Magnetocardiogram data processing device and method, electronic equipment and medium
By designing a device that uses deep learning models to process cardiac magnetic data, the problem of difficulty in obtaining the degree of cardiac vascular stenosis and the degree of ischemia in the area of the heart position through a single examination is solved, and efficient prediction results are achieved, filling the gap in the clinical lack of this method.
Patent Information
- Application Number
- CN202311805204.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-25
- Publication Date
- 2025-06-27
AI Technical Summary
The prior art is difficult to obtain predictions of the degree of cardiac vascular stenosis and the degree of ischemia in the area of the heart through a single examination, resulting in increased patient examination burden and inconsistent information.
A cardiac magnetic data processing device is designed to process cardiac magnetic data using deep learning models. Through training data, the predicted results of cardiac vascular stenosis degree and cardiac position area ischemia index, and cardiac position area area ischemia index are obtained.
The prediction results of simultaneously obtaining the degree of cardiac vascular stenosis and the degree of ischemia in the area of cardiac position based on a single cardiac magnetic examination are realized, filling the gap in clinical lack of this method, and the method is based on non-invasive cardiac magnetic examination data.
Smart Images
Figure CN120203591A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the technical field of magnetocardiogram signal processing, and relates to a magnetocardiogram data processing device, in particular to a magnetocardiogram data processing device, method, electronic device and medium. Background Art
[0002] Magnetocardiogram (MCG) is a method for detecting and recording the magnetic field signals emitted by the human heart itself. The examination process is non-invasive, radiation-free and does not require drug injection, and has the characteristics of fast imaging speed and high sensitivity. It is applicable to the clinical diagnosis of coronary heart disease, prognosis evaluation, and the auxiliary positioning of arrhythmia. Recent clinical studies have shown that magnetocardiogram can be used in the diagnosis and evaluation of coronary microcirculation disorders. Coronary angiography (CAG) is an invasive examination that evaluates the stenosis of coronary arteries by injecting contrast agents into the coronary arteries and using X-ray imaging. Single photon emission computed tomography (SPECT) is a nuclear medicine examination that evaluates the degree of myocardial ischemia by radioactive isotope tracing of myocardial perfusion. The latter two examinations have certain side effects and risks, such as allergic reactions and vascular injuries.
[0003] Currently, multiple examinations are usually required to obtain comprehensive evaluation information. However, this approach not only increases the examination burden on patients but also may result in inconsistent information. Therefore, there is still a lack of a processing device that can simultaneously obtain the prediction results of the stenosis degree of the cardiac blood vessels and the ischemia degree of the cardiac position area based on a single examination. Summary of the Invention
[0004] In a first aspect, this application provides a magnetocardiogram data processing device, which includes: a data acquisition module for acquiring magnetocardiogram data; a data processing module for obtaining a processing result of the magnetocardiogram data by using a deep learning model; wherein, the training data of the deep learning model includes magnetocardiogram data, as well as an index of the stenosis degree of the cardiac blood vessels and an index of the ischemia degree of the cardiac position area, and the processing result includes a prediction result of the stenosis degree of the cardiac blood vessels and a prediction result of the ischemia degree of the cardiac position area.
[0005] In this application, the cardiac magnetic data is used as the model input, and the stenosis degree index of the cardiac blood vessels and the ischemia degree index of the cardiac position area are used as labels to train a deep learning model, so as to use the trained deep learning model to process the cardiac magnetic data and obtain the stenosis degree prediction result of the cardiac blood vessels and the ischemia degree prediction result of the cardiac position area. Such a cardiac magnetic data processing device can simultaneously obtain the stenosis degree of the cardiac blood vessels and the ischemia degree prediction result of the cardiac position area based on a single cardiac magnetic examination, filling the gap in clinical methods that lack a way to simultaneously obtain the stenosis degree of the cardiac blood vessels and the ischemia degree of the cardiac position area based on one examination, and this method is completed based on non-invasive cardiac magnetic examination data.
[0006] In one implementation manner of the first aspect, the deep learning model includes a three-dimensional convolution module, a connection layer, a graph convolution module, a ViT module, and a prediction head.
[0007] In one implementation manner of the first aspect, the input of the graph convolution module includes the adjacency matrix of the cardiac blood vessels and the cardiac position area.
[0008] In one implementation manner of the first aspect, the connection layer includes a first convolution unit, a second convolution unit, and a third convolution unit, and the convolution kernel sizes of the convolution units are different.
