Health status prediction system utilizing single-lead electrocardiogram equipment

Abstract

The present invention discloses a health condition prediction system using a single-lead electrocardiogram device, which includes a single-lead electrocardiogram measurement unit (110) having two electrodes and measuring electrocardiograms of two or more electrical axes with a time lag to obtain asynchronous electrocardiogram data, and a prediction unit (120) that predicts the presence or absence of a disease and the severity of a disease from the asynchronous electrocardiogram data input from the single-lead electrocardiogram measurement unit (110) through a diagnostic algorithm (121) previously constructed by learning a standard lead electrocardiogram measured on the same electrical axis and a standard multiple electrocardiogram data set matching the presence or absence of a disease and the severity of a disease corresponding to the standard lead electrocardiogram, and can predict a disease from asynchronous electrocardiogram data of two or more leads measured by the single-lead electrocardiogram device.

Description

Technical Field

[0001] The present invention relates to a health status prediction system using a single-lead electrocardiogram device, which can easily measure asynchronous electrocardiograms of two or more leads by utilizing a single-lead electrocardiogram device or a six-lead electrocardiogram device, and can predict diseases from the measured asynchronous electrocardiogram data of two or more leads.

[0002]

Background Art

[0003] As is well known, after the development of electrocardiograms, electrocardiogram-related knowledge has expanded geometrically, and electrocardiogram examinations can obtain information about the electrical function of the heart and diagnose various heart diseases such as arrhythmias, coronary artery diseases, and myocardial diseases.

[0004] Recently, research on AI algorithms for electrocardiograms has been actively conducted, and the AI algorithm may sense heart failure, predict atrial fibrillation from an arrhythmic rhythm, or determine gender.

[0005] Thus, it is possible to overcome human limitations, sense subtle changes in electrocardiogram waveforms by an AI algorithm, and thus improve electrocardiogram interpretation.

[0006] On the other hand, the electrocardiograms used in the medical field are 12-lead electrocardiograms, which are measured by attaching 10 electrodes including 3 limb electrodes, 6 chest electrodes, and 1 ground electrode, and the measured electrocardiogram data can be remotely transmitted.

[0007] However, since it is inconvenient to expose the chest area and attach 10 electrodes for use in daily life, portable pad measuring devices, Galaxy Watches, Apple Watches, etc. that can measure single-lead electrocardiograms may also be used.

[0008] While single-lead electrocardiograms measured in this way can be used to diagnose arrhythmias, they have limitations in their use for diagnosing diseases that require electrocardiogram information from multiple leads, such as myocardial infarction.

[0009] Therefore, there is a need for a technology that can easily measure asynchronous electrocardiograms of two or more leads using a single-lead or six-lead electrocardiogram device, and that can predict diseases from the measured asynchronous electrocardiogram data of two or more leads.

[0010] [Overview of the project] [Problems that the invention aims to solve]

[0011] The technical problem that the present invention aims to solve is to provide a health status prediction system utilizing a single-lead electrocardiogram (ECG) device that can easily measure two or more asynchronous ECGs using a single-lead ECG device or a six-lead ECG device, and can predict diseases from the measured two or more asynchronous ECG data.

[0012] [Means for solving the problem]

[0013] To achieve the aforementioned objectives, an embodiment of the present invention provides a health status prediction system utilizing a single-lead electrocardiogram device, comprising: a single-lead electrocardiogram measurement unit equipped with two electrodes that acquires asynchronous electrocardiogram data by measuring electrocardiograms of two or more electrical axes with a time difference; and a prediction unit that predicts the presence or absence and severity of a disease from asynchronous electrocardiogram data input from the single-lead electrocardiogram measurement unit via a diagnostic algorithm pre-constructed by learning a standard multiple electrocardiogram dataset that matches the presence or absence and severity of disease corresponding to the standard electrocardiogram measured on the same electrical axis.

[0014]

[0015] Here, the diagnostic algorithm can be constructed to take asynchronously measured standard lead electrocardiogram data as input and output and predict the presence or absence and severity of a disease corresponding to the standard lead electrocardiogram data.

