WO2024147506A1

WO2024147506A1 - System for providing health care services to worker on basis of electrocardiogram and method therefor

Info

Publication number: WO2024147506A1
Application number: PCT/KR2023/020375
Authority: WO
Inventors: 권준명
Original assignee: 주식회사 메디컬에이아이
Priority date: 2023-01-03
Filing date: 2023-12-12
Publication date: 2024-07-11

Abstract

The present disclosure relates to a system for providing health care services to a worker on the basis of an electrocardiogram and a method therefor. The system may comprise: at least one electrocardiogram meter that measures an electrocardiogram for a worker and generates electrocardiogram data including electrocardiogram information and worker identification information; and a server that generates electrocardiogram reading result data for managing the worker's diseases or health conditions on the basis of the electrocardiogram data by using a pre-trained first neural network model.

Description

System and method for providing worker health care services based on electrocardiogram

The content of this disclosure relates to a system and method for providing workers' health management services based on electrocardiograms, and specifically relates to a system that can integratedly manage workers' health conditions using electrocardiograms.

Artificial intelligence technology aims for the practical implementation of intelligence through several technological changes, and technology development is in progress through convergence with various fields. Additionally, expectations for artificial intelligence technology are increasing due to the recent emergence of deep learning and the rapid development of ICT technology.

Artificial intelligence technology is developing through integration with technologies such as the Internet of Things, big data, and cloud computing, providing five senses recognition, reflex behavior, and reasoning similar to humans, creating new integrated services, and destructive innovation in all fields. It is expected that it will be accompanied by

Recently, artificial intelligence technology has been intensively applied to fields such as medicine, finance, and culture due to its technical characteristics and impact. However, in the industrial safety field, intelligent information services using artificial intelligence technology have not been attempted to be utilized quickly, so attempts to utilize them quickly are difficult. It is being demanded.

Among tasks at work, there are some tasks in which the employee's health is an important factor. For example, workers working at construction sites are more likely to suffer accidents such as falls if they have heart disease that causes dizziness or fainting. Alternatively, when office workers encounter situations such as long-term meetings, short-term business trips, or frequent job changes that are expected to cause excessive strain, their health may put strain on their heart, which can lead to serious accidents such as death from overwork, such as acute myocardial infarction.

To prevent such accidents, visiting a hospital before starting work and performing an electrocardiogram test using the 12-lead heart measurement method is not only difficult to apply in a general work environment, but can only add to the inconvenience to workers.

Currently, many industrial workers are performing their duties in industrial sites that can directly cause diseases such as stroke and myocardial infarction. However, in order to detect workers' diseases early, medical examinations are required once every two years for office workers and once a year for production workers. However, the development of technology to prevent safety accidents in advance by regularly checking the health status of workers using artificial intelligence technology has not been properly developed.

Looking at the 2019 industrial accident statistics from the Ministry of Employment and Labor, we can see that the mortality rate for patients with cerebrovascular diseases accounts for a much higher proportion of deaths from diseases. The number of cerebrovascular disease deaths among workers at industrial sites is increasing every year, and compared to the past 10 years, it can be seen that it has almost doubled.

If we develop disease risk prediction technology and worker health management technology according to occupational group using artificial intelligence technology in the field of industrial safety, we can carry out effective preventive activities for safety management and worker health management in industrial sites. It is thought that it will be possible to secure the health of workers from various industrial diseases.

However, for industrial safety, only technologies that take into account environmental factors of the workplace have been developed, and the development of technology to determine the health status of industrial workers such as fatigue, concentration, stress, heart abnormality, lung capacity, and carbon dioxide poisoning has not been properly developed. It is not being done.

This disclosure was made in response to the above-described background technology, and measures the worker's electrocardiogram before starting work through a kiosk-type electrocardiogram meter, smart watch, or miniaturized electrocardiogram meter placed in a working environment such as a company or industrial site, and measures the measured electrocardiogram. Based on electrocardiogram data, workers' health status is checked in real time and workers with health problems found are excluded from work, thereby preventing accidents that may occur due to workers' health problems in advance. The purpose is to provide a system that provides management services.

