WO2025198378A1 - Method, apparatus, and program for identifying state of user on basis of abnormal electrocardiogram signal - Google Patents

Abstract

The present disclosure provides a method, an apparatus, and a program for identifying a state of a user on the basis of an abnormal electrocardiogram signal. The method for identifying a state of a user on the basis of an abnormal electrocardiogram signal according to one embodiment of the present disclosure comprises the steps of: obtaining first lead electrocardiogram data for a user on the basis of a first measurement method of a single lead electrocardiogram measurement device; determining whether an electrocardiogram signal corresponding to the obtained first lead electrocardiogram data is abnormal; and suggesting to obtain second lead electrocardiogram data for the user on the basis of a second measurement method of the single lead electrocardiogram measurement device when it is determined that the electrocardiogram signal is abnormal.

Description

Method, device and program for determining a user's condition based on an abnormal electrocardiogram signal

The present disclosure relates to a method, device, and program for determining a user's condition based on an abnormal electrocardiogram (ECG) signal. Specifically, the present disclosure relates to a method for proposing a measurement method for measuring a normal ECG signal by determining whether the measured ECG signal is abnormal using a single-lead ECG measurement device.

Electrocardiogram (ECG) data is a record of electrocardiogram signals generated by microcurrents within the heart. Accurate analysis of ECG data is essential in the medical field, as it can lead to early detection of cardiovascular disease or structural abnormalities in patients.

Recently, there has been growing interest in technologies that analyze a user's electrocardiogram (ECG) data in real time using single-lead ECG devices, such as smartwatches and portable ECG monitors. These single-lead ECG devices offer excellent portability, allowing users to easily monitor their health in their daily lives and contributing to early detection of abnormalities and preventive health management.

However, when measuring a user's ECG using a single-lead ECG measurement device, the accuracy of the analysis results can be reduced if the measured ECG signal contains any abnormalities. This includes, for example, ECG signals containing noise, unclear ECG signals, or low-voltage ECG signals.

In particular, when analyzing ECG signals using AI models, low-voltage ECG signals not only degrade the performance of the AI model, but can also lead to inaccurate analysis results regarding the user's condition. The cause of low-voltage ECG signals is often due to the user's physical factors (e.g., subcutaneous fat, pericardial effusion, dehydration, etc.). If these factors are not identified, there is a high probability that low-voltage ECG data will continue to be acquired, which can also lead to problems such as reducing the reliability of the AI model's analysis results.

The present disclosure has been made in response to the aforementioned background technology, and aims to provide a method, device and computer program for determining a user's condition based on an abnormal electrocardiogram signal.

However, the problems to be solved in this disclosure are not limited to the problems mentioned above, and other problems not mentioned can be clearly understood based on the description below.

In accordance with an embodiment of the present disclosure for realizing the task as described above, a method for determining a user's condition based on an abnormal electrocardiogram signal, performed by a computing device including at least one processor, comprises the steps of: obtaining first-lead electrocardiogram data for the user based on a first measurement method of a single-lead electrocardiogram measuring device; determining whether an electrocardiogram signal corresponding to the obtained first-lead electrocardiogram data is abnormal; and proposing to obtain second-lead electrocardiogram data for the user based on a second measurement method of the single-lead electrocardiogram measuring device if the electrocardiogram signal is determined to be abnormal.

Alternatively, the first measurement method and the second measurement method are set differently depending on the combination of the user's body parts that are each in contact with the plurality of electrodes included in the single-induction electrocardiogram measuring device.

Alternatively, the proposed step includes displaying body part information in contact with the first electrode and the second electrode included in the single-induction electrocardiogram measuring device through a display.

Alternatively, when second-derived electrocardiogram data for the user is acquired based on the second measurement method, the method includes a step of determining whether an electrocardiogram signal corresponding to the acquired second-derived electrocardiogram data is abnormal, and a step of setting the second measurement method as a reference measurement method for determining the status of the user when the electrocardiogram signal corresponding to the acquired second-derived electrocardiogram data is determined not to be abnormal.

Alternatively, the step of determining whether the electrocardiogram signal corresponding to the acquired second-derived electrocardiogram data is abnormal includes the step of acquiring second-derived electrocardiogram data for the user based on the second measurement method, receiving lead type information of the second measurement method through an input interface, and the step of determining whether the electrocardiogram signal corresponding to the acquired second-derived electrocardiogram data for the lead type is abnormal.

Alternatively, the method includes a step of detecting motion of a single-induction electrocardiogram measuring device in the second measurement method through a sensing unit to obtain a sensing value, and a step of determining whether the user obtains electrocardiogram data for determining the status using the set reference measurement method based on the sensing value.

Alternatively, if the electrocardiogram signal is determined to be normal, the step of inputting the acquired first-derived electrocardiogram data into a pre-trained neural network model to obtain a score corresponding to the user's condition, and the step of analyzing the user's condition by comparing the acquired score with a reference value.

Alternatively, the method comprises a step of selecting a pre-trained neural network model corresponding to a first measurement method among a plurality of pre-trained neural network models corresponding to different measurement methods of the single-induction electrocardiogram measuring device.

Alternatively, the step of proposing to acquire second lead electrocardiogram data for the user comprises the step of proposing to disinfect a body part of the user that comes into contact with the first measurement method or to adjust the position of the single lead electrocardiogram measurement device in the first measurement method, if the first measurement method is a lead 1 measurement method and the electrocardiogram signal is determined to be abnormal due to low voltage.

Alternatively, if the electrocardiogram signal corresponding to the acquired second-derived electrocardiogram data is determined to be normal, the step of determining that the electrical axis of the user's heart is outside the normal range is included.

In order to achieve the above-described task, according to one embodiment of the present disclosure, a single-lead electrocardiogram (ECG) measuring device for determining a user's condition based on an abnormal ECG signal comprises a memory including a program code, a sensing unit for acquiring ECG data of a user, a display, and at least one processor for acquiring first-lead ECG data for the user based on a first measurement method of the single-lead ECG measuring device, determining whether an ECG signal corresponding to the acquired first-lead ECG data is abnormal, and proposing, through the display, to acquire second-lead ECG data for the user based on a second measurement method of the single-lead ECG measuring device if the ECG signal is determined to be abnormal.

A computer program stored in a computer-readable storage medium according to one embodiment of the present disclosure for realizing the task as described above, wherein the computer program, when executed on one or more processors, performs an operation for determining a user's condition based on an abnormal electrocardiogram signal, the operation including an operation for obtaining first-lead electrocardiogram data for the user based on a first measurement method of a single-lead electrocardiogram measurement device, an operation for determining whether an electrocardiogram signal corresponding to the obtained first-lead electrocardiogram data is abnormal, and an operation for proposing to obtain second-lead electrocardiogram data for the user based on a second measurement method of the single-lead electrocardiogram measurement device if the electrocardiogram signal is determined to be abnormal.