[0016]

[0017] Furthermore, the diagnostic algorithm can be constructed to take standard lead electrocardiogram data obtained by removing time-series information from synchronously measured standard lead electrocardiogram data as input, and to output and predict the presence or absence and severity of a disease corresponding to the standard lead electrocardiogram data.

[0018]

[0019] Furthermore, the diagnostic algorithm may be constructed to convert multiple asynchronous electrocardiogram data, measured and input with a time delay from the single-lead electrocardiogram measurement unit, into synchronous electrocardiogram data, and then predict the presence or absence of disease and the severity of the disease from the converted synchronous electrocardiogram data.

[0020]

[0021] The system further includes an electrocardiogram data generation unit that generates multiple standard lead electrocardiogram data that are not measured by the single lead electrocardiogram measurement unit and do not match the standard lead electrocardiogram, based on multiple asynchronous electrocardiogram data measured by the single lead electrocardiogram measurement unit. The diagnostic algorithm can take the measured asynchronous electrocardiogram data and the generated standard lead electrocardiogram data as input, or take the generated standard lead electrocardiogram data as input, and output and predict the presence or absence of disease and the degree of disease.

[0022]

[0023] Furthermore, the diagnostic algorithm learns by reflecting the individual characteristics information of the person being examined, and the prediction unit can predict the presence or absence of disease and the severity of the disease from asynchronous electrocardiogram data from the single-lead electrocardiogram measurement unit, reflecting the individual characteristics information.

[0024]

[0025] In addition, the individual characteristic information can include demographic information on gender and age, basic health information on weight, back, and obesity level, disease-related information on past history, drug history, and family history, and examination information on vital signs and biological signals such as blood tests, genetic tests, blood pressure, and oxygen saturation.

[0026]

[0027] In addition, the single-lead electrocardiogram measurement unit can include a wearable electrocardiogram patch, a smartwatch, or an electrocardiogram bar.

[0028]

[0029] In addition, the prediction unit further includes a data conversion module that generates numerical data from the asynchronous electrocardiogram data measured by the single-lead electrocardiogram measurement unit through a specific mathematical formula, and the diagnostic algorithm can take the numerical data corresponding to the asynchronous electrocardiogram data as input.

[0030]

Advantages of the Invention

[0031] According to the present invention, it is possible to easily measure asynchronous electrocardiograms of two or more leads by utilizing a single-lead electrocardiogram device or a six-lead electrocardiogram device, predict diseases from the measured asynchronous electrocardiogram data of two or more leads, and generate multi-channel electrocardiogram data from the asynchronous electrocardiogram data of two or more leads, thereby outputting and predicting the presence or absence of diseases and the degree of diseases with higher accuracy.

[0032] <s

Brief Description of the Drawings

[0033] [Figure 1] It is a configuration diagram of a health status prediction system utilizing a single-lead electrocardiogram device according to an embodiment of the present invention.

[0034] [Figure 2] Figure 1 is a flowchart illustrating the prediction method using a health status prediction system that utilizes a single-lead electrocardiogram device.

[0035] [Modes for carrying out the invention]

[0036] Hereinafter, embodiments of the present invention having the features described above will be described in more detail based on the attached drawings.

[0037]

[0038] The health status prediction system utilizing a single-lead electrocardiogram device according to an embodiment of the present invention includes a single-lead electrocardiogram measurement unit 110 equipped with two electrodes that acquires asynchronous electrocardiogram data by measuring electrocardiograms of two or more electrical axes with a time difference, and a prediction unit 120 that predicts the presence or absence and severity of disease from the asynchronous electrocardiogram data input from the single-lead electrocardiogram measurement unit 110 via a diagnostic algorithm 121 that has been pre-constructed by learning a standard multiple electrocardiogram dataset that matches the presence or absence and severity of disease corresponding to the standard electrocardiogram measured on the same electrical axis. The gist of the system is to predict disease from asynchronous electrocardiogram data of two or more leads measured by a single-lead electrocardiogram device.