[0009] In one implementation manner of the first aspect, the cardiac magnetic data processing device further includes a model training module, and the model training module is used to train the deep learning model, including: a first index acquisition unit for acquiring the ischemia degree index of the cardiac position area; a second index acquisition unit for acquiring the stenosis degree index of the cardiac blood vessels; an adjacency matrix acquisition unit for acquiring the adjacency matrix, which is obtained by using the bull's-eye diagram model of the heart, and the bull's-eye diagram model is used to represent the relative relationship between the cardiac blood vessels and the heart; a model training unit for training the deep learning model by using the cardiac magnetic signal, the stenosis degree index of the cardiac blood vessels, and the ischemia degree index of the cardiac position area.
[0010] In one implementation manner of the first aspect, each of the cardiac position areas and the cardiac blood vessels corresponds to at least one segment of the bull's-eye diagram model of the heart.
[0011] In one implementation of the first aspect, the prediction results of the stenosis degree of the cardiac blood vessels include: the prediction result of the stenosis degree of the left anterior descending artery, the prediction result of the stenosis degree of the left circumflex artery, and the prediction result of the stenosis degree of the right coronary artery; the prediction results of the ischemia degree of the cardiac position area include: the prediction result of the ischemia degree of the apical position area, the prediction result of the ischemia degree of the anterior wall position area, the prediction result of the ischemia degree of the septal wall position area, the prediction result of the ischemia degree of the inferior wall position area, and the prediction result of the ischemia degree of the lateral wall position area.
[0012] In a second aspect, the present application provides a method for processing magnetocardiogram data, including: acquiring magnetocardiogram data; obtaining a processing result of the magnetocardiogram data by using a deep learning model; wherein, the training data of the deep learning model includes magnetocardiogram signals and indexes of the stenosis degree of cardiac blood vessels and indexes of the ischemia degree of the cardiac position area, and the processing result includes the prediction result of the stenosis degree of the cardiac blood vessels and the prediction result of the ischemia degree of the cardiac position area.
[0013] In a third aspect, the present application provides an electronic device, which includes: a memory for storing a computer program; a processor for executing the computer program stored in the memory, so that the electronic device executes the magnetocardiogram data processing method as described in the second aspect.
[0014] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the magnetocardiogram data processing method as described in the second aspect is implemented. Description of the Drawings
[0015] Figure 1 It shows a schematic diagram of an application scenario of the magnetocardiogram data processing device described in the present application.
[0016] Figure 2 It shows a schematic diagram of the structure of the magnetocardiogram data processing device according to an embodiment of the present application.
[0017] Figure 3 It shows a schematic diagram of the structure of the deep learning model according to an embodiment of the present application.
[0018] Figure 4 It shows a schematic diagram of the structure of the connection layer module according to an embodiment of the present application.
[0019] Figure 5 It shows a schematic diagram of the structure of the magnetocardiogram data processing device according to an embodiment of the present application.
[0020] Figure 6A It shows a schematic diagram of the structure of the bull's-eye diagram model according to an embodiment of the present application.
[0021] Figure 6BSchematic diagram showing the adjacency matrix according to an embodiment of the present application.
[0022] Figure 6C Schematic diagram showing the adjacency matrix according to an embodiment of the present application.
[0023] Figure 7 Schematic flowchart showing the method for processing magnetocardiogram data according to an embodiment of the present application.
[0024] Figure 8 Schematic diagram showing the structure of the electronic device according to an embodiment of the present application.
[0025] Description of component numbers
[0026] 1 Coronary artery stenosis degree index and myocardial ischemia degree prediction device
[0027] 11 Magnetocardiogram examination device
[0028] 12 Processor
[0029] 13 Display terminal
[0030] 100 Magnetocardiogram data processing device
[0031] 110 Data acquisition module
[0032] 120 Data processing module
[0033] 130 Model training module
[0034] 131 First index acquisition unit
[0035] 132 Second index acquisition unit
[0036] 133 Adjacency matrix acquisition unit
[0037] 134 Model training unit
[0038] 800 Electronic device
[0039] 810 Memory
[0040] 820 Processor
[0041] 830 Display
[0042] Steps S11 - S12 Detailed implementation manner
[0043] The following describes the implementation manners of the present application through specific specific examples. Those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.