A method for determining a user's condition based on a low-voltage electrocardiogram signal according to an embodiment of the present disclosure can improve the accuracy of artificial intelligence-based electrocardiogram analysis by inducing the user to acquire electrocardiogram data according to a different measurement method when low-voltage electrocardiogram data is generated.

FIG. 1 is an exemplary diagram of a computing device that determines a user's condition based on a low-voltage electrocardiogram signal according to one embodiment of the present disclosure.

FIG. 2 is a block diagram of a computing device that determines a user's condition based on a low-voltage electrocardiogram signal according to one embodiment of the present disclosure.

FIG. 3 is a flowchart of a method for determining a user's status based on a low-voltage electrocardiogram signal according to an embodiment of the present disclosure.

FIG. 4 is an exemplary diagram illustrating multiple measurement methods using a computing device according to one embodiment of the present disclosure.

FIG. 5 is an exemplary diagram illustrating a method for a computing device according to an embodiment of the present disclosure to propose a second measurement method.

FIG. 6 is a sequence diagram illustrating a method for determining a user's condition based on a low-voltage electrocardiogram signal of a computing device implemented as an artificial intelligence server according to another embodiment of the present disclosure.

FIG. 7 is a detailed configuration diagram of a computing device according to another embodiment of the present disclosure.

Below, embodiments of the present disclosure are described in detail with reference to the attached drawings 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 utilize or implement the contents of the present disclosure. Accordingly, 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 implemented in various different forms and is not limited to the embodiments described below.

Throughout the specification of this disclosure, identical or similar drawing numbers refer to identical or similar components. Furthermore, for the purpose of clearly describing the disclosure, drawing numbers for parts in the drawings that are not relevant to the description of the disclosure may be omitted.

The term "or" as used herein is intended to mean an inclusive "or" rather than an exclusive "or." That is, unless otherwise specified herein or clear from context, "X employs A or B" should be understood to mean either of the natural inclusive permutations. For example, unless otherwise specified herein or clear from context, "X employs A or B" can be interpreted to mean either X employs A, X employs B, or X employs both A and B.

The term "and/or" as used herein should be understood to refer to and include all possible combinations of one or more of the related concepts listed.

The terms "comprises" and/or "comprising" as used herein should be understood to mean the presence of certain features and/or components. However, it should be understood that the terms "comprises" and/or "comprising" do not exclude the presence or addition of one or more other features, other components, and/or combinations thereof.

Unless otherwise specified in this disclosure or unless the context makes it clear that the singular form is intended to be referred to, the singular should generally be construed to include “one or more.”

The term "Nth (N is a natural number)" used in this disclosure can be understood as an expression used to mutually distinguish components of this disclosure based on a predetermined standard such as a functional perspective, a structural perspective, or convenience of explanation. For example, components performing different functional roles in this disclosure can be distinguished as a first component or a second component. However, components that are substantially the same within the technical spirit of this disclosure but must be distinguished for convenience of explanation may also be distinguished as a first component or a second component.

The term "acquisition" as used in this disclosure may be understood to mean not only receiving data through a wired or wireless communication network with an external device or system, but also generating data in an on-device form.

Meanwhile, the term "module" or "unit" used in the present disclosure can be understood as a term referring to an independent functional unit that processes computing resources, such as a computer-related entity, firmware, software or a part thereof, hardware or a part thereof, or a combination of software and hardware. At this time, the "module" or "unit" may be a unit composed of a single element, or a unit expressed as a combination or set of multiple elements. For example, as a narrow concept, a "module" or "unit" may refer to a hardware element of a computing device or a set thereof, an application program that performs a specific function of software, a processing process implemented through software execution, or a set of instructions for program execution, etc. In addition, as a broad concept, a "module" or "unit" may refer to the computing device itself that constitutes the system, or an application running on the computing device, etc. However, since the above-described concept is only an example, the concept of “module” or “part” may be defined in various ways within a range understandable to those skilled in the art based on the contents of the present disclosure.

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

The term "data" used in this disclosure may include "images," signals, and the like. The term "image" used in this disclosure may refer to multidimensional data composed of discrete image elements. In other words, "image" may be understood as a term referring to a digital representation of an object visible to the human eye. For example, "image" may refer to multidimensional data composed of elements corresponding to pixels in a two-dimensional image. "Image" may refer to multidimensional data composed of elements corresponding to voxels in a three-dimensional image.

The explanation of the above terms is intended to aid understanding of the present disclosure. Therefore, unless explicitly stated as limiting the contents of the present disclosure, it should be noted that the above terms are not intended to limit the technical ideas of the contents of the present disclosure.

FIG. 1 is an exemplary diagram of a computing device that determines a user's condition based on a low-voltage electrocardiogram signal according to one embodiment of the present disclosure.

Referring to FIG. 1, a computing device (100) according to an embodiment of the present disclosure may be a hardware device or a part of a hardware device that performs comprehensive processing and calculation of data, or may be a software-based computing environment connected via a communication interface. For example, the computing device (100) may be a server that performs intensive data processing functions and shares resources, or may be a client that shares resources through interaction with a server. In addition, the computing device (100) may be a cloud system in which multiple servers and clients interact to comprehensively process data. Since the above description is only one example related to the type of computing device (100), the type of computing device (100) may be configured in various ways within a category understandable to those skilled in the art based on the contents of the present disclosure. For example, the computing device (100) may be implemented as a server device, a desktop, a laptop, a smartphone, a smart watch, a smart ring, etc. In particular, the computing device (100) may be implemented as a biosignal measuring device that measures a user's electrocardiogram signal. More specifically, the computing device (100) may be a biosignal measuring device that measures a single-lead electrocardiogram signal (hereinafter, referred to as a single-lead electrocardiogram measuring device). The single-lead electrocardiogram measuring device is a device that records the electrical activity of the heart using a single electrode pair. Since it is possible to acquire electrocardiogram data (10) for one lead of the single-lead electrocardiogram measuring device, the user can acquire multiple electrocardiogram data (10) for different leads by varying the measuring method. Here, the measuring method may be set according to the position of the computing device (100) in contact with the user's body and the combination of body parts. However, the present invention is not limited thereto, and according to an embodiment, the computing device (100) may also be implemented as an artificial intelligence server that is linked with the single-lead electrocardiogram measuring device.

Meanwhile, the computing device (100) can determine whether the ECG signal corresponding to the acquired ECG data (10) is abnormal when the computing device (10) obtains the ECG data (10) for a specific induction for the user based on a specific measurement method. For example, the computing device (100) can determine whether the ECG signal is abnormal by determining whether the noise in the ECG signal is higher than a preset value, whether the ECG signal corresponds to a low voltage, or whether the graph corresponding to the ECG signal is unclear. Specifically, the computing device (100) can determine whether the ECG signal is abnormal by determining the degree of noise included in the ECG signal, calculating a score corresponding to the degree of noise, and comparing the score with a preset value. At this time, the computing device (100) can determine that the ECG signal is abnormal if the noise score is higher than a preset value. In addition, the computing device (100) can determine whether the ECG signal is low voltage by comparing the value (e.g., voltage value) of feature information (e.g., QRS wave) included in the ECG signal with a preset voltage value, and can also determine whether the ECG signal is abnormal. At this time, if the value of the feature information is lower than the preset voltage value, the computing device (100) can determine that the ECG signal is abnormal. In addition, the computing device (100) can determine whether preset feature information (e.g., feature point) is identified on a graph corresponding to the ECG signal, and if the preset feature information is not identified, the computing device (100) can determine that the ECG signal is abnormal. Here, the feature information can be feature points such as P waves and Q waves observed on the ECG graph.