[0039]

[0040] The following section provides a detailed description of a health status prediction system utilizing a single-lead electrocardiogram device with the configuration described above, based on the diagrams.

[0041]

[0042] First, the single-lead electrocardiogram measurement unit 110, equipped with two electrodes, contacts two locations on the patient's body to measure electrocardiograms of two or more electrical axes with a time difference, acquires asynchronous electrocardiogram data, and transmits it to the prediction unit 120 via a short-range network.

[0043] For example, by placing electrodes on each hand to measure a lead I electrocardiogram corresponding to one electrical axis, and then placing electrodes on the right hand and left ankle (different from the aforementioned electrode contact body combination) to measure a lead II electrocardiogram corresponding to another electrical axis, it is possible to measure multiple asynchronous electrocardiograms for two or more different electrical axes.

[0044] Furthermore, the single-lead electrocardiogram measurement unit 110 measures an asynchronous or synchronous electrocardiogram including a wearable electrocardiogram patch 111, a smartwatch 112, or a 6-lead electrocardiogram bar that measures in a short time, which can perform contact or non-contact electrocardiogram measurements during daily life. The electrocardiogram data generation unit 130 further generates multiple electrocardiograms, which are then input into a diagnostic algorithm 121 constructed from standard lead electrocardiogram data, thereby predicting the corresponding health condition.

[0045] Here, the single-lead electrocardiogram measurement unit 110 can either measure a continuous electrocardiogram of the person being examined and transmit it to the prediction unit 120, or measure two electrocardiograms at time intervals and transmit them to the prediction unit 120.

[0046]

[0047] Subsequently, the prediction unit 120 predicts the health status from the asynchronous electrocardiogram data of the single-lead electrocardiogram measurement unit 110 via a pre-learned diagnostic algorithm 121. The diagnostic algorithm 121, which was previously constructed by learning standard lead electrocardiograms measured on the same electrical axis as the electrocardiogram measured by the single-lead electrocardiogram measurement unit 110, and a standard multi-electrocardiogram dataset that matches the presence or absence and severity of diseases corresponding to the standard lead electrocardiograms, predicts the presence or absence and severity of diseases from the asynchronous electrocardiogram data input from the single-lead electrocardiogram measurement unit 110 to understand the health status.

[0048] In other words, the diagnostic algorithm 121 is constructed to take standard lead electrocardiogram data measured asynchronously for any electrical axis as input and output the presence or absence and severity of disease corresponding to the standard lead electrocardiogram data, thereby enabling the prediction of health status from asynchronous electrocardiogram data transmitted from the single lead electrocardiogram measurement unit 110. Here, the standard lead electrocardiogram data can be divided and stored in 2.5-second units or stored asynchronously.

[0049] Alternatively, the diagnostic algorithm 121 can be constructed to take standard lead electrocardiogram data obtained by removing time-series information from standard lead electrocardiogram data measured synchronously for multiple electrical axes as input, and output the presence or absence and severity of disease corresponding to the standard lead electrocardiogram data to predict the health status from asynchronous electrocardiogram data transmitted from the single lead electrocardiogram measurement unit 110.

[0050] Alternatively, the diagnostic algorithm 121 can be constructed to predict the presence or absence of disease and the severity of the disease from the converted synchronous electrocardiogram data by converting multiple asynchronous electrocardiogram data, which are measured by the single-lead electrocardiogram measurement unit 110 and input with a time delay, into synchronous electrocardiogram data.

[0051] Therefore, the diagnostic algorithm 121 can match multiple asynchronous electrocardiogram data, measured by the single-lead electrocardiogram measurement unit 110 and input with a time lag, to asynchronously measured standard lead electrocardiogram data, standard lead electrocardiogram data from which time-series information has been removed, or synchronous standard lead electrocardiogram data, respectively.

