Myocardial infarction diagnosis method, program, and device using electrocardiogram

Abstract

According to an embodiment of the present disclosure, there is disclosed a method, program, and apparatus for diagnosing myocardial infarction using an electrocardiogram, which is performed by a computing device, and includes generating training data based on electrocardiogram data of a patient who has undergone coronary angiography; and training a deep learning model using the generated training data to predict the onset or progression of myocardial infarction in the patient.

Description

[Technical Field]

[0001] The present disclosure relates to artificial intelligence technology in the medical field, and more particularly to a method for training a model for diagnosing myocardial infarction using an electrocardiogram and utilizing the trained model. [Background technology]

[0002] An electrocardiogram (ECG) is a test that records the electrical activity of the heart. An ECG is a relatively simple and cost-effective test that can confirm the health of the heart, which plays an important role in the early diagnosis and management of heart disease. For example, ECG signals measured through an ECG can be used to determine whether each part of the heart is functioning normally, whether the size and position of the heart are normal, and whether there is any damage to the myocardium. Based on these determinations, ECG signals can be used to diagnose various heart-related problems and predict a person's health status.

[0003] Meanwhile, as artificial intelligence technology advances, attempts to use it to analyze electrocardiograms are increasing. In particular, there has been much research on models that use electrocardiograms to diagnose arrhythmias. However, unlike arrhythmias, there has been almost no research on models that diagnose myocardial infarction. Furthermore, myocardial infarction, which occurs when the coronary artery is 80% to 90% blocked, is known to be difficult to detect even by human medical staff using electrocardiograms alone. Therefore, there is a need to apply artificial intelligence technology to the field of myocardial infarction diagnosis using electrocardiograms. Summary of the Invention [Problem to be solved by the invention]

[0004] The present disclosure aims to provide a method for training a model for diagnosing myocardial infarction using an electrocardiogram and diagnosing myocardial infarction using the trained model.

[0005] However, the problems to be solved by the present disclosure are not limited to those mentioned above, and other problems not mentioned will be clearly understood based on the following description. [Means for solving the problem]

[0006] To achieve the above object, an embodiment of the present disclosure provides a method for diagnosing myocardial infarction using an electrocardiogram, the method including the steps of generating training data based on electrocardiogram data of a patient who has undergone coronary angiography, and training a deep learning model using the generated training data to predict the onset or progression of myocardial infarction in the patient.

[0007] Alternatively, the step of generating training data based on electrocardiogram data of a patient who has undergone coronary angiography may include the step of generating training data by labeling the degree of coronary artery stenosis in the electrocardiogram data of the patient who has undergone coronary angiography.

[0008] Alternatively, the labeling can be performed by distinguishing between the types of coronary arteries.

[0009] Alternatively, the step of training a deep learning model that predicts the onset or progression of myocardial infarction using the generated training data may include a step of classifying the training data to generate subgroups based on at least one of a first criterion indicating demographics, a second criterion indicating factors that influence the onset or progression of myocardial infarction, a third criterion indicating the type of disease and diseased subject, or a fourth criterion indicating the type of electrocardiogram; and a step of training the deep learning model to predict the onset or progression of myocardial infarction for each of the generated subgroups.

[0010] Alternatively, the step of training a deep learning model that predicts the onset or progression of myocardial infarction using the generated training data may include a step of selecting a group whose performance is below a reference value based on validation results derived during the training process of the generated subgroups; and a step of performing additional learning on the deep learning model on which the training has been performed based on the selected group.

[0011] Alternatively, the additional learning may be performed based on few-shot learning.

[0012] Alternatively, the second group may include at least one of a risk factor for disease onset, a past history, whether the patient is an early-onset patient, or a type of chest pain.

[0013] Alternatively, the third group may include at least one of a type of myocardial infarction, a type of acute coronary syndrome, or a type of coronary artery.