[0044] Figure 1 Shown is a schematic diagram of an application scenario of the magnetocardiogram data processing device described in the present application. The coronary artery stenosis degree index and myocardial ischemia degree prediction device 1 can be used to deploy the magnetocardiogram data processing device provided in the embodiments of the present application, but the application scenarios of the magnetocardiogram data processing device provided in the embodiments of the present application are not limited to Figure 1 the coronary artery stenosis degree index and myocardial ischemia degree prediction device 1 shown. As Figure 1 shown, the coronary artery stenosis degree index and myocardial ischemia degree prediction device 1 includes a magnetocardiogram examination device 11, a processor 12, and a display terminal 13. The magnetocardiogram data processing device 1 provided in the embodiments of the present application can be deployed in the processor 12.
[0045] Among them, Figure 1 the server 11 in can be a processor or a processor cluster composed of multiple processors or a cloud computing center, etc., and are not specifically limited here. Although Figure 1 only one magnetocardiogram examination device 11, 1 processor 12, and one display terminal 13 are shown in, it should be understood that Figure 1 the examples in are only used to understand the solution, and the specific numbers of the display terminal and the server should be flexibly determined according to the actual situation. In some other implementation manners, the myocardial ischemia index prediction device 1 may not include the display terminal 13, but only include the processor 12 with a display function and the magnetocardiogram examination device 11. The processor 12 with a display function may include a tablet computer, a notebook computer, a palm computer, a mobile phone, a personal computer (abbreviated as PC), and are not limited here.
[0046] Next, the technical solutions in the embodiments of the present application will be described in detail with reference to the accompanying drawings in the embodiments of the present application.
[0047] The following embodiments of the present application provide a magnetocardiogram data processing device 100, and this method can be implemented, for example, through Figure 1 the processor 12 shown. Figure 2 Shown is a schematic structural diagram of the magnetocardiogram data processing device described in the embodiments of the present application. As Figure 2 shown, the magnetocardiogram data processing device 100 includes a data acquisition module 110 and a data processing module 120.
[0048] The data acquisition module 110 is used to acquire magnetocardiogram data. Optionally, a magnetocardiogram scanner is used to acquire the magnetocardiogram data.
[0049] The data processing module 120 is used to obtain the processing result of the magnetocardiogram data by using a deep learning model; wherein, the training data of the deep learning model includes magnetocardiogram signals, as well as the stenosis degree index of the cardiac blood vessels and the ischemia degree index of the cardiac position area, and the processing result includes the stenosis degree prediction result of the cardiac blood vessels and the ischemia degree prediction result of the cardiac position area.
[0050] Optionally, coronary angiography is used to obtain the stenosis degree index of the blood vessels, and single photon emission computed tomography is used to obtain the ischemia degree index of the cardiac position area.
[0051] In the embodiment of the present application, a deep learning model is trained by using magnetocardiogram signals, as well as the stenosis degree index of the cardiac blood vessels and the ischemia degree index of the cardiac position area, so as to process the magnetocardiogram data by using the trained deep learning model to obtain the stenosis degree prediction result of the cardiac blood vessels and the ischemia degree prediction result of the cardiac position area. Such a magnetocardiogram data processing device can simultaneously obtain the stenosis degree of the cardiac blood vessels and the ischemia degree prediction result of the cardiac position area according to a single magnetocardiogram data, filling the gap in the clinic that there is no method to simultaneously obtain the stenosis degree of the cardiac blood vessels and the ischemia degree of the cardiac position area according to one examination, and this method is completed based on non-invasive magnetocardiogram examination data.
[0052] Figure 3 Shown is a schematic structural diagram of the deep learning model described in the embodiment of the present application. As Figure 3 shown, the deep learning model includes a three-dimensional convolution module, a basic module, a connection layer, a graph convolution module (GCN module), a ViT module, and a prediction head.
[0053] In some possible implementation manners, the QRS band of the magnetocardiogram data is input into the three-dimensional convolution module, the output of the three-dimensional convolution module is input into the basic module, and the output of the basic module is input into the connection layer. The T band of the magnetocardiogram data is input into the three-dimensional convolution module, the output of the three-dimensional convolution module is input into the basic module, and the output of the basic module is input into the connection layer. The outputs of the connection layer module are merged and then input into eight parallel graph convolution modules, ViT modules, and prediction heads to obtain five prediction results on the degree of ischemia in the cardiac position area and three prediction results on the degree of stenosis of the cardiac blood vessels. The three-dimensional convolution module includes a three-dimensional convolutional layer, a batch normalization layer, and a ReLU activation layer. The basic module includes three three-dimensional convolution modules, where the output of the first three-dimensional convolution module is used as the input of the second three-dimensional convolution module, and the output of the third three-dimensional convolution module and the input of the first three-dimensional convolution module together serve as the output of the basic module. The prediction head includes a fully connected layer, a ReLU activation layer, and a sigmoid activation layer, and the input of the prediction head passes through a fully connected layer, a ReLU activation layer, a fully connected layer, a ReLU activation layer, a fully connected layer, and a sigmoid activation layer.