However, for the convenience of explanation of the present disclosure, in the following description, it is assumed that the abnormality of the electrocardiogram signal is determined by determining whether the electrocardiogram signal corresponding to the acquired electrocardiogram data (10) is low voltage. Since a low voltage electrocardiogram signal may reduce the accuracy of the analysis result, the computing device (100) may determine whether the electrocardiogram signal corresponding to the electrocardiogram data (10) is low voltage prior to analyzing the electrocardiogram data (10). At this time, if the computing device (100) determines that the electrocardiogram signal corresponding to the electrocardiogram data (10) is low voltage, the computing device (100) may suggest to the user to acquire the electrocardiogram data (10) through another measurement method. Referring to FIG. 1, if the computing device (100) determines that the electrocardiogram signal corresponding to the electrocardiogram data (10) acquired according to the first measurement method is low voltage, the computing device (100) may request acquisition of the electrocardiogram data (10') through the second measurement method. At this time, the computing device (100) may output a voice message guiding the second measurement method through the speaker of the computing device (100) (or a text message through the display of the computing device (100). Meanwhile, the computing device (100) may also determine whether the electrocardiogram signal corresponding to the electrocardiogram data (10') acquired through the second measurement method is low voltage. At this time, if it is determined that the electrocardiogram signal corresponding to the electrocardiogram data (10') acquired according to the second measurement method is low voltage, the computing device (100) may request acquisition of electrocardiogram data for other leads through the third measurement method. The third measurement method may be a measurement method corresponding to a lead type different from the first and second measurement methods. At this time, the computing device (100) may set the first measurement method to Lead I, the second measurement method to Lead II, and the third measurement method to Lead III, and sequentially suggest the measurement methods to the user.

It is determined that the electrocardiogram signal corresponding to the electrocardiogram data (10') acquired according to the second measurement method is normal and not low voltage, and the computing device (100) can analyze the electrocardiogram data (10') acquired according to the second measurement method to determine the user's condition. In particular, the computing device (100) can determine the user's condition by inputting the acquired electrocardiogram data (10') into a pre-trained neural network model to obtain an output value corresponding to the user's condition.

In addition, the computing device (100) may analyze the cause of the low voltage of the electrocardiogram signal corresponding to the electrocardiogram data (10). The cause of the low voltage of the electrocardiogram signal corresponding to the electrocardiogram data (10) may be due to the electric axis of the user's heart being outside the normal range or a muscle abnormality in a body part of the user that the computing device (100) has contacted. Accordingly, if the computing device (100) determines that the electrocardiogram signal corresponding to the acquired electrocardiogram data (10) is low voltage, it may analyze the cause of the low voltage and suggest a different measurement method or remove an interference factor to the user. Hereinafter, embodiments of the present disclosure related to this will be described in detail.

FIG. 2 is a block diagram of a computing device (100) that determines a user's status based on a low-voltage electrocardiogram signal according to one embodiment of the present disclosure.

Referring to FIG. 2, a computing device (100) includes one or more processors (hereinafter, processors) (110), memory (120), a sensing unit (130), and a display (140).

However, since FIG. 1 is only an example, the computing device (100) may include other components for implementing a computing environment. In addition, only some of the disclosed components may be included in the computing device (100). The processor (110) according to one embodiment of the present disclosure may be understood as a configuration unit 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 computational processes such as processing input data 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), or a field programmable gate array (FPGA). The type of processor (110) described above is only one example, and thus the type of processor (110) can be configured in various ways within a range understandable to those skilled in the art based on the contents of the present disclosure.

The processor (110) is connected to other components of the computing device (100) (i.e., the memory (120), the sensing unit (130), and the display (140)) to control the overall operation of the computing device (100). The memory (120) according to one embodiment of the present disclosure may be understood as a configuration unit including hardware and/or software for storing and managing data processed in the computing device (100). That is, the memory (120) may store any form of data generated or determined by the processor (110) and any form of data received by the communication interface of the computing device (100). For example, the memory (120) may include at least one type of storage medium among a flash memory type, a hard disk type, a multimedia card micro type, a card type memory, a RAM (random access memory), an SRAM (static random access memory), a ROM (read-only memory), an EEPROM (electrically erasable programmable read-only memory), a PROM (programmable read-only memory), a magnetic memory, a magnetic disk, and an optical disk. In addition, the memory (120) may include a database system that controls and manages data in a predetermined system. Since the type of the memory (120) described above is only one example, the type of the memory (120) may be configured in various ways within a range understandable to those skilled in the art based on the contents of the present disclosure.

The memory (120) can structure and organize and manage data, combinations of data, and program codes executable by the processor (110) required for the processor (110) to perform operations. For example, the memory (120) can store electrocardiogram data (10) received through a sensing unit (130) (or communication interface) to be described later. In addition, the memory (120) can store a neural network model (20) learned to identify the condition of a patient, and can store program codes that operate the neural network model (20) to perform learning, program codes that operate the neural network model (20) to receive electrocardiogram data (10) and perform inference according to the purpose of use of the computing device (100), and processed data generated as the program code is executed.

The sensing unit (130) senses the patient's electrocardiogram signal to obtain electrocardiogram data (10). For example, the electrical signal of the patient's heartbeat can be detected through a pair of electrodes included in the sensing unit (130), thereby obtaining the patient's single-induced electrocardiogram data (10). Meanwhile, the sensing unit (130) may also include at least one of an acceleration sensor, a gyro sensor, and an IMU sensor to detect the motion of the computing device (100).

The display (140) can display various images. The images include both still images and moving images. The display (140) can output information regarding the measurement method and guide information.

The display (140) may be implemented as a display in various forms, such as an LCD (Liquid Crystal Display Panel), an OLED (Organic Light Emitting Diodes), an LCoS (Liquid Crystal on Silicon), a DLP (Digital Light Processing), etc. In addition, the display (140) may also include a driving circuit, a backlight unit, etc., which may be implemented in forms such as an a-si TFT, an LTPS (low temperature poly silicon) TFT, an OTFT (organic TFT), etc.

Meanwhile, the display (140) may be implemented as a touch screen by being combined with a touch panel, and in this case, the display (140) may perform the function of not only an output interface that outputs an image through the touch screen, but also an input interface that receives a user's touch input.

FIG. 3 is a flowchart of a method for determining a user's status based on a low-voltage electrocardiogram signal according to an embodiment of the present disclosure.