[0014] To achieve the above-described object, one embodiment of the present disclosure provides a method for diagnosing myocardial infarction using an electrocardiogram, the method being performed by a computing device. The method includes the steps of acquiring electrocardiogram data; and inputting the electrocardiogram data into a pre-trained deep learning model to predict the onset or progression of myocardial infarction in a subject whose electrocardiogram data was measured. In this case, the deep learning model may be pre-trained to predict the onset or progression of myocardial infarction in the subject using training data generated based on the electrocardiogram data of the patient who underwent coronary angiography.

[0015] To achieve the above object, one embodiment of the present disclosure provides a computer program stored on a computer-readable storage medium. When executed by one or more processors, the computer program performs operations for diagnosing myocardial infarction using electrocardiograms. The operations include generating training data based on electrocardiogram data of a patient who has undergone coronary angiography; and training a deep learning model that predicts the onset or progression of myocardial infarction in the patient using the generated training data.

[0016] To achieve the above-described object, one embodiment of the present disclosure provides a computing device for diagnosing myocardial infarction using an electrocardiogram. The device may include a processor including at least one core; a memory including program code executable by the processor; and a network unit for the processor. In this case, the processor generates training data based on electrocardiogram data of a patient who has undergone coronary angiography, and uses the generated training data to train a deep learning model that predicts the onset or progression of myocardial infarction in the patient. [Effects of the Invention]

[0017] According to the method of the present disclosure, it is possible to analyze not only the presence or absence of myocardial infarction disease but also how the myocardial infarction is progressing through an artificial intelligence model.

[0018] In addition, through few-shot learning of the aforementioned artificial intelligence model, it is possible to compensate for the shortcomings of the model that show insufficient performance in a specific patient population. [Brief explanation of the drawings]

[0019] [Figure 1] FIG. 1 is a block diagram of a computing device according to one embodiment of the present disclosure. [Figure 2]FIG. 1 is a block diagram illustrating a process for training a deep learning model according to one embodiment of the present disclosure. [Figure 3] FIG. 10 is a block diagram illustrating a process for training a deep learning model according to an additional embodiment of the present disclosure. [Figure 4] 1 is a flowchart illustrating a method for training a deep learning model according to an embodiment of the present disclosure. [Figure 5] 1 is a flowchart illustrating an inference method for a deep learning model according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0020] Hereinafter, with reference to the accompanying drawings, embodiments of the present disclosure will be described in detail so that those skilled in the art can easily implement the present disclosure. The embodiments presented in this disclosure are provided to enable those skilled in the art to use or practice the contents of the present disclosure. Therefore, various modifications to the embodiments of the present disclosure will be apparent to those skilled in the art. That is, the present disclosure may be embodied in various different forms and is not limited to the following embodiments.

[0021] Throughout the specification of the present disclosure, the same or similar reference numerals refer to the same or similar components. In addition, in order to clearly explain the present disclosure, reference numerals of parts in the drawings that are not related to the explanation of the present disclosure may be omitted.

[0022] The term "or" as used in this disclosure is intended to mean an inclusive "or" rather than an exclusive "or." That is, unless otherwise specified in this disclosure or the context makes the meaning clear, "X utilizes A or B" should be understood to mean one of the natural inclusive permutations. For example, unless otherwise specified in this disclosure or the context makes the meaning clear, "X utilizes A or B" can be interpreted as either X utilizes A, X utilizes B, or X utilizes both A and B.

[0023] The term "and / or" as used in this disclosure should be understood to refer to and include all possible combinations of one or more of the associated listed concepts.

[0024] The terms "comprises" and / or "comprises" as used in this disclosure should be understood to mean that the specified feature and / or component is present. However, the terms "comprises" and / or "comprises" should be understood not to exclude the presence or addition of one or more other features, other components and / or combinations thereof.

[0025] Unless otherwise specified in this disclosure or clear from the context as indicating the singular form, the singular should generally be construed as including "one or more."

[0026] The term "Nth (N is a natural number)" used in the present disclosure may be understood as an expression used to distinguish components of the present disclosure from one another based on a predetermined criterion, such as functional, structural, or convenience of description. For example, components performing different functional roles in the present disclosure may be classified as a first component or a second component. However, components that are substantially the same within the technical concept of the present disclosure but must be distinguished for convenience of description may also be classified as a first component or a second component.