[0054] In an embodiment of the present application, the input of the graph convolution module includes the adjacency matrices of the cardiac blood vessels and the cardiac position area.
[0055] Figure 4 Shown is a schematic structural diagram of the connection layer module according to an embodiment of the present application. The connection layer includes a first convolution unit, a second convolution unit, and a third convolution unit, and the convolution kernel sizes of each convolution unit are different.
[0056] Optionally, the output and input of two three-dimensional convolution modules together serve as the output of the first convolution unit, and the outputs of the first convolution unit, the second convolution unit, and the third convolution unit serve as the output of the connection layer module.
[0057] In some possible implementation manners, the convolution kernel size of the first convolution unit is 3×3, the convolution kernel size of the second convolution unit is 5×5, and the convolution kernel size of the third convolution unit is 7×7. The present application is not limited thereto.
[0058] Figure 5 Shown is a schematic structural diagram of the magnetocardiogram data processing device according to an embodiment of the present application. The magnetocardiogram data processing device 100 further includes a model training module 130. The module for training the deep learning model by the model training module 130 includes a first index acquisition unit 131, a second index acquisition unit 132, an adjacency matrix acquisition unit 133, and a model training unit 134.
[0059] The first index acquisition unit 131 is configured to acquire an ischemia degree index of the heart position area.
[0060] The second index acquisition unit 132 is configured to acquire a stenosis degree index of the heart blood vessels.
[0061] The adjacency matrix acquisition unit 133 is configured to acquire the adjacency matrix, which is obtained by using a bull's-eye diagram model of the heart, and the bull's-eye diagram model is used to represent the relative relationship between the heart blood vessels and the heart.
[0062] The model training unit 134 is configured to train the deep learning model by using the magnetocardiogram signal, the stenosis degree index of the heart blood vessels, and the ischemia degree index of the heart position area.
[0063] Figure 6A Shown is a schematic structural diagram of the bull's-eye diagram model described in the embodiments of the present application. As Figure 6A shown, each of the heart position areas and heart blood vessels corresponds to at least one segment of the bull's-eye diagram model. The heart positions include Apex, Anterior, Septal, Inferior, and Lateral. The heart blood vessels include the left anterior descending branch (LAD), the left circumflex branch (LCX), and the right coronary artery (RCA).
[0064] The apex corresponds to the 17th segment in the bull's-eye diagram model. The anterior wall corresponds to the 1st, 7th, and 13th segments in the bull's-eye diagram model. The septal wall corresponds to the 2nd, 3rd, 8th, 9th, and 14th segments in the bull's-eye diagram model. The inferior wall corresponds to the 4th, 10th, and 15th segments in the bull's-eye diagram model. The lateral wall corresponds to the 5th, 6th, 11th, 12th, and 16th segments in the bull's-eye diagram model.
[0065] The left anterior descending branch corresponds to the 1st, 2nd, 7th, 8th, 13th, 14th, and 17th segments in the bull's-eye diagram model. The left circumflex branch corresponds to the 3rd, 4th, 9th, 10th, and 15th segments in the bull's-eye diagram model. The right coronary artery corresponds to the 5th, 6th, 11th, 12th, and 16th segments in the bull's-eye diagram model.
[0066] Figure 6B Shown is a schematic diagram of the adjacency matrix described in the embodiments of the present application. As Figure 6B shown, the result indexes of the heart positions and blood vessels corresponding to different segments are as shown in the figure. For example, if the apex corresponds to the 17th segment of the bull's-eye diagram model, the value corresponding to the 17th segment of the apex is 1.
[0067] Figure 6C Shown is a schematic diagram of the adjacency matrix described in the embodiments of the present application. For example, Apex is in Figure 6BIf the 17th segment in it is 1 and the LAD in the column where the 17th segment is located is also 1, then Figure 6C the LAD in the row and column where Apex is located in it is 1. For another example, Inferior is in Figure 6B If the 4th, 10th, and 15th segments in it are 1, the RCA in the column where the 4th segment is located is 1, the RCA in the column where the 10th segment is located is 1, and the RCA in the column where the 15th segment is located is 1, then Figure 6C the RCA in the row and column where Inferior is located in it is 1.