[0052]

[0053] Furthermore, via the electrocardiogram data generation unit 130, multiple standard lead electrocardiogram data that are not measured by the single lead electrocardiogram measurement unit 110 and do not match the standard lead electrocardiogram are generated based on multiple asynchronous electrocardiogram data measured for two or more electrical axes by the single lead electrocardiogram measurement unit 110. The diagnostic algorithm 121 can take the asynchronous electrocardiogram data measured by the single lead electrocardiogram measurement unit 110 and the standard lead electrocardiogram data generated by the electrocardiogram data generation unit 130 as input, or take the standard lead electrocardiogram data generated by the electrocardiogram data generation unit 130 as input, and output and predict the presence or absence of disease and the degree of disease.

[0054] Here, when generating multiple standard lead electrocardiogram data, the unique characteristics of each lead electrocardiogram can be grasped, the asynchronous electrocardiogram data input from the single-lead electrocardiogram measurement unit 110 can be matched to the corresponding specific standard lead electrocardiogram data, and the remaining standard lead electrocardiogram data that does not match the specific standard lead electrocardiogram data can be generated to create new multiple standard lead electrocardiogram data.

[0055] Furthermore, the electrocardiogram data generation unit 130 can generate standard lead electrocardiogram data that is either asynchronous or synchronous. As mentioned above, when the diagnostic algorithm 121 uses asynchronously measured standard lead electrocardiogram data, or uses standard lead electrocardiogram data from which time-series information has been removed, it can generate asynchronous electrocardiogram data or generate electrocardiogram data without considering synchronization.

[0056] This allows for prediction of the presence or absence of diseases that should be diagnosed by multi-lead electrocardiograms, and diagnosis of the severity of those diseases, by utilizing the asynchronous electrocardiogram measured by the single-lead electrocardiogram measurement unit 110. This enables more accurate analysis of diseases based on multiple electrocardiogram pieces of information, allowing for measurement, diagnosis, screening, and prediction of health status.

[0057] For example, in the case of diseases that can be diagnosed with a single lead, such as arrhythmias, a two-lead electrocardiogram and multiple electrocardiograms generated based on it can be used to generate electrocardiograms for each bit with various electrical axes, enabling more accurate measurement, diagnosis, screening, and prediction of health status. In the case of diseases that can be diagnosed with multiple leads, such as myocardial infarction, multiple electrocardiograms can be generated to further diagnose myocardial infarction.

[0058] Alternatively, the diagnostic algorithm 121 could be a model that measures, diagnoses, screens, and predicts health status using only the electrocardiogram data corresponding to the two leads to be used from the standard lead electrocardiogram data for disease diagnosis. However, as mentioned above, it is not limited to two leads, and electrocardiogram data from additional generated leads can also be used.

[0059] In other words, the electrocardiogram data accumulated at most medical institutions is standard 12-lead electrocardiogram data. The electrocardiogram data generation unit 130 generates a standard 12-lead electrocardiogram from a 2-lead electrocardiogram. By inputting this generated electrocardiogram into an algorithm that measures, diagnoses, screens, and predicts health status using the medical institution's standard 12-lead electrocardiogram data, it becomes possible to output more accurate prediction results by utilizing the 2-lead electrocardiogram, and to predict a wider range of health conditions.

[0060] For example, when measuring with the single-lead electrocardiogram measurement unit 110, if the electrocardiogram data contains a lot of noise in some leads or sections, or if the electrode contact is separated and measurement cannot be performed properly, the electrocardiogram data generation unit 130 generates an electrocardiogram of a noise-free lead and fills in the missing electrocardiogram data, thereby enabling more accurate measurement, diagnosis, screening, and prediction of health status.

[0061] Furthermore, the single-lead electrocardiogram measurement unit 110 and the prediction unit 120 measure a reference electrocardiogram of the subject in their normal, healthy state to monitor their health. Subsequently, by comparing the electrocardiogram measured and input in real time by the single-lead electrocardiogram measurement unit 110 during daily life with the reference electrocardiogram, the system predicts whether the electrocardiogram was measured without errors or whether there are any abnormalities in the subject's health. In the event of an error or prediction of an abnormality, the warning unit 140 generates warning information, which can be transmitted along with a beep sound via the smartwatch-type single-lead electrocardiogram measurement unit 110 or a separate smart device.