[0027] The term "acquire" as used in this disclosure may be understood to mean generating data in an on-device manner as well as receiving data over a wired or wireless communication network with an external device or system.

[0028] Meanwhile, the terms "module" or "unit" used in this disclosure may refer to an independent functional unit that processes computing resources, such as a computer-related entity, firmware, software or a portion thereof, hardware or a portion thereof, or a combination of software and hardware. In this regard, a "module" or "unit" may refer to a unit composed of a single element or a unit expressed as a combination or set of multiple elements. For example, in a narrow sense, a "module" or "unit" may refer to a hardware element or a set of elements of a computing device, an application program that performs a specific software function, a processing procedure implemented through the execution of software, or a set of instructions for program execution. In a broad sense, a "module" or "unit" may refer to a computing device itself that constitutes a system, or an application executed on a computing device. However, the above concepts are merely examples, and the concepts of "module" and "unit" may be defined in various ways within the scope of those skilled in the art based on the contents of this disclosure.

[0029] The term "model" as used in this disclosure may be understood as a system implemented using mathematical concepts and language to solve a specific problem, a set of software units for solving a specific problem, or an abstract model of a processing process for solving a specific problem. For example, a neural network "model" may refer to a system implemented as a neural network that has problem-solving capabilities through learning. In this case, a neural network can acquire problem-solving capabilities by optimizing parameters connecting nodes or neurons through learning. A neural network "model" may include a single neural network or a neural network ensemble in which multiple neural networks are combined.

[0030] The explanations of the above terms are intended to aid in understanding the present disclosure. Therefore, unless the above terms are explicitly stated as matters that limit the contents of the present disclosure, please note that they are not used to limit the technical ideas of the contents of the present disclosure.

[0031] FIG. 1 is a block diagram of a computing device according to an embodiment of the present disclosure.

[0032] The computing device 100 according to an embodiment of the present disclosure may be a hardware device or part of a hardware device that performs comprehensive data processing and calculations, or may be a software-based computing environment connected via a communication network. For example, the computing device 100 may be a server that performs intensive data processing functions and shares resources, or a client that shares resources through interaction with the server. The computing device 100 may also be a cloud system in which multiple servers and clients interact to process data comprehensively. The above description is merely an example of a type of computing device 100, and various types of computing devices 100 may be configured within a scope that would be understandable to one skilled in the art based on the contents of this disclosure.

[0033] 1, a computing device 100 according to an embodiment of the present disclosure may include a processor 110, a memory 120, and a network unit 130. However, since FIG. 1 is merely an example, the computing device 100 may include other components for implementing a computing environment. Also, only some of the disclosed components may be included in the computing device 100.

[0034] The processor 110 according to an embodiment of the present disclosure may be understood as a component including hardware and / or software for performing computing operations. For example, the processor 110 may read a computer program to perform data processing for machine learning. The processor 110 may process operations such as input data processing for machine learning, feature extraction for machine learning, and error calculation based on backpropagation. The processor 110 for performing such data processing may include a central processing unit (CPU), a general-purpose graphics processing unit (GPGPU), a tensor processing unit (TPU), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), etc. The types of processor 110 described above are merely examples, and various types of processor 110 may be configured within the scope of what would be understood by one skilled in the art based on the present disclosure.

[0035] The processor 110 may generate training data for a deep learning model for diagnosing myocardial infarction based on electrocardiogram data of a patient who has undergone coronary angiography. The processor 110 may generate training data by labeling the degree of coronary artery stenosis in the electrocardiogram data of the patient who has undergone coronary angiography. In this case, the degree of coronary artery stenosis is information confirmed by performing coronary angiography and may be acquired together with the electrocardiogram data. That is, the computing device 100 may input the electrocardiogram data of the patient who has undergone coronary angiography and generate training data by configuring the degree of coronary artery stenosis of the patient as a label.