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

According to an embodiment of the present disclosure for realizing the above-described tasks, the present disclosure seeks to provide a system that provides workers' health management services based on electrocardiograms. The system includes at least one electrocardiogram meter that measures an electrocardiogram for a worker and generates electrocardiogram data including the electrocardiogram and worker identification information; And it may include a server that generates ECG reading result data for managing the worker's disease or health condition based on the ECG data, using a pre-trained first neural network model.

Alternatively, the server generates the ECG reading result data using the worker's work information along with the ECG data, and the worker's work information is generated based on user input through the ECG meter, or It can be calculated by the server based on the worker identification information.

Alternatively, the server determines a disease or health condition that is a target for electrocardiogram reading of the worker based on worker work information including at least one of job type, work intensity, or working hours, and the determined disease or health condition. Based on the state, a checklist to be used for analysis of the first neural network model can be created.

Alternatively, if the worker is a field worker, the disease determined based on the worker's work information is heart disease, and if the worker is an office worker, the disease determined based on the worker's work information is stress, depression, Or it could be a mental illness, including anxiety.

Alternatively, the server inputs the electrocardiogram data into the first neural network model, generates a primary reading result regarding the determined disease or health state, and provides a user interface for expert in-depth reading to the expert terminal. , generating a secondary reading result including the expert reading information, wherein the ECG reading result data may include at least one of the first reading result or the secondary reading result.

Alternatively, the first reading result or the second reading result may include a prediction value of at least one of a determination level indicating the worker's health status, whether there is a disease, or a disease probability score related to the disease.

Alternatively, the decision level included in the first read result is classified into a normal level or an abnormal level according to the output of the first neural network model, and the server determines that the decision level included in the first read result is abnormal. In the case of the level, ECG reading result data including first guide information indicating the worker's exclusion from work can be generated.

Alternatively, the determination level included in the secondary reading result is classified into a normal level, abnormal level, or critical level depending on the degree of deterioration of the worker's health condition, and the server determines the level included in the secondary reading result. If the determination level is an abnormal level, electrocardiogram reading result data including second guide information that instructs the worker to stop working and recommends hospital treatment is generated, and the determination level included in the secondary reading result is a risk level. In this case, electrocardiogram reading result data containing third guide information that prohibits the worker from working and directs immediate hospital treatment can be generated.

Alternatively, if the predicted value included in the primary reading result is greater than or equal to a preset threshold, the server may provide a user interface for the expert in-depth reading to the expert terminal to generate a secondary reading result.

Alternatively, if the predicted value included in the current primary reading result differs from the predicted value included in the past primary reading result by more than a predetermined ratio, the server provides a user interface for the expert in-depth reading to the expert terminal Secondary reading results can be generated.

Alternatively, the electrocardiogram measuring device determines the disease or health condition that is the target of the worker's electrocardiogram reading based on the worker's work information including at least one of job type, work intensity, or working hours, and pre-learned Using a second neural network model, a primary reading result for the determined disease or health condition may be generated based on the ECG data.

Alternatively, if the reliability of the first reading result is less than or equal to a preset reference value, the server inputs the ECG data into the first neural network model to generate a secondary reading result related to the determined disease or health condition, and , While the secondary reading result is generated, fourth guide information instructing the worker to wait for work can be generated.

Alternatively, the server determines whether the predicted value included in the primary read result is greater than or equal to a preset threshold, or the predicted value included in the current primary read result differs from the predicted value included in the past primary read result by a predetermined ratio or more. In this case, the user interface for the expert in-depth reading can be provided to the expert terminal to generate a third reading result.

Alternatively, the server may generate health management data including at least one of heart health index information or wellness information for each worker based on the worker's ECG data and ECG reading result data stored for a preset period of time. there is.

Alternatively, the server may provide different numbers of diseases or health conditions that are targets for ECG reading, depending on the level of the product used for the worker's health care service.

Meanwhile, as a method of providing a worker's health care service based on an electrocardiogram, which is performed by a computing device according to an embodiment of the present disclosure and is performed by a computing device including at least one processor, the electrocardiogram measuring device is used by the user. Measuring the electrocardiogram for the worker based on the input and providing electrocardiogram data including the electrocardiogram and worker identification information to the server; And the server uses a pre-trained first neural network model to generate ECG reading result data for managing the worker's disease or health condition based on the ECG data.