The processor (110) can obtain first induction electrocardiogram data (10) for the user based on the first measurement method of the single-induction electrocardiogram measuring device (S310). Specifically, the processor (110) can detect the user's electrocardiogram signal through a plurality of electrodes included in the single-induction electrocardiogram measuring device, and obtain electrocardiogram data (10) corresponding to the electrocardiogram signal through digital processing (e.g., A/D conversion) of the electrocardiogram signal.

At this time, the plurality of electrodes may include a first electrode that contacts the user's first body and a second electrode that contacts the second body. The processor (110) may obtain first induced electrocardiogram data (10) for the user using the first electrode and the second electrode.

For example, when a single-inductor electrocardiogram (ECG) measurement device is implemented as a smartwatch, the first electrode may be placed on the back of the smartwatch, and the second electrode may be placed on the side of the smartwatch.

The first measurement method may be a measurement method of Lead 1 among multiple measurement methods for obtaining electrocardiogram data (10) set in a single-inductor electrocardiogram measurement device. At this time, the first measurement method may be a method in which the first electrode arranged on the back of the smartwatch contacts the user's left wrist and the second electrode arranged on the side of the smartwatch contacts the user's right finger. However, the present invention is not limited thereto, and the first measurement method may be set to the last measurement method in which the user's normal electrocardiogram data (10) that is not low voltage was obtained. That is, if the user recently obtained normal electrocardiogram data (10) that is not low voltage according to the measurement method of Lead II, the first measurement method may be a measurement method of Lead II.

Meanwhile, the processor (110) may also obtain electrocardiogram data (10) from an external single-induction electrocardiogram measuring device through a communication interface of the computing device (100).

In addition, the processor (110) can determine whether the electrocardiogram signal corresponding to the acquired first-induced electrocardiogram data (10) is abnormal (S320). Specifically, the processor (110) can extract feature information from the acquired first-induced electrocardiogram data (10) based on the acquired first measurement method, and analyze whether the electrocardiogram signal corresponding to the first-induced electrocardiogram data is low voltage to determine whether the electrocardiogram signal is abnormal. For example, the processor (110) can determine whether the first-induced electrocardiogram data (10) is low voltage by determining whether the amplitude of the QRS wave is 0.5 mV from the electrocardiogram data. At this time, the processor (110) may apply different criteria for determining whether there is a low voltage depending on the measurement method. In the case of the limb lead method (lead I, lead II, lead III, lead Avr, lead aVL, and lead aVF), the processor may determine whether there is a low voltage by determining whether the amplitude of the QRS wave is 0.5 mV, and in the case of the precordial lead method (lead V1, lead V2, lead V3, lead V4, lead V5, and lead V6), the processor may determine whether there is a low voltage of the electrocardiogram signal corresponding to the first lead electrocardiogram data (10) obtained by determining whether the amplitude of the QRS wave is 1.0 mV.

Meanwhile, the processor (110) inputs the acquired first-derived electrocardiogram data (10) into a neural network model (hereinafter, the first neural network model) trained to determine whether the electrocardiogram signal corresponding to the electrocardiogram data is low-voltage, thereby determining whether the electrocardiogram signal corresponding to the first-derived electrocardiogram data (10) is low-voltage. Specifically, the first neural network model may be pre-trained based on learning data composed of input data including a plurality of single-derived electrocardiogram data and label data indicating whether the electrocardiogram signal corresponding to the plurality of single-derived electrocardiogram data is low-voltage. During the learning process, the processor (110) may calculate a loss function based on the difference between the output value of the first neural network model and the label data corresponding to the input data. The loss function may be defined as a cross entropy loss or an objective function that optimizes the balance between precision and recall. Based on the generated loss function, the processor (110) can adjust the weights of the model through backpropagation. By repeating this process, the processor (110) can obtain a pre-trained first neural network model. For example, the first neural network model can be implemented as a multi-layer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network, etc. In particular, the first neural network model can include a plurality of residual blocks that extract potential feature information of the input electrocardiogram data and a classifier that classifies whether there is a low voltage based on the extracted potential feature information. The processor (110) can determine whether there is a low voltage of an electrocardiogram signal corresponding to the first induced electrocardiogram data (10) based on the probability value of the low voltage output from the first neural network model.

And, if the processor (110) determines that the electrocardiogram signal is low voltage, it may propose to acquire second lead electrocardiogram data (10') for the user based on the second measurement method of the single lead electrocardiogram measurement device (S330).

Specifically, if the processor (110) determines that the electrocardiogram signal corresponding to the first induction electrocardiogram data (10) is low voltage, the processor (110) may suggest re-acquiring electrocardiogram data for the user according to another measurement method. To this end, the processor (110) may provide information guiding another measurement method using a single-induction electrocardiogram measurement device. For example, the processor (110) may display information on multiple measurement methods using a single-induction electrocardiogram measurement device through the display (140) of the single-induction electrocardiogram measurement device. Hereinafter, the present disclosure related thereto will be described in detail.

FIG. 4 is an exemplary diagram for explaining multiple measurement methods using a computing device (100) according to one embodiment of the present disclosure.

Meanwhile, according to one embodiment of the present disclosure, the first measurement method and the second measurement method may be set differently depending on the combination of the user's body parts that are respectively contacted by the plurality of electrodes included in the single-lead electrocardiogram measurement device. Referring to FIG. 4, the method of measuring an electrocardiogram signal using the single-lead electrocardiogram measurement device may be set in various ways depending on the lead type. To explain again with the above example, if the single-lead electrocardiogram measurement device is a smartwatch, the measurement method of Lead 1 may be a method in which the first electrode disposed on the back of the smartwatch contacts the user's left wrist and the second electrode disposed on the side of the smartwatch contacts the user's right finger. In addition, the measurement method of Lead II may be a method in which the first electrode disposed on the back of the smartwatch contacts the user's left lower abdomen and the second electrode disposed on the side of the smartwatch contacts the user's right finger. In addition, the measurement method of Lead III may be a method in which the first electrode disposed on the back of the smartwatch contacts the user's left lower abdomen and the user's left finger contacts the second electrode disposed on the side of the smartwatch. And, the measurement method of lead V1 may be a method in which the first electrode arranged on the back of the smartwatch is contacted with the sternal side of the user's right fourth intercostal space and the user's right finger (or left finger) is contacted with the second electrode arranged on the side of the smartwatch. And, the measurement method of lead V2 may be a method in which the first electrode arranged on the back of the smartwatch is contacted with the sternal side of the user's left fourth intercostal space and the user's right finger (or left finger) is contacted with the second electrode arranged on the side of the smartwatch. And, the measurement method of lead V3 may be a method in which the first electrode is contacted with a body part located between the contact position of the first electrode in the measurement method of V2 and the contact position of the first electrode in the measurement method of V4 described below, and the user's right finger (or left finger) is contacted with the second electrode. And, the measurement method of lead V4 may be a method in which the first electrode arranged on the back of the smartwatch is contacted with the 5th intercostal space along the center line of the user's left clavicle and the user's right finger (or left finger) is contacted with the second electrode. And, the measurement method of Lead V5 may be a method in which the first electrode placed on the back of the smartwatch is contacted to the fifth intercostal space along the user's left anterior axillary line (or midaxillary line), and the user's right finger (or left finger) is contacted to the second electrode. And, the measurement method of Lead V6 may be a method in which the first electrode placed on the back of the smartwatch is contacted to the fifth intercostal space along the user's left middle axillary line (or midaxillary line), and the user's right finger (or left finger) is contacted to the second electrode.