[0036] The processor 110 can train a deep learning model for diagnosing myocardial infarction using the generated training data. The processor 110 can train the deep learning model so that the deep learning model predicts the onset or progression of myocardial infarction in a patient based on the training data. For example, the processor 110 can input electrocardiogram data included in the training data into the deep learning model and output the degree of coronary artery stenosis for predicting the onset or progression of myocardial infarction in a patient. In this case, the output may be in the form of a score, which is a numerical value representing the degree of coronary artery stenosis. The processor 110 can compare the output of the deep learning model with the label and adjust the parameters of the neural network included in the deep learning model. The processor 110 can train the deep learning model by repeatedly performing the above process until the error between the output of the deep learning model and the label satisfies a minimum criterion.

[0037] The processor 110 can predict the onset or progression of myocardial infarction using the trained deep learning model. The processor 110 can input electrocardiogram data into the trained deep learning model and output the degree of coronary artery stenosis for predicting the onset or progression of myocardial infarction. In this case, the electrocardiogram data input into the deep learning model is not limited to data of patients who have undergone coronary angiography, but can include all electrocardiogram data measured for various purposes in a medical environment or in real life. The processor 110 can predict whether myocardial infarction has occurred or, if myocardial infarction has occurred, how far it has progressed based on the degree of coronary artery stenosis output by the deep learning model.

[0038] The memory 120 according to an embodiment of the present disclosure may be understood as a component including hardware and / or software for storing and managing data processed by the computing device 100. That is, the memory 120 may store any type of data generated or determined by the processor 110 and any type of data received by the network unit 130. For example, the memory 120 may include at least one type of storage medium selected from the group consisting of flash memory, hard disk, multimedia card micro, card-type memory, random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, and optical disk. The memory 120 may also include a database system for managing data in a predetermined manner. The types of memory 120 described above are merely examples, and the types of memory 120 may be configured in various ways within the scope of what would be understood by a person skilled in the art based on the contents of this disclosure.

[0039] The memory 120 may structure, organize, and manage data, combinations of data, and program code executable by the processor 110 required for the processor 110 to perform calculations. For example, the memory 120 may store medical data received through the network unit 130 (described below). The memory 120 may store program code for operating a neural network model to receive input medical data and perform learning, program code for operating a neural network model to receive input medical data and perform inference according to the intended use of the computing device 100, and processed data generated by executing the program code.

[0040] The network unit 130 according to an embodiment of the present disclosure may be understood as a component that transmits and receives data through any type of publicly-known wired / wireless communication system. For example, the network unit 130 may transmit and receive data using a wired / wireless communication system such as a local area network (LAN), wideband code division multiple access (WCDMA), long term evolution (LTE), wireless broadband internet (WiBro), 5th generation mobile communication (5G), ultra wideband, ZigBee, radio frequency (RF) communication, wireless LAN, wireless fidelity (WiFi), near field communication (NFC), or Bluetooth. The above-mentioned communication systems are merely examples, and various wired / wireless communication systems for transmitting and receiving data by the network unit 130 may be applied other than the above-mentioned examples.

[0041] The network unit 130 may receive data necessary for the processor 110 to perform calculations via wired / wireless communication with any system or any client. The network unit 130 may also transmit data generated through calculations by the processor 110 via wired / wireless communication with any system or any client. For example, the network unit 130 may receive medical data through communication with a database in a hospital environment, a cloud server that performs tasks such as medical data standardization, a client such as a smart watch, or a medical computing device. The network unit 130 may transmit output data of the neural network model, and intermediate data and processed data derived during the calculation process of the processor 110 via communication with the aforementioned database, server, client, or computing device.

[0042] Figure 2 is a block diagram illustrating a process for training a deep learning model according to one embodiment of the present disclosure. Figure 3 is a block diagram illustrating a process for training a deep learning model according to an additional embodiment of the present disclosure.