[0062] Specifically, after saving a reference electrocardiogram measured by the single-lead electrocardiogram measurement unit 110 at the beginning of use, additional lead electrocardiograms can be generated, allowing for continuous monitoring of a 12-lead electrocardiogram. Here, the smartwatch 112 is taken with the right hand and placed against the abdomen to measure lead II electrocardiogram and save it as a reference electrocardiogram. Then, the smartwatch 112 is normally worn on the left hand, and lead I electrocardiogram is measured by touching it with the right hand. Simultaneously, this is used to generate multiple electrocardiogram leads, including lead II electrocardiogram, thereby enabling the smartwatch 112 to be used for a wider range of health condition measurements, diagnoses, screenings, and predictions.

[0063] Alternatively, prior to attaching the wearable ECG patch 111, it can be removed with both hands, a lead I ECG can be measured and saved as a reference ECG, then the wearable ECG patch 111 can be attached and a lead V ECG can be measured. By generating a synchronized multi-lead ECG based on the continuously measured or monitored lead V ECGs, more accurate measurement, diagnosis, screening, and prediction of health status can be achieved.

[0064] Furthermore, the diagnostic algorithm 121 learns by reflecting the individual characteristics information of the person being examined, and the prediction unit 120 can predict the presence or absence of disease and the severity of the disease from asynchronous electrocardiogram data from the single-lead electrocardiogram measurement unit 110, reflecting the individual characteristics information.

[0065]

[0066] For example, when applying standard lead electrocardiogram data used in training the diagnostic algorithm 121, it is possible to predict the presence or absence and severity of a corresponding disease by reflecting individual characteristic information, or by generating standard lead electrocardiogram data by the electrocardiogram data generation unit 130 while reflecting the individual characteristic information of the person whose electrocardiogram is measured by the single lead electrocardiogram measurement unit 110, thereby predicting the presence or absence and severity of a corresponding disease.

[0067] Here, individual characteristic information may include demographic information such as the subject's sex and age, basic health information such as weight, height, and obesity level, disease-related information such as past history, medication history, and family history, and test information such as blood tests, genetic tests, blood pressure, oxygen saturation, and vital signs and biological signals.

[0068]

[0069] On the other hand, the prediction unit 120 can accept not only graphed electrocardiograms but also digitized electrocardiograms as input. For example, it further includes a data conversion module 122 that converts asynchronous electrocardiogram data measured by the single-lead electrocardiogram measurement unit 110 into numerical data equivalent to an electrocardiogram using a specific mathematical formula, and the diagnostic algorithm 121 can also predict health status by taking numerical data equivalent to asynchronous electrocardiogram data as input.

[0070]

[0071] Furthermore, the diagnostic algorithm 121 can be constructed using a variety of deep learning models such as composite product neural networks, LSTM, RNN, and MLP, and can also be constructed using a variety of machine learning models such as lodgestick regression, principle-based models, random forests, and support vector machines.

[0072] For example, the diagnostic algorithm 121 can be constructed using a deep learning model that utilizes two or more electrocardiogram data points measured at different times with a time difference. Here, two or more electrocardiograms can be integrated into a single electrocardiogram data point and input, or two or more electrocardiograms can be input into a deep learning model separately, and then intermediate semantic features, i.e., spatial and temporal features, can be extracted. Based on these, the two electrocardiograms can be compared to perform measurement, diagnosis, screening, and prediction of the health status at the time of measurement by the single-lead electrocardiogram measurement unit 110 or of future health status.

[0073] Here, a method for mixing electrocardiograms measured at two or more time points is to split a single electrocardiogram bit by bit, then input pairs of bits from the same lead measured at different time points, or to compare the extracted semantic features or result values ​​after input.