[0068] In some possible implementation manners, the adjacency matrix is directly stored as a tensor, and the tensor is a self-attention tensor. The GCN module calculates by using the self-attention tensor and the features after the current input sample is processed. For example, the dimension of the input sample x after being processed by the previous network layer is [b, c, n], the dimension of the adjacency matrix is [1, 1, 8, 8], and after being processed, it becomes [1, k, m, m]. The input feature [b, c, n] is transformed into the dimension [b, c, m, g], where g * m = n. Then, the matrix multiplication operation [1, k, m, m] × [b, c, m, g] is performed on the two, and the output result of the GCN module can be obtained.
[0069] It should be noted that the above is only a possible implementation manner of the embodiments of the present application, and the present application is not limited thereto.
[0070] In an embodiment of the present application, the prediction result of the stenosis degree of the cardiovascular includes the prediction result of the stenosis degree of the left anterior descending artery, the prediction result of the stenosis degree of the left circumflex artery, and the prediction result of the stenosis degree of the right coronary artery.
[0071] The prediction result of the ischemia degree of the heart position area includes the prediction result of the ischemia degree of the apex position area, the prediction result of the ischemia degree of the anterior wall position area, the prediction result of the ischemia degree of the septal wall position area, the prediction result of the ischemia degree of the inferior wall position area, and the prediction result of the ischemia degree of the lateral wall position area.
[0072] Figure 7 Shown is a schematic flowchart of the magnetocardiogram data processing method according to the embodiments of the present application. As Figure 7 shown, the magnetocardiogram data processing method includes the following steps S11 to S12.
[0073] Step S11, obtain magnetocardiogram data.
[0074] Step S12: Obtain the processing result of the magnetocardiogram data by using a deep learning model. Among them, the training data of the deep learning model includes magnetocardiogram data, as well as indicators of the stenosis degree of the cardiac blood vessels and the ischemia degree of the cardiac position area. The processing result includes the prediction result of the stenosis degree of the cardiac blood vessels and the prediction result of the ischemia degree of the cardiac position area.
[0075] It should be noted that steps S11 to S12 in the magnetocardiogram data processing method correspond one by one to the above-mentioned modules 110 to 120 included in the magnetocardiogram data processing device 100, and will not be elaborated here. Figure 2 The magnetocardiogram data processing device 100 includes the above-mentioned modules 110 to 120 in one-to-one correspondence, and will not be elaborated here.
[0076] In several embodiments provided in the present application, it should be understood that the disclosed system, device or method can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of modules / units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, and the indirect coupling or communication connection of devices or modules or units may be in electrical, mechanical or other forms.
[0077] The modules / units described as separate components may or may not be physically separated, and the components shown as modules / units may or may not be physical modules, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules / units can be selected according to actual needs to achieve the purpose of the embodiments of the present application. For example, in each embodiment of the present application, the various functional modules / units can be integrated in a processing module, or each module / unit can exist physically alone, or two or more modules / units can be integrated in one module / unit.
[0078] Those of ordinary skill in the art should also be further aware that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but this implementation should not be considered to exceed the scope of the present application.
[0079] The embodiments of the present application also provide an electronic device. Figure 8It shows a schematic structural diagram of the electronic device 800 described in the embodiments of the present application. As Figure 8 shown, in this embodiment, the electronic device 800 includes a memory 810 and a processor 820.
[0080] The memory 810 is used to store computer programs; preferably, the memory 810 includes: various media such as ROM, RAM, magnetic disks, USB flash drives, memory cards, or optical discs that can store program codes.
[0081] Specifically, the memory 810 may include a computer system readable medium in the form of a volatile memory, such as a random access memory (RAM) and / or a cache memory. The electronic device 800 may further include other removable / non-removable, volatile / non-volatile computer system storage media. The memory 810 may include at least one program product, and this program product has a set of (for example, at least one) program modules, and these program modules are configured to execute the functions of the embodiments of the present application.
[0082] The processor 820 is connected to the memory 810 and is used to execute the computer program stored in the memory 810, so that the electronic device 800 executes the magnetocardiogram data processing method described in any embodiment of the present application.