[0043] 2, a computing device 100 according to an embodiment of the present disclosure may generate learning data 20 by labeling the degree of stenosis of the coronary arteries of a patient who has undergone coronary angiography to electrocardiogram data 10 of the patient. At this time, the computing device 100 may label the degree of stenosis of the coronary arteries by distinguishing between different types of coronary arteries. For example, the computing device 100 may label the degree of stenosis of each type of coronary artery to the electrocardiogram data 10 based on medical classifications of the coronary arteries, such as right coronary artery (RCA), left anterior descending artery (LAD), left circumflex artery (LCx), etc. Through this labeling, it may be possible to analyze which blood vessels have narrowed to what extent during the learning process of a deep learning model.

[0044] The computing device 100 can input the training data 20 into the deep learning model 200 and train the deep learning model 200 to output the degree of coronary artery stenosis for diagnosing myocardial infarction. In this case, the deep learning model 200 can include a neural network that extracts features related to an electrocardiogram waveform, such as a P wave, a QRS complex, and a T wave, from an electrocardiogram, and a neural network that outputs a numerical value representing the degree of coronary artery stenosis based on the extracted features. For example, the computing device 100 can input electrocardiogram data included in the training data 20 into the deep learning model 200 and output a numerical value representing the degree of coronary artery stenosis. The computing device 100 can calculate an error using a loss function with the numerical value output by the deep learning model 200 and the labels included in the training data 20 as input variables. The computing device 100 can adjust parameters of the neural network included in the deep learning model 200 based on the calculated error. The computing device 100 can then repeatedly adjust the parameters of the neural network until the error satisfies a minimum criterion. The learning of the deep learning model 200 can be performed based on supervised learning as in the above example, but may also be performed based on self-supervised learning due to the neural network structure of the deep learning model 200.

[0045] On the other hand, if the deep learning model 200 is unable to learn many electrocardiograms of patients with myocardial infarction, the output accuracy may be low because learning on the electrocardiogram data of patients with myocardial infarction was not sufficiently performed. Generally, the electrocardiogram data of patients with diseases is smaller in number than the electrocardiogram data of people in a normal, healthy state. Therefore, there is a possibility that a sufficient amount of data for learning the deep learning model 200 for diagnosing diseases such as myocardial infarction may not be secured. Therefore, to improve the performance of the deep learning model 200, the computing device 100 may perform additional learning on data groups with low output accuracy based on verification evaluation of specific data groups identified during the learning process.

[0046] Referring to FIG. 3, the computing device 100 can classify the training data 20 into multiple subgroups based on medically used criteria for diagnosing myocardial infarction. For example, the computing device 100 can classify the training data 20 into a first subgroup 31, a second subgroup 33, a third subgroup 35, ..., an Nth subgroup 39, each corresponding to a variety of criteria that influence the diagnosis of myocardial infarction. Specifically, the classification criteria corresponding to each of the subgroups 31, 33, 35, and 39 can include a first criterion indicating demographics, a second criterion indicating factors that influence the onset or progression of myocardial infarction, a third criterion indicating the type of disease and diseased subject, and a fourth criterion indicating the type of electrocardiogram. The second criterion can include at least one of risk factors for disease onset, past history, whether the patient is an early-onset patient, or the type of chest pain. The third criterion can include at least one of the type of myocardial infarction, the type of acute coronary syndrome, or the type of coronary artery. The computing device 100 may classify data included in the training data 20 into one of the first to fourth criteria to generate subgroups 31, 33, 35, and 39.

[0047] [Table 1]

[0048] Specific items of each criterion can be organized as shown in Table 1. However, since Table 1 is merely an example, the criteria of the present disclosure are not limited to Table 1.

[0049] The computing device 100 may perform a first learning process in which the deep learning model 200 predicts the onset or progression of myocardial infarction for each subgroup. The computing device 100 may perform a first learning process in which each subgroup is input into the deep learning model 200 and the deep learning model 200 outputs the degree of coronary artery stenosis for predicting the onset or progression of myocardial infarction. In this case, the first learning process may correspond to the learning process of FIG. 2 described above.