[0074] Alternatively, the electrocardiogram data itself can be integrated without dividing the electrocardiogram measured at two or more time points into bits. Here, the electrocardiogram data from two or more time points can be directly merged and input, or it can be separated by lead and merged before input. Alternatively, the electrocardiograms measured at different time points can be input into the deep learning layer of a deep learning model, semantic features or output values ​​can be extracted, and then these can be merged to output the final conclusion. Alternatively, the electrocardiograms at different time points can be separated by lead, input into the deep learning layer, and the features or output values ​​extracted by deep learning can be merged in a later stage to output the final conclusion.

[0075] When using electrocardiograms from two or more time points in time, the input can be asynchronous, synchronized bit by bit, or synchronized and fused based on deep learning.

[0076] In other words, a deep learning model can be constructed using standard lead electrocardiogram data. Here, after developing a deep learning model that measures, diagnoses, screens, and predicts health status by inputting one electrocardiogram at each time point, the deep learning model can be used to input electrocardiograms measured at two or more time points, and the results can be predicted by comprehensively analyzing the semantic features or final output values ​​output from the deep learning model.

[0077] As mentioned above, various deep learning methods such as composite neural networks, LSTM, RNN, and MLP can be used to synthesize the final output values, and various machine learning methods such as lodgestick regression, principle-based models, random forests, and support vector machines can also be used.

[0078]

[0079] As described above, the prediction unit 120 can diagnose and predict diseases of the circulatory system, endocrine, nutritional and metabolic diseases, neoplasms, mental and behavioral disorders, nervous system diseases, eye and adnexal diseases, ear and mastoid process diseases, respiratory system diseases, digestive system diseases, skin and cutaneous tissue diseases, musculoskeletal and connective tissue diseases, urogenital system diseases, pregnancy, childbirth and postpartum diseases, and congenital malformations, deformities and chromosomal abnormalities.

[0080] In addition, the prediction unit 120 can confirm injuries caused by physical trauma, check the prognosis, measure pain levels, predict the risk of death or worsening of injuries, detect or predict concomitant complications, and identify specific pathological conditions that appear in the early or late stages of birth.

[0081] Furthermore, the prediction unit 120 can measure, diagnose, screen, and predict the health status of a person being screened, which can lead to services in the healthcare field such as aging, sleep, weight, blood pressure, blood glucose, oxygen saturation, metabolism, stress, tension, fear, alcohol consumption, smoking, problem behaviors, lung capacity, exercise volume, pain management, obesity, body mass, body composition, menus, exercise types, lifestyle pattern recommendations, emergency situation management, late-onset disease management, drug prescriptions, test recommendations, health check recommendations, nursing care, remote health management, telemedicine, vaccinations, and post-vaccination management.

[0082]

[0083] Figure 2 shows a flowchart illustrating the prediction method using the health status prediction system that utilizes the single-lead electrocardiogram device shown in Figure 1. A detailed explanation based on this flowchart is as follows:

[0084] First, the single-lead electrocardiogram measurement unit 110, which is equipped with two electrodes, measures electrocardiograms of two or more electrical axes at different time intervals by contacting two locations on the patient's body with the two electrodes, acquires asynchronous electrocardiogram data, and transmits it to the prediction unit 120 via short-range communication (S110).

[0085] Subsequently, the prediction unit 120 predicts the health status from the asynchronous electrocardiogram data of the single-lead electrocardiogram measurement unit 110 using a pre-learned diagnostic algorithm 121. The diagnostic algorithm 121, which was previously constructed by learning standard lead electrocardiograms measured on the same electrical axis as the electrocardiogram measured by the single-lead electrocardiogram measurement unit 110, and a standard multi-lead electrocardiogram dataset that matches the presence or absence and severity of diseases corresponding to the standard lead electrocardiograms, predicts the presence or absence and severity of diseases from the asynchronous electrocardiogram data input from the single-lead electrocardiogram measurement unit 110 and understands the health status.