A system that provides an electrocardiogram reading service according to an embodiment of the present disclosure measures the worker's electrocardiogram before starting work using an electrocardiogram meter placed in the working environment, and reads the measured electrocardiogram data using a neural network model or in-depth analysis by an expert. By performing readings, you can check the health status of workers in real time, and if health problems such as heart disease or mental illness are discovered, you can prevent accidents that may occur due to health problems by excluding the worker from work. In addition, it has the effect of providing customized health management and wellness information to workers.

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

Figure 2 is a block diagram illustrating the configuration of a system that provides health care services for workers based on electrocardiograms according to an embodiment of the present disclosure.

Figure 3 is a flowchart explaining a method of providing health care services to workers based on electrocardiograms according to the first embodiment of the present disclosure.

Figure 4 is a flowchart explaining a method of providing health care services to workers based on electrocardiograms according to a second embodiment of the present disclosure.

Below, with reference to the attached drawings, embodiments of the present disclosure are described in detail so that those skilled in the art (hereinafter referred to as skilled in the art) can easily practice the present disclosure. The embodiments presented in this disclosure are provided to enable any person skilled in the art to use or practice the subject matter of this 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 can be implemented in various different forms and is not limited to the following embodiments.

The same or similar reference numerals refer to the same or similar elements throughout the specification of this disclosure. Additionally, in order to clearly describe the present disclosure, reference numerals of parts in the drawings that are not related to the description of the present disclosure may be omitted.

As used in this disclosure, the term “or” is intended to mean an inclusive “or” and not an exclusive “or.” That is, unless otherwise specified in the present disclosure or the meaning is not clear from the context, “x uses a or b” should be understood to mean one of natural implicit substitutions. For example, unless otherwise specified in the present disclosure or the meaning is not clear from the context, “x uses a or b” means that x uses a, x uses b, or x uses a and It can be interpreted as one of the cases where both b are used.

The term “and/or” as used in this disclosure should be understood to refer to and include all possible combinations of one or more of the listed related concepts.

The terms “comprise” and/or “comprising” as used in this disclosure should be understood to mean that certain features and/or elements are present. However, the terms “comprise” and/or “including” should be understood as not excluding the presence or addition of one or more other features, other components, and/or combinations thereof.

Unless otherwise specified in this disclosure or the context is clear to indicate a singular form, the singular should generally be construed to include “one or more.”

The term "th nth (n is a natural number)" used in the present disclosure can be understood as an expression used to distinguish the components of the present disclosure according to a predetermined standard such as a functional perspective, a structural perspective, or explanatory convenience. there is. For example, in the present disclosure, components performing different functional roles may be distinguished as first components or second components. However, components that are substantially the same within the technical spirit of the present disclosure but must be distinguished for convenience of explanation may also be distinguished as first components or second components.

The term “acquisition” used in this disclosure is 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. It can be.

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

As used in this disclosure, the term "model" refers to a system implemented using mathematical concepts and language to solve a specific problem, a set of software units to solve a specific problem, or a process to solve a specific problem. It can be understood as an abstract model of a process. For example, a neural network “model” may refer to an overall system implemented as a neural network that has problem-solving capabilities through learning. At this time, the neural network can have problem-solving capabilities by optimizing parameters connecting nodes or neurons through learning. A neural network “model” may include a single neural network or a neural network set in which multiple neural networks are combined.

The explanation of the foregoing terms is intended to aid understanding of the present disclosure. Therefore, if the above-mentioned terms are not explicitly described as limiting the content of the present disclosure, it should be noted that the content of the present disclosure is not used in the sense of limiting the technical idea.

