovascular system through the outlet; the desired position in the cardiovascular system comprises at least one of an aorta, a pulmonary artery, a left ventricle, a right ventricle, a left atrium, and a right atrium; the perfusion liquid pressure data characterizing perfusion liquid pressure generated from a pressure collection position to the outlet.

6. The method of claim 1, wherein, The flow detection processing according to the perfusion data, the blood pressure data, and the motor operation data to obtain the flow data corresponding to the trans-catheter ventricular assist device comprises: inputting the perfusion data, the blood pressure data, and the motor operation data into a preset flow detection model for flow detection processing, and outputting the flow data; wherein the flow detection model is a machine learning model obtained by training on a plurality of groups of sample data, and each group of sample data comprises corresponding perfusion sample data, blood pressure sample data, motor operation sample data, and sample flow data.

7. The method of claim 6, wherein, The inputting the perfusion data, the blood pressure data, and the motor operation data into a preset flow detection model for flow detection processing, and outputting the flow data comprises: The perfusion liquid pressure data, the arterial pressure data, the motor speed data and the motor current data are input into a preset Gaussian model for flow detection processing, and the flow data is output. The perfusion data includes the perfusion liquid pressure data, the blood pressure data includes the arterial pressure data, the motor operation data includes the motor speed data and the motor current data, the preset flow detection model includes the preset Gaussian model, and the preset Gaussian model is a Gaussian model obtained after training based on multiple sets of sample data. 8.A method for training a traffic detection model, the method comprising: The method comprises: obtaining a sample data set collected during operation of a trans-catheter ventricular assist device, each set of sample data in the sample data set comprising perfusion sample data, blood pressure sample data, motor operation sample data corresponding to the trans-catheter ventricular assist device, and corresponding flow measurement data; training a preset machine learning model based on the perfusion sample data, the blood pressure sample data, the motor operation sample data, and the flow measurement data to obtain a flow detection model; The flow detection model is used to detect the flow data corresponding to the trans-catheter ventricular assist device based on perfusion data, blood pressure data, and motor operation data generated during application of the trans-catheter ventricular assist device.

9. The method of claim 8, wherein, The sample data set includes multiple sets of training data and multiple sets of test data, and the training of the preset machine learning model based on the perfusion sample data, the blood pressure sample data, the motor operation sample data, and the flow measurement data to obtain a flow detection model comprises: modeling based on perfusion sample data, blood pressure sample data, motor operation sample data, and flow measurement data in the multiple sets of training data to generate a first machine learning model; inputting the perfusion sample data, the blood pressure sample data, and the motor operation sample data in each set of test data into the first machine learning model for flow detection processing, and outputting the flow detection data corresponding to each set of test data; comparing the flow measurement data in each set of test data with the flow detection data corresponding to each set of test data to determine loss information corresponding to the first machine learning model; adjusting the model parameters of the first machine learning model based on the loss information to obtain the flow detection model.

10. The method of claim 9, wherein, The first machine learning model includes a Gaussian model, and the comparison of the flow measurement data in each set of test data with the flow detection data corresponding to each set of test data to determine the loss information corresponding to the first machine learning model comprises: determining average flow data from the flow measurement data in each set of sample data; comparing the average flow data, the flow detection data corresponding to each set of test data, and the flow measurement data in each set of test data to obtain variance data and root mean square error data corresponding to the Gaussian model, and the loss information includes the variance data and the root mean square error data.

11. The method according to any one of claims 8 to 10, characterized in that, The flow measurement data is determined by the change in liquid weight per unit time.

12. A flow determining device, characterized in that The device is applied to a trans-catheter ventricular assist device, and the device comprises: The data acquisition module is configured to acquire perfusion data, blood pressure data corresponding to the trans-catheter ventricular assist device, and motor operation data corresponding to the trans-catheter ventricular assist device. The flow rate determination module is configured to perform flow rate detection processing according to the perfusion data, the blood pressure data, and the motor operation data to obtain flow rate data corresponding to the trans-catheter ventricular assist device. 13.A device for training a traffic detection model, characterized in that, The device comprises: The data acquisition module is configured to acquire a sample data set collected during operation of the trans-catheter ventricular assist device, each set of sample data in the sample data set comprising perfusion sample data, blood pressure sample data, motor operation sample data corresponding to the trans-catheter ventricular assist device, and corresponding flow rate measurement data. The model training module is configured to perform model training on a preset machine learning model according to the perfusion sample data, the blood pressure sample data, the motor operation sample data, and the flow rate measurement data to obtain a flow rate detection model. The flow rate detection model is configured to detect flow rate data corresponding to the trans-catheter ventricular assist device according to perfusion data, blood pressure data, and motor operation data generated during application of the trans-catheter ventricular assist device.

14. A catheter-based ventricular assist device, characterized by The trans-catheter ventricular assist device comprises a computer device, the computer device comprising a processor and a memory; the memory stores a program, the program is loaded and executed by the processor to implement the flow rate determination method according to any one of claims 1 to 6; or, implement the training method of the flow rate detection model according to any one of claims 7 to 10.

15. A computer device, comprising: The device comprises a processor and a memory; the memory stores a program, the program is loaded and executed by the processor to implement the flow rate determination method according to any one of claims 1 to 6; or, implement the training method of the flow rate detection model according to any one of claims 7 to 10.

16. A computer readable storage medium characterized by: The storage medium stores a program, the program is loaded and executed by the processor to implement the flow rate determination method according to any one of claims 1 to 6; or, implement the training method of the flow rate detection model according to any one of claims 7 to 10.

Images