FIG. 5 is an exemplary diagram illustrating a method for a computing device (100) to propose a second measurement method according to an embodiment of the present disclosure.

If the processor (110) determines that the electrocardiogram signal corresponding to the first induced electrocardiogram data (10) acquired through the first measurement method is low voltage, the processor (110) may suggest to the user the remaining measurement methods other than the first measurement method. Specifically, the processor (110) may display the lead types of the remaining measurement methods and information on the user's body part in contact with each electrode on the display (140).

In particular, the processor (110) may display information about a body part in contact with the first electrode and the second electrode included in the single-induction electrocardiogram measurement device through the display (140). Referring to FIG. 5, the processor (110) may notify the user of a low voltage of an electrocardiogram signal through the display (140) of the smartwatch and recommend re-measuring the electrocardiogram data through another measurement method (i.e., the second measurement method). At this time, the display (140) may display text and graphics indicating the body parts of the user in contact with the first electrode and the second electrode included in the smartwatch. Specifically, when the second measurement method is a Lead II measurement method, the processor (110) may display graphic objects corresponding to the user's body and the smartwatch on the display (140), and display text indicating the location where the smartwatch is placed on the user's body and the body parts of the user in contact with the first and second electrodes.

According to one embodiment of the present disclosure, referring to FIG. 3, if the processor (110) determines that the electrocardiogram signal is normal and not low voltage, it inputs the acquired first induced electrocardiogram data (10) into a pre-trained neural network model (20), obtains a score corresponding to the user's condition, and analyzes the user's condition by comparing the obtained score with a reference value. Hereinafter, the neural network model (20) trained to determine the user's condition and output a score corresponding to the user's condition is referred to as a second neural network model.

The processor (110) can analyze the first-derived electrocardiogram data (10) using a second neural network model (20) trained based on learning data including a plurality of electrocardiogram data. The processor (110) can extract feature information related to the user's condition from the first-derived electrocardiogram data (10) using the second neural network model (20), and can identify the patient's condition based on the extracted feature information. Here, the user's condition may be a heart disease such as left ventricular systolic dysfunction, hypovolemia, etc. The processor (110) can obtain learning data including input data composed of a plurality of electrocardiogram data obtained from different patients and label data in which a label indicating whether each patient has a heart disease (e.g., left ventricular systolic dysfunction) is assigned, and can train the second neural network model (20) in advance using the obtained learning data. In particular, the input data may be composed of electrocardiogram data corresponding to the same lead type as the first-derived electrocardiogram data (10).

The training data may include feature information (e.g., P wave, QRS complex, T wave, etc.) extracted from an electrocardiogram signal (more specifically, an electrocardiogram signal corresponding to the electrocardiogram data). The processor (110) may train a second neural network model (20) to determine the possibility of a specific heart disease in a patient based on this feature information. The processor (110) may input the training data into the second neural network model (20) and calculate a loss function based on the difference between the output value of the second neural network model (20) and the label data during the training process. The loss function may be defined as a cross entropy loss or an objective function that optimizes the balance between precision and recall. Based on the calculated loss function, the processor (110) can adjust the weights of the second neural network model (20) through backpropagation. By repeating this process, the processor (110) can improve the classification performance of the second neural network model (20) for heart disease, and ultimately obtain a second neural network model (20) trained to identify the possibility of a specific heart disease in a patient based on electrocardiogram data. The second neural network model (20) can produce a probability value or score for the possibility of a specific heart disease.

Meanwhile, the electrocardiogram signal corresponding to the electrocardiogram data included in the training data used to train the second neural network model (20) may have a commonly preset length. Accordingly, when the processor (110) acquires electrocardiogram data from a patient, the processor (110) may divide the electrocardiogram signal corresponding to the acquired electrocardiogram data into preset lengths or divide the electrocardiogram signal into multiple segments by applying a window of preset length along the time axis. For example, the electrocardiogram signal may be measured in 12 leads of 10 seconds in length and may be a signal measured in 500 points per second.

Meanwhile, the second neural network model (20) can be implemented as a multi-layer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network, etc.

When the processor (110) obtains a score corresponding to the user's condition from the second neural network model (20), the processor (110) can determine the user's condition by comparing the obtained score with a reference value. For example, if the user's condition is a heart disease such as left ventricular systolic dysfunction, the processor (110) can determine that the user belongs to a high-risk group for left ventricular systolic dysfunction if the score obtained from the second neural network model (20) is equal to or greater than the reference value, and can determine that the user belongs to a low-risk group for left ventricular systolic dysfunction if the obtained score is less than the reference value.

Meanwhile, the process of identifying the user's status using the second neural network model (20) may be performed in the computing device (100) when the computing device (100) is implemented as a smart watch, or may be performed through an external artificial intelligence server including the second neural network model (20) connected to the computing device (100) through a network. Alternatively, in the case where the computing device (100) is the aforementioned external artificial intelligence server, the process may be performed by the computing device (100) after acquiring the first induced electrocardiogram data (10) from a smart watch connected to the computing device (100) through a network.

At this time, among the plurality of pre-trained neural network models corresponding to different measurement methods of the single-lead electrocardiogram measuring device, the pre-trained neural network model (20) corresponding to the first measurement method can be selected. That is, the computing device (100) selects the second neural network model (20) corresponding to the first measurement method used to obtain the first-lead electrocardiogram data (10) from among the plurality of neural network models corresponding to different measurement methods (or different lead types), and can identify the user's condition using the selected second neural network model (20). At this time, the plurality of neural network models can be trained using the plurality of electrocardiogram data of different lead types obtained using the single-lead electrocardiogram device as training data, respectively. In this regard, since it overlaps with the method of training the above-described second neural network model (20), a detailed description thereof will be omitted.

Meanwhile, according to one embodiment of the present disclosure, when second-induced electrocardiogram data (10') for a user is acquired based on the second measurement method, the processor (110) can determine whether the electrocardiogram signal corresponding to the acquired second-induced electrocardiogram data (10') is low voltage. In addition, when the processor (110) determines that the electrocardiogram signal corresponding to the acquired second-induced electrocardiogram data (10') is not low voltage, the processor (110) can set the second measurement method as a reference measurement method for determining the user's status.