[0086] On the other hand, the electrocardiogram data generation unit 130 generates multiple standard lead electrocardiogram data that are not measured by the single lead electrocardiogram measurement unit 110 and do not match the standard lead electrocardiogram, based on multiple asynchronous electrocardiogram data measured for two or more electrical axes by the single lead electrocardiogram measurement unit 110 (S130). The diagnostic algorithm 121 can take either the asynchronous electrocardiogram data measured by the single lead electrocardiogram measurement unit 110 and the standard lead electrocardiogram data generated by the electrocardiogram data generation unit 130 as input, or the standard lead electrocardiogram data generated by the electrocardiogram data generation unit 130 as input, and output and predict the presence or absence of disease and the degree of disease.

[0087] Subsequently, the warning unit 140 measures a reference electrocardiogram of the subject in their normal, healthy state using the single-lead electrocardiogram measurement unit 110 and the prediction unit 120 to monitor their health status. Then, it compares the electrocardiogram measured and input in real time by the single-lead electrocardiogram measurement unit 110 during daily life with the reference electrocardiogram to predict whether the electrocardiogram was measured without errors or whether there are any abnormalities in the subject's health status. In the event of errors or predicted abnormalities, it generates warning information and can transmit the warning information along with a beep sound via the smartwatch-type single-lead electrocardiogram measurement unit 110 or a separate smart device.

[0088]

[0089] Therefore, with the configuration of the health status prediction system utilizing a single-lead electrocardiogram device as described above, it is possible to easily measure asynchronous electrocardiograms of two or more leads using a single-lead or six-lead electrocardiogram device, predict diseases from the measured asynchronous electrocardiogram data of two or more leads, and generate multi-channel electrocardiogram data from the asynchronous electrocardiogram data of two or more leads, thereby enabling the output and prediction of the presence or absence and severity of diseases with higher accuracy.

[0090]

[0091] The embodiments and configurations shown in the drawings described herein represent only one best embodiment of the present invention and do not represent the entire technical concept of the invention. It should be understood that there are various equivalents and modifications that can be substituted for them at the time of filing this application. [Explanation of Symbols]

[0092] 110: Single-lead electrocardiogram measurement unit 111: Wearable ECG Patch 112: Smartwatch 120: Prediction Department 121: Diagnostic Algorithm 122: Data conversion module 130: ECG data generation unit 140: Warning section

Claims

1. The system includes a prediction unit that predicts the presence and severity of a disease by inputting multiple asynchronous electrocardiogram data measured using a single-lead electrocardiogram device equipped with two electrodes and corresponding to multiple electrical axes into a pre-built diagnostic algorithm. The aforementioned multiple asynchronous electrocardiogram data are acquired at different time points with time intervals between them. The diagnostic algorithm is configured by learning based on a training dataset that includes multiple first standard lead electrocardiogram data points corresponding to multiple electrical axes and measured at different time intervals, and outputs the presence or absence and severity of a disease corresponding to the input data. A health status prediction system utilizing a single-lead electrocardiogram device.

2. The system further includes an electrocardiogram data generation unit configured to generate second standard lead electrocardiogram data corresponding to at least one of the multiple electrical axes that is not measured by the single-lead electrocardiogram device, based on the unique morphological characteristics of the lead electrocardiogram signals corresponding to each of the multiple electrical axes. The prediction unit is configured to take the plurality of asynchronous electrocardiogram data and the plurality of generated second standard lead electrocardiogram data as input to the diagnostic algorithm, and to output and predict the presence or absence of disease and the severity of the disease. A health status prediction system utilizing a single-lead electrocardiogram device as described in claim 1.

3. The single-lead electrocardiogram device is characterized by including a wearable electrocardiogram patch and a smartwatch. A health status prediction system utilizing a single-lead electrocardiogram device as described in claim 1.

4. The prediction unit further includes a data conversion module that generates numerical data corresponding to the plurality of asynchronous electrocardiogram data measured by the single-lead electrocardiogram device, and the diagnostic algorithm is characterized by taking numerical data corresponding to the plurality of asynchronous electrocardiogram data as input. A health status prediction system utilizing a single-lead electrocardiogram device as described in claim 1.

Images