The computing device 100 according to an embodiment of the present disclosure may be a hardware device or part of a hardware device that performs comprehensive processing and calculation of data, or may be a software-based computing environment connected to a communication network. For example, the computing device 100 may be a server 100 that performs intensive data processing functions and shares resources, or it may be a client that shares resources through interaction with the server. Additionally, the computing device 100 may be a cloud system in which a plurality of servers and clients interact to comprehensively process data. Since the above description is only an example related to the type of computing device 100, the type of computing device 100 may be configured in various ways within a range understandable to those skilled in the art based on the contents of the present disclosure.

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

The processor 110 according to an embodiment of the present disclosure may be understood as a structural unit including hardware and/or software for performing computing operations. For example, the processor 110 may read a computer program and perform data processing for machine learning. The processor 110 can process computational processes such as processing input data for machine learning, extracting features for machine learning, and calculating errors based on backpropagation. The processor 110 for performing such data processing includes a central processing unit (CPU), a general purpose graphics processing unit (GPGPU), a tensor processing unit (TPU), and a custom processing unit (TPU). It may include a semiconductor (ASIC: application specific integrated circuit), or a field programmable gate array (FPGA: field programmable gate array). Since the type of processor 110 described above is only an example, the type of processor 110 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 processor 110 can learn a neural network model that diagnoses heart disease based on electrocardiogram data, and can provide worker health management services that can manage the worker's disease or health condition based on the electrocardiogram. For example, the processor 110 may train a neural network model to estimate arrhythmia and other heart diseases based on biological information including information such as gender, age, weight, height, etc., along with electrocardiogram data for each worker. Specifically, the processor 110 may input electrocardiogram data and various biological information into the neural network model and train the neural network model to detect changes in the electrocardiogram due to arrhythmia or other heart diseases. At this time, the neural network model can perform learning based on an ECG dataset that includes features extracted from ECG data and diagnostic data for arrhythmia and other heart diseases. The processor 110 may perform an operation representing at least one neural network block included in the neural network model during the learning process of the neural network model.

The processor 110 may estimate ECG reading result data based on ECG data input by the user using a neural network model generated through the above-described learning process. The processor 110 inputs electrocardiogram data and biological information including information such as gender, age, weight, height, etc. into a neural network model learned through the above-described process, and inferred data representing the result of estimating the probability of heart disease. can be created. For example, the processor 110 can predict the presence and progress of arrhythmia or other heart disease by inputting electrocardiogram data into a trained neural network model.

In addition to the examples described above, the types of medical data and the output of the neural network model 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 according to an embodiment of the present disclosure may be understood as a structural unit including hardware and/or software for storing and managing data processed in the computing device 100. That is, the memory 120 can store any type of data generated or determined by the processor 110 and any type of data received by the network unit 130. For example, the memory 120 may be a flash memory type, hard disk type, multimedia card micro type, card type memory, or random access memory (RAM). ), SRAM (static random access memory), ROM (read-only memory), EEPROM (electrically erasable programmable read-only memory), PROM (prom: programmable read-only memory), magnetic memory , may include at least one type of storage medium among a magnetic disk and an optical disk. Additionally, the memory 120 may include a database system that controls and manages data in a predetermined system. Since the type of memory 120 described above is only an example, the type of 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 manage data required for the processor 110 to perform operations, a combination of data, and program code executable on the processor 110 by structuring and organizing them. For example, the memory 120 may store ECG data received through the network unit 130, which will be described later. The memory 120 includes program code that operates the neural network model to receive medical data and perform learning, program code that operates the neural network model to receive medical data and perform inference according to the purpose of use of the computing device 100, and Processed data generated as the program code is executed can be saved.

The network unit 130 according to an embodiment of the present disclosure may be understood as a structural unit that transmits and receives data through any type of known wired or wireless communication system. For example, the network unit 130 may be connected to a local area network (LAN), wideband code division multiple access (WCDMA), long term evolution (LTE), or wireless (WIBRO). broadband internet, 5th generation mobile communication (5g), ultra wide-band wireless communication, zigbee, radio frequency (RF) communication, wireless LAN, wireless fidelity ), data transmission and reception can be performed using a wired or wireless communication system such as near field communication (NFC), or Bluetooth. Since the above-described communication systems are only examples, the wired and wireless communication systems for data transmission and reception of the network unit 130 may be applied in various ways other than the above-described examples.