[0050] The computing device 100 may derive a verification result for each subgroup during the first learning process. The computing device 100 may then select a subgroup with low performance based on the verification result. For example, the computing device 100 may select a subgroup with performance below a reference value based on the verification result value during the first learning process. The computing device 100 may select a subgroup with performance below a reference value based on a performance index used in hospitals or emergency rooms.

[0051] The computing device 100 may perform second learning on the deep learning model 200 that has completed the first learning based on the selected group. The computing device 100 may input a group whose performance is below a reference value into the deep learning model 200 that has completed the first learning and perform additional learning based on short-shot learning to improve the performance of the group. Here, short-shot learning is a learning method that performs inference based on a small amount of data and may be understood as a method of learning to which class of support data query data belongs by learning similarities between small amounts of data. That is, since the group whose performance is below a reference value is a small amount of data, the computing device 100 may input the group whose performance is below the reference value, which is a small amount of data, into the deep learning model 200 that has completed the first learning, and perform short-shot learning-based second learning to predict the degree of coronary artery stenosis. Through this second learning, the computing device 100 may supplement the performance of the deep learning model 200 that shows insufficient performance in specific conditions or diagnostic factors.

[0052] FIG. 4 is a flowchart illustrating a method for training a deep learning model according to an embodiment of the present disclosure.

[0053] Referring to FIG. 4, a computing device 100 according to an embodiment of the present disclosure may generate training data based on electrocardiogram data of a patient who has undergone coronary angiography (S110). If the computing device 100 is a server that performs electrocardiogram interpretation, the computing device 100 may acquire the electrocardiogram data of the patient who has undergone coronary angiography through communication with a database in a medical environment that manages the patient's electrocardiogram data. If the computing device 100 is a server that manages electrocardiogram data in a medical environment, once measurement of the patient's electrocardiogram signal is completed, the computing device 100 may generate and manage the electrocardiogram data of the patient. The computing device 100 may generate training data by labeling the degree of stenosis of the patient's coronary artery to the electrocardiogram data of the patient who has undergone coronary angiography. When labeling the degree of stenosis of the coronary artery, the computing device 100 may distinguish and label the type of coronary artery.

[0054] The computing device 100 can train a deep learning model that predicts the onset or progression of myocardial infarction in a patient using the training data generated through step S110 (S120). The computing device 100 can train the deep learning model through first learning based on supervised learning, self-supervised learning, or the like, depending on the structure of the deep learning model. The computing device 100 can then determine data sets with low prediction accuracy of the deep learning model and perform additional training on the deep learning model through second learning based on short-shot learning. For this training, the computing device 100 can use various classifications of the training data based on factors related to myocardial infarction diagnosis. Specifically, the computing device 100 can perform first learning by inputting subgroups generated through classification of the training data into the deep learning model. The computing device 100 can select groups with low learning performance based on the verification results for each subgroup confirmed through the first learning. The computing device 100 can then perform second learning on the deep learning model based on the selected groups.

[0055] FIG. 5 is a flowchart illustrating an inference method for a deep learning model according to an embodiment of the present disclosure.

[0056] 5, a computing device 100 according to an embodiment of the present disclosure may acquire electrocardiogram data (S210). If the computing device 100 is a server that performs electrocardiogram interpretation, the computing device 100 may acquire electrocardiogram data of a person who desires to interpret an electrocardiogram through communication with equipment in a medical environment that measures the patient's electrocardiogram data or a wearable device capable of measuring electrocardiogram data. If the computing device 100 is equipment in a medical environment that measures the patient's electrocardiogram data or a wearable device capable of measuring electrocardiogram data, the computing device 100 may measure an electrocardiogram signal to generate electrocardiogram data.

[0057] The computing device 100 can input electrocardiogram data into a pre-trained deep learning model to predict the onset or progression of myocardial infarction in a subject whose electrocardiogram data was measured (S220). The computing device 100 can input electrocardiogram data into the pre-trained deep learning model to generate an output value indicating the degree of stenosis of the coronary artery. The computing device 100 can analyze the degree of stenosis of the coronary artery output through the deep learning model based on clinical evidence to derive the onset or progression of myocardial infarction. The computing device 100 can generate a report including information on the derived onset or progression of myocardial infarction. In this case, if the computing device 100 is a server that performs electrocardiogram interpretation, the computing device 100 can share the report through communication with a client that can view the electrocardiogram interpretation results. If the computing device 100 has an input / output unit that provides a user interface, the computing device 100 can output the report through the input / output unit.