[0083] Optionally, the processor 820 may be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), etc.; it may also be a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), a field programmable gate array (FPGA for short), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
[0084] Optionally, in this embodiment, the electronic device 800 may further include a display 830. The display 830 is communicatively connected to the memory 810 and the processor 820, and is used to display a relevant graphical user interface (GUI for short) interaction interface of the magnetocardiogram data processing method described in the embodiments of the present application.
[0085] The embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the magnetocardiogram data processing method described in any embodiment of the present application.
[0086] The descriptions of the processes or structures corresponding to the above-mentioned respective drawings each have their own focuses. For the parts not detailed in a certain process or structure, reference may be made to the relevant descriptions of other processes or structures.
[0087] The above embodiments are only illustrative of the principles and effects of the present application and are not intended to limit the present application. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by those with ordinary knowledge in the technical field without departing from the spirit and technical ideas disclosed in the present application should still be covered by the claims of the present application.
Claims
1. A magnetocardiogram data processing device, characterized in that, Comprising: A data acquisition module, configured to acquire magnetocardiogram data; A data processing module, configured to obtain a processing result of the magnetocardiogram data by using a deep learning model; Wherein, the training data of the deep learning model includes magnetocardiogram data, as well as an index of the stenosis degree of the cardiac blood vessels and an index of the ischemia degree of the cardiac position area, and the processing result includes a prediction result of the stenosis degree of the cardiac blood vessels and a prediction result of the ischemia degree of the cardiac position area.
2. The magnetocardiogram data processing device according to claim 1, characterized in that, The deep learning model includes a three-dimensional convolution module, a connection layer, a graph convolution module, a ViT module, and a prediction head.
3. The magnetocardiogram data processing device according to claim 2, wherein The input of the graph convolution module includes the adjacency matrices of the cardiac blood vessels and the cardiac position area.
4. The magnetocardiogram data processing device according to claim 2, characterized in that, The connection layer includes a first convolution unit, a second convolution unit, and a third convolution unit, and the convolution kernel sizes of the convolution units are different.
5. The magnetocardiogram data processing device according to claim 1, wherein It further includes a model training module, and the model training module is configured to train the deep learning model, including: A first index acquisition unit, configured to acquire the index of the ischemia degree of the cardiac position area; A second index acquisition unit, configured to acquire the index of the stenosis degree of the cardiac blood vessels; An adjacency matrix acquisition unit, configured to acquire the adjacency matrix, which is obtained by using a bull's-eye diagram model of the heart, and the bull's-eye diagram model is used to represent the relative relationship between the cardiac blood vessels and the heart; A model training unit, configured to train the deep learning model by using the magnetocardiogram signal, the index of the stenosis degree of the cardiac blood vessels, and the index of the ischemia degree of the cardiac position area.
6. The magnetocardiogram data processing device according to claim 5, wherein Each of the cardiac position areas and the cardiac blood vessels corresponds to at least one segment of the bull's-eye diagram model of the heart.
7. The magnetocardiogram data processing device according to claim 1, characterized in that The prediction result of the stenosis degree of the cardiac blood vessels includes: a prediction result of the stenosis degree of the left anterior descending artery, a prediction result of the stenosis degree of the left circumflex artery, and a prediction result of the stenosis degree of the right coronary artery; The prediction result of the ischemia degree of the cardiac position area includes: a prediction result of the ischemia degree of the apical position area, a prediction result of the ischemia degree of the anterior wall position area, a prediction result of the ischemia degree of the septal wall position area, a prediction result of the ischemia degree of the inferior wall position area, and a prediction result of the ischemia degree of the lateral wall position area.
8. A method for processing magnetocardiogram data, characterized in that, Comprising: Acquire magnetocardiogram data; Obtain a processing result of the magnetocardiogram data by using a deep learning model; Wherein, the training data of the deep learning model includes magnetocardiogram signals, as well as an index of the stenosis degree of the cardiac blood vessels and an index of the ischemia degree of the cardiac position area, and the processing result includes a prediction result of the stenosis degree of the cardiac blood vessels and a prediction result of the ischemia degree of the cardiac position area.
9. An electronic device, characterized in that, The electronic device includes: A memory, configured to store a computer program; A processor, and the processor is configured to execute the computer program stored in the memory, so that the electronic device executes the magnetocardiogram data processing method as claimed in claim 8.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the magnetocardiogram data processing method as claimed in claim 8.