That is, when the second-induced electrocardiogram data (10') for the user is acquired based on the proposed second measurement method, the processor (110) can determine whether the electrocardiogram signal corresponding to the acquired second electrocardiogram data is also low voltage. Here, the first-induced electrocardiogram data (10) and the second-induced electrocardiogram data (10') may be electrocardiogram data of different lead types. In addition, when the processor (110) determines that the electrocardiogram signal corresponding to the second-induced electrocardiogram data (10') is normal and not low voltage, the processor (110) may set the second measurement method as a reference measurement method for determining the user's status, and thereafter, when the user's status is determined or a request for determining the user's status is received, the processor may recommend measuring the electrocardiogram data by the second measurement method set as the reference measurement method.

According to one embodiment of the present disclosure, the processor (110) may obtain second-induced electrocardiogram data (10') for a user based on a second measurement method, receive lead type information of the second measurement method through an input interface, and determine whether an electrocardiogram signal corresponding to the obtained second-induced electrocardiogram data (10') for the lead type is low voltage. When the processor (110) obtains the second-induced electrocardiogram data (10'), it may receive information on the measurement method of the second-induced electrocardiogram data (10') through the input interface. For example, the processor (110) may determine the second measurement method and lead type performed by the user through a measurement method UI selected by the user from among a plurality of measurement method type (or lead type) UIs displayed on the display (140). Alternatively, information on the second measurement method and lead type may be input through a virtual keyboard displayed on the display (140). This may be to select the remaining measurement methods that have not been performed when the electrocardiogram signal corresponding to the second induced electrocardiogram data (10') acquired through the second measurement method is judged to be low voltage.

Meanwhile, the processor (110) may identify the type of measurement method by detecting the movement of the computing device (100) using a gyro sensor, acceleration sensor, IMU sensor, etc. of the sensing unit included in the computing device (100). To this end, the memory (120) may store sensing values related to the movement of the computing device (100) when each measurement method is performed.

In addition, the processor (110) can obtain and store the sensing value of the sensor in the second measurement method, and determine whether the user obtains electrocardiogram data for status determination using the set reference measurement method based on the stored sensing value. The processor (110) can obtain sensing values regarding the movement of the computing device (100) while performing the second measurement method using a gyro sensor, an acceleration sensor, an IMU sensor, or the like of a sensing unit included in the computing device (100). In addition, when the second measurement method is set as the reference measurement method, the processor (110) can store the sensing value regarding the second measurement method, and thereafter determine whether the user uses the second measurement method for obtaining normal electrocardiogram data whenever obtaining electrocardiogram data. At this time, if it is not the second measurement method, the processor (110) can sound an alarm or display a message requesting to perform the second measurement method through the display (140).

Meanwhile, if the processor (110) determines that the electrocardiogram signal corresponding to the electrocardiogram data acquired according to the second measurement method is low voltage, the processor (110) may request acquisition of electrocardiogram data for other leads through a third measurement method corresponding to a lead type different from the first and second measurement methods. The processor (110) may set the first measurement method to Lead I, the second measurement method to Lead II, and the third measurement method to Lead III, and then sequentially suggest a measurement method to the user each time the electrocardiogram signal corresponding to the electrocardiogram data is determined to be low voltage.

According to one embodiment of the present disclosure, if the processor (110) determines that the electrocardiogram signal is low voltage, the processor (110) may analyze the electrocardiogram data to determine the cause of the low voltage. The cause of the low voltage of the electrocardiogram signal may include 1) internal factors of the heart (e.g., deviation of the electrical axis of the user's heart from the normal range, myocardial infarction, heart failure, etc.), 2) physical factors other than the user's heart (obesity, muscle abnormality, emphysema, hypothyroidism, dehydration, etc.), and 3) external factors (poor electrode contact, electrode position error, etc.). At this time, the processor (110) may analyze the low voltage electrocardiogram signal to guide the removal of interference factors of the electrocardiogram signal. In particular, the processor (110) may analyze the cause of the low voltage of the electrocardiogram data (e.g., first-derived electrocardiogram data (10)) acquired by a neural network model (third neural network model) trained with a plurality of electrocardiogram data classified according to the cause of the low voltage as learning data. Here, the third neural network model may be a model trained based on training data including multiple electrocardiogram data, each labeled with a different cause of low voltage. During the training process, the processor (110) may calculate a loss function based on the difference between the output value of the third neural network model and the label data corresponding to the input data. The loss function may be defined as a cross entropy loss or an objective function that optimizes the balance between precision and recall. Through this, the third neural network model may extract feature information related to low voltage within the input electrocardiogram data and classify the cause of low voltage in the electrocardiogram data. The third neural network model may be implemented as a multi-layer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network, etc.

When the processor (110) determines that the electrical axis of the user's heart is outside the normal range (+30° to -90°), if the first measurement method is a measurement method of Lead 1 and the second measurement method is a measurement method of Lead Ⅱ, and the electrocardiogram signal corresponding to the first induction electrocardiogram data (10) acquired according to the first measurement method is determined to be low voltage, and the electrocardiogram signal corresponding to the second induction electrocardiogram data (10') acquired according to the second measurement method is determined to be normal rather than low voltage, the processor (110) may determine that the electrical axis of the user's heart is outside the normal range (+30° to -90°), and may then propose to acquire electrocardiogram data for determining the user's condition using the second measurement method.

According to one embodiment of the present disclosure, if the first measurement method is a Lead 1 measurement method and the electrocardiogram signal is determined to be low voltage, the processor (110) may propose to disinfect the body part of the user that comes into contact with the first measurement method or adjust the position of the single-lead electrocardiogram measurement device in the first measurement method. At this time, the processor (110) may re-acquire third-lead electrocardiogram data for the user based on the first measurement method and determine whether the electrocardiogram signal corresponding to the acquired third-lead electrocardiogram data is low voltage.

Meanwhile, if the processor (110) determines that the electrocardiogram signal is low voltage, it can apply a weight to the electrocardiogram data and determine the user's condition based on the electrocardiogram data to which the weight has been applied. That is, the electrocardiogram data to which the weight has been applied can be used in a pre-trained second neural network model to obtain a score corresponding to the user's condition. In particular, if the processor (110) determines that the electrocardiogram signal is repeatedly voltage despite the proposed multiple measurement methods, it can apply a weight to the electrocardiogram data.

In addition, the processor (110) may acquire normal electrocardiogram data and then determine the user's condition by using a neural network model (hereinafter, the fourth neural network model) trained to convert abnormal electrocardiogram data into normal electrocardiogram data and output it. In particular, if the processor (110) determines that the electrocardiogram signal is repeatedly voltage despite the proposed multiple measurement methods, the processor (110) may use the fourth neural network model. In addition, if the processor (110) determines that the cause of the low voltage of the electrocardiogram signal cannot be resolved, such as 1) internal factors of the heart (e.g., diseases such as myocardial infarction, heart failure, etc.) and 2) physical factors other than the user's heart (e.g., obesity), the processor (110) may use the fourth neural network model.

The fourth neural network model can be trained using training data consisting of pairs of multiple normal ECG data from a single lead and multiple low-voltage ECG data corresponding to the multiple normal ECG data, based on the same patient. The low-voltage ECG data can also be generated by adjusting the signal attenuation of the normal ECG data. In particular, the fourth neural network model can include sub-neural network models corresponding to each measurement method.