[0058] The various embodiments of the present disclosure described above may be combined with additional embodiments and may be modified within the scope that would be understandable to one skilled in the art in light of the above detailed description. The embodiments of the present disclosure are illustrative in all respects and should not be considered limiting. For example, components described as a single type may be implemented in a distributed form, and similarly, components described as distributed may be implemented in a combined form. Therefore, all modifications and variations derived from the meaning, scope, and equivalent concepts of the claims of the present disclosure should be construed as being within the scope of the present disclosure.

Claims

1. 1. A method for diagnosing myocardial infarction using an electrocardiogram, the method being performed by a computing device including at least one processor, the method comprising: generating training data based on electrocardiogram data of patients who have undergone coronary angiography; and A method comprising: using the generated training data to train a deep learning model that predicts the onset or progression of myocardial infarction in the patient.

2. The step of generating learning data based on electrocardiogram data of patients who have undergone coronary angiography includes: The method according to claim 1 , further comprising: generating training data by labeling the degree of coronary artery stenosis in electrocardiogram data of a patient who has undergone coronary angiography.

3. The labeling may include: The method according to claim 2, wherein the method is performed by distinguishing between types of coronary arteries.

4. The step of training a deep learning model that predicts the onset or progression of myocardial infarction using the generated training data includes: generating subgroups by classifying the training data based on at least one of a first criterion indicating demographics, a second criterion indicating factors affecting the onset or progression of myocardial infarction, a third criterion indicating the type of disease and subject, and a fourth criterion indicating the type of electrocardiogram; and The method of claim 3 , further comprising: training the deep learning model to predict the onset or progression of myocardial infarction for each of the generated subgroups.

5. The step of training a deep learning model that predicts the onset or progression of myocardial infarction using the generated training data includes: selecting a group whose performance is below a reference value based on a validation result derived during a learning process of the generated subgroups; The method of claim 3 , further comprising: performing additional learning on the deep learning model on which the learning has been performed based on the selected group.

6. The additional learning includes: The method of claim 5 , wherein the method is performed on the basis of few-shot learning.

7. The second criterion is:

5. The method of claim 4, wherein the risk factors for developing the disease include at least one of a past history, whether the patient is an early-onset patient, or a type of chest pain.

8. The third criterion is:

5. The method of claim 4, comprising at least one of a type of myocardial infarction, a type of acute coronary syndrome, or a type of coronary artery.

9. 1. A method for diagnosing myocardial infarction using an electrocardiogram, the method being performed by a computing device including at least one processor, the method comprising: acquiring electrocardiogram data; and inputting the electrocardiogram data into a pre-trained deep learning model to predict the onset or progression of myocardial infarction in the subject whose electrocardiogram data was measured; Including, The deep learning model is A method that is pre-trained to predict the onset or progression of myocardial infarction in a patient using training data generated based on electrocardiogram data of the patient who underwent coronary angiography.

10. A computer program stored in a computer-readable storage medium, the computer program performing an operation for diagnosing myocardial infarction using an electrocardiogram when executed by one or more processors, The operation is generating training data based on electrocardiogram data of patients who have undergone coronary angiography; and A computer program comprising: an operation of using the generated learning data to train a deep learning model that predicts the onset or progression of myocardial infarction in the patient.

11. 1. A computing device for diagnosing myocardial infarction using an electrocardiogram, comprising: a processor including at least one core; a memory containing program code executable by the processor; and a network unit for acquiring electrocardiogram data; Including, The processor: Generate training data based on electrocardiogram data of patients who have undergone coronary angiography, An apparatus that uses the generated learning data to train a deep learning model that predicts the onset or progression of myocardial infarction in the patient.

Images