Meanwhile, the processor (110) may input normal electrocardiogram data acquired from the fourth neural network model into a pre-trained neural network model (the fifth neural network model) to acquire electrocardiogram data of the remaining leads. In particular, the processor (110) may input specific single-lead electrocardiogram data into the fifth neural network model to acquire the remaining multiple-lead electrocardiogram data among the standard 12-lead electrocardiogram data, and may also identify the user's condition based on the standard 12-lead electrocardiogram data. At this time, the fifth neural network model may be a model trained to input specific single-lead electrocardiogram data and output the remaining lead electrocardiogram data. In particular, the fifth neural network model may acquire multiple electrocardiogram data of the remaining leads by identifying the style of each single-lead electrocardiogram data. To this end, the fifth neural network model may also receive lead type information, etc. of the input single-lead electrocardiogram data. Here, the style may be latent feature information that the single-lead electrocardiogram data has in common for each lead type. The fifth neural network model can learn the correlation between the style of a specific input single-lead ECG data and the remaining lead ECG data. The fifth neural network model can be trained based on training data in which any one standard 12-lead ECG data is set as input data and the remaining ECG data is set as label data. The fifth neural network model can be implemented using a multi-layer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network, etc.

FIG. 6 is a sequence diagram illustrating a method for determining a user's condition based on a low-voltage electrocardiogram signal of a computing device (100) implemented as an artificial intelligence server according to another embodiment of the present disclosure. The description of the embodiment of the present disclosure described above may be applied equally to the embodiment of FIG. 6 within the scope of overlap.

The computing device (300) can obtain first-induction electrocardiogram data (10) according to a first measurement method from a smart watch (400) (or other single-induction electrocardiogram measurement device) through a communication interface of the computing device (300). At this time, the computing device (300) can also obtain first-induction electrocardiogram data and information about the measurement method together. In addition, the computing device (300) can determine whether the electrocardiogram signal corresponding to the obtained first-induction electrocardiogram data (10) is low voltage. If the computing device (300) determines that the electrocardiogram signal corresponding to the first-induction electrocardiogram data (10) is low voltage, the computing device (300) can request the smart watch (400) to obtain second-induction electrocardiogram data (10') according to a different measurement method. At this time, the smart watch (400) can display information guiding the different measurement method through the display (140). When the smart watch (400) acquires the second-induced electrocardiogram data (10') according to the second measurement method, the computing device (300) acquires the second-induced electrocardiogram data (10') according to the second measurement method from the smart watch (400), determines whether the electrocardiogram signal corresponding to the acquired second-induced electrocardiogram data (10') is low voltage, and if the electrocardiogram signal corresponding to the second-induced electrocardiogram data (10') is determined not to be low voltage, the computing device (300) can analyze the second-induced electrocardiogram data (10') using the pre-learned neural network model (20) and determine the user's status. In addition, the computing device (300) can transmit the user's status information to the smart watch (400).

Meanwhile, as described above in FIG. 1, according to one embodiment of the present disclosure, whether an electrocardiogram signal is abnormal may be determined based on the degree of noise included in the electrocardiogram signal and whether feature information is identified on a graph corresponding to the electrocardiogram signal.

Specifically, the processor (110) inputs electrocardiogram data into a pre-trained neural network model (hereinafter, the sixth neural network model) to obtain a noise score indicating the degree of noise included in the electrocardiogram signal, and can identify the degree of noise included in the electrocardiogram signal based on the obtained noise score. If the noise score is greater than a preset value, the processor (110) can determine that the electrocardiogram signal is abnormal.

Here, the sixth neural network model may be a model trained based on input data composed of a plurality of clean electrocardiogram data with or without noise, and training data including label data composed of values that quantify noise information of the plurality of electrocardiogram data constituting the input data. During the training process, the processor (110) may calculate a loss function based on the difference between the output value of the sixth neural network model and the label data corresponding to the input data. The loss function may be defined as a cross entropy loss or an objective function that optimizes the balance between precision and recall. Through this, the sixth neural network model may extract feature information related to noise in the input electrocardiogram data and output a noise score of the electrocardiogram data. The sixth neural network model may be implemented as a multi-layer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network, etc.

In addition, the processor (110) can check whether feature information (e.g., feature points such as P waves and Q waves) is identified on an electrocardiogram graph corresponding to the electrocardiogram data, and if it is confirmed that at least one of the feature information is not identified, the electrocardiogram signal can be determined to be abnormal.

The seventh neural network model may be a model trained based on input data consisting of multiple electrocardiogram data and training data including label data to which feature information included in the multiple electrocardiogram data has been assigned a level. During the training process, the processor (110) may calculate a loss function based on the difference between the output value of the seventh neural network model and the label data corresponding to the input data. The loss function may be defined as a cross entropy loss or an objective function that optimizes the balance between precision and recall. Through this, the seventh neural network model may output classification information (e.g., a probability value of each feature information) regarding each feature information included in the input electrocardiogram data (i.e., identified on the electrocardiogram graph). In addition, the processor (110) may determine whether each feature information is identified on the electrocardiogram graph based on the classification information. Whether the feature information is identified may be determined by comparing the probability value of each feature information with a preset value set for each feature information. The seventh neural network model can be implemented as a multi-layer perceptron (MLP), convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network, or transformer neural network.

FIG. 7 is a detailed configuration diagram of a computing device (700) according to another embodiment of the present disclosure. Referring to FIG. 7, a computing device (700) according to an embodiment of the present disclosure includes a processor (710), a memory (720), a sensing unit (730), a display (740), a user interface (750), a communication interface (760), and a speaker (770). Among the configurations illustrated in FIG. 7, a detailed description of configurations that overlap with those illustrated in FIG. 2 will be omitted.

The user interface (750) is a component used by the computing device (700) to perform interaction with the user, and may include at least one of a touch sensor, a motion sensor, a button, a jog dial, and a switch, but is not limited thereto. The processor (710) may receive information regarding a measurement method and lead type through the user interface (750).

A communication interface (760) according to an embodiment of the present disclosure may be understood as a component that transmits and receives data through any known wired or wireless communication system. For example, the communication interface (760) may perform data transmission and reception using a wired or wireless communication system such as a local area network (LAN), wideband code division multiple access (WCDMA), long term evolution (LTE), wireless broadband internet (WiBro), fifth generation mobile communication (5G), ultrawide-band, ZigBee, radio frequency (RF) communication, wireless LAN, wireless fidelity, near field communication (NFC), or Bluetooth. Since the above-described communication systems are only examples, the wired and wireless communication system for data transmission and reception of the communication interface (760) may be applied in various ways other than the above-described examples. The communication interface (760) can receive data necessary for the processor (110) to perform calculations through wired or wireless communication with any system or any client, etc. In addition, the communication interface (760) can transmit data generated through calculations of the processor (710) through wired or wireless communication with any system or any client, etc. For example, the communication interface (760) can receive electrocardiogram data of a patient through communication with a database in a hospital environment, a cloud server that performs tasks such as standardization of medical data, or another computing device (e.g., a smart watch, etc.). The communication interface (760) can transmit output data of the neural network model (20), and intermediate data, processed data, etc. derived from the calculation process of the processor (110) through communication with the aforementioned database, server, or computing device, etc. Data transmission and reception can be performed using a wired or wireless communication system such as a wireless LAN, Wi-Fi (wireless fidelity), near field communication (NFC), or Bluetooth. The above-described communication systems are only examples, and thus, wired and wireless communication systems for data transmission and reception of the communication interface (760) can be applied in various ways other than the above-described examples.

The communication interface (760) can receive data necessary for the processor (110) to perform calculations through wired or wireless communication with any system or any client, etc. In addition, the communication interface (760) can transmit data generated through calculations of the processor (710) through wired or wireless communication with any system or any client, etc. For example, the communication interface (760) can receive electrocardiogram data of a patient through communication with a cloud server that performs tasks such as databases in a hospital environment, standardization of medical data, or a computing device, etc. The communication interface (760) can transmit output data of the neural network model (20), intermediate data, processed data, etc. derived from the calculation process of the processor (710), etc. through communication with the aforementioned database, server, or computing device, etc.

The speaker (780) is a component that outputs various audio data that have undergone various processing operations, such as decoding, amplification, and noise filtering, by an audio processing unit (not shown). The speaker (780) can output various notification sounds or voice messages. According to one embodiment of the present disclosure, the processor (910) can convert an electrical signal into a user's (1) voice and output it through the speaker (780). For example, the speaker (780) can output a voice message that notifies whether there is a low voltage or guides a measurement method.

Meanwhile, according to one embodiment of the present disclosure, a non-transitory computer-readable medium storing a program for performing a method of determining a user's condition based on the aforementioned low-voltage electrocardiogram signal may be provided. Here, the non-transitory computer-readable medium refers to a medium that semi-permanently stores data and can be read by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, the various applications or programs described above may be stored and provided on a non-transitory computer-readable medium, such as a CD, DVD, hard disk, Blu-ray disk, USB, memory card, or ROM.

The various embodiments of the present disclosure described above can be combined with additional embodiments and modified within the scope understood by those skilled in the art in light of the detailed description above. It should be understood that the embodiments of the present disclosure are illustrative in all respects and not restrictive. For example, each component described as a single component may be implemented in a distributed manner, and likewise, components described as distributed may be implemented in a combined manner. Accordingly, all changes or modifications derived from the meaning, scope, and equivalent concepts of the claims of the present disclosure should be construed as being included within the scope of the present disclosure.

Claims (12)

A method for determining a user's condition based on an abnormal electrocardiogram signal, the method being performed by a computing device including at least one processor, A step of obtaining first lead electrocardiogram data for a user based on a first measurement method of a single lead electrocardiogram measuring device; A step of determining whether an electrocardiogram signal corresponding to the first derived electrocardiogram data obtained above is abnormal; and a step of proposing to acquire second lead electrocardiogram data for the user based on a second measurement method of the single lead electrocardiogram measurement device, if the electrocardiogram signal is determined to be abnormal; method. In the first paragraph, The above first measurement method and the above second measurement method, It is set differently depending on the combination of the user's body parts that are each in contact with the multiple electrodes included in the single-induction electrocardiogram measuring device. method. In the first paragraph, The steps suggested above are: A step of displaying body part information in contact with the first electrode and the second electrode included in the single-induction electrocardiogram measuring device through a display; method. In the first paragraph, When second-induced electrocardiogram data for the user is acquired based on the second measurement method, a step of determining whether an electrocardiogram signal corresponding to the acquired second-induced electrocardiogram data is abnormal; and If it is determined that the electrocardiogram signal corresponding to the acquired second induced electrocardiogram data is not abnormal, a step of setting the second measurement method as a reference measurement method for determining the user's condition is included; method. In paragraph 4, The step of determining whether the electrocardiogram signal corresponding to the second-derived electrocardiogram data obtained above is abnormal is as follows: A step of obtaining second-induced electrocardiogram data for the user based on the second measurement method and receiving lead type information of the second measurement method through an input interface; and A step of determining whether an electrocardiogram signal corresponding to the acquired second-derived electrocardiogram data for the above lead type is abnormal; method. In paragraph 4, A step of detecting the motion of a single-induction electrocardiogram measuring device in the second measurement method through a sensing unit to obtain a sensing value; and A step of determining whether the user obtains electrocardiogram data for determining the condition using the set reference measurement method based on the sensing value; method. In the first paragraph, If the above electrocardiogram signal is determined to be normal, a step of inputting the acquired first-derived electrocardiogram data into a pre-trained neural network model to obtain a score corresponding to the user's condition; A step of analyzing the status of the user by comparing the obtained score with a reference value; including; method. In paragraph 6, A step of selecting a pre-learned neural network model corresponding to a first measurement method among a plurality of pre-learned neural network models corresponding to different measurement methods of the single-induction electrocardiogram measuring device; comprising; method. In the first paragraph, The step of proposing to obtain second-derived electrocardiogram data for the above user is: The first measurement method is a measurement method of Lead 1, and if the electrocardiogram signal is determined to be abnormal due to low voltage, a step of suggesting disinfecting the body part of the user that comes into contact with the first measurement method or adjusting the position of the single-lead electrocardiogram measurement device in the first measurement method is included. method. In paragraph 4, If the electrocardiogram signal corresponding to the acquired second electrocardiogram data is determined to be normal, a step of determining that the electrical axis of the user's heart is out of the normal range is included; method. In a single-lead electrocardiogram measuring device that determines the user's condition based on an abnormal electrocardiogram signal, Memory containing program codes; A sensing unit that acquires the user's electrocardiogram data; display; and One or more processors for obtaining first-lead electrocardiogram data for a user based on a first measurement method of the single-lead electrocardiogram measurement device, determining whether an electrocardiogram signal corresponding to the obtained first-lead electrocardiogram data is abnormal, and proposing to obtain second-lead electrocardiogram data for the user based on a second measurement method of the single-lead electrocardiogram measurement device through the display if the electrocardiogram signal is determined to be abnormal; Single lead electrocardiogram measuring device. A computer program stored in a computer-readable storage medium, wherein the computer program, when executed on one or more processors, performs an operation of determining a user's condition based on an abnormal electrocardiogram signal. The above action is, An operation of obtaining first lead electrocardiogram data for a user based on a first measurement method of a single lead electrocardiogram measuring device; An operation for determining whether an electrocardiogram signal corresponding to the first derived electrocardiogram data obtained above is abnormal; and An operation of proposing to acquire second lead electrocardiogram data for the user based on a second measurement method of the single lead electrocardiogram measurement device, if the electrocardiogram signal is determined to be abnormal; Computer program.