The network unit 130 may receive data necessary for the processor 110 to perform operations through wired or wireless communication with any system or client. Additionally, the network unit 130 may transmit data generated through the calculation of the processor 110 through wired or wireless communication with any system or any client. For example, the network unit 130 may receive medical data through communication with a kiosk-type electrocardiogram meter, a smart watch, or a miniaturized electrocardiogram meter. The network unit 130 may transmit output data of the neural network model, intermediate data derived from the calculation process of the processor 110, processed data, etc. through communication with the above-mentioned devices.

A system that provides workers' health management services based on electrocardiograms includes, but is not limited to, at least one electrocardiogram meter 200, a server 100, and an expert terminal 300.

At least one electrocardiogram measuring device 200 measures the worker's electrocardiogram, generates electrocardiogram data including the electrocardiogram and worker identification information, and then transmits the generated electrocardiogram data to the server 100.

The electrocardiogram measuring device 200 can use various devices capable of measuring electrocardiograms, such as wearable devices that are worn on the user's body and can measure and collect various health indicators such as heart rate, body fat percentage, and blood pressure, and electrocardiogram kiosks.

At this time, the electrocardiogram measuring device 200 can measure the electrocardiogram using various electrode combinations, such as a 12-guide method and a 6-guide method, as well as a single-guide method using a wearable device such as a wristwatch or a patch. It is desirable that the ECG measurement time is also set by adding or subtracting depending on the signal to be obtained.

The server 100 uses a pre-trained first neural network model to generate ECG reading result data for managing the worker's disease or health condition based on the ECG data and the worker's work information. At this time, the worker identification information may include a unique identification number such as the worker's employee number, ID, and email address, and the server 100 identifies the worker through information such as work records and department name based on the unique identification number. Workers' work information such as job type, work intensity, and working hours can be calculated. Meanwhile, the worker's work information may be generated based on user input through the electrocardiogram monitor 200. That is, the worker may input worker identification information into the electrocardiogram measuring device 200 and separately input work information such as the worker's job type, the intensity of the job felt by the worker, and the worker's average working hours.

The server 100 may link ECG data including worker identification information and ECG reading result data and store them in the database 150. In addition, the server 100 can cumulatively store health management data, including electrocardiogram reading result data, using the database 150 to periodically provide a cerebrovascular disease risk report for workers, and cumulatively store the worker's wellness information. Therefore, it can be provided as management and basis data in the event of a major disaster.

Here, the first neural network model learns ECG data for each feature using a deep learning algorithm, and uses the learned model to derive ECG reading information, including the diagnosis of heart disease. Additionally, the first neural network model may be learned based on the correlation between left ventricular systolic dysfunction and changes in characteristics such as electrocardiogram, gender, age, weight, and height. Specifically, the neural network model may be learned based on a learning dataset including electrocardiogram and heart disease diagnosis results and the correlation between various factors in the learning dataset.

The first neural network model may be learned based on the electrocardiogram measured with 12 leads obtained from electrodes of the electrocardiogram measuring device 200 connected to the human body. For example, an electrocardiogram can be measured with 12 leads of 10 seconds in length and stored at 500 points per second. Additionally, the neural network model can be learned based on partial information extracted from only 6 limb lead ECGs and a single lead (lead I) ECG among the 12 lead ECGs.

Specifically, the first neural network model can receive ECG data as input and output ECG reading information, and includes at least one convolutional neural network (CNN), batch normalization, and ReLU activation function layer. and may include a dropout layer. The neural network model may include a fully connected layer in which biological information such as age, gender, height, and weight is input as auxiliary information.

The first neural network model may include a neural network corresponding to each of a plurality of leads of ECG data. That is, the first neural network model may include an individual neural network into which electrocardiograms measured through individual leads are input.

Meanwhile, since the structure and type of neural network of the above-described first neural network model are only examples, the neural network model according to an embodiment of the present invention may be configured in various ways based on the above-described examples.

Meanwhile, the server 100 determines the disease or health condition that is the target of the worker's electrocardiogram reading based on the worker's work information including at least one of job type such as field work/office work, work intensity, or work hours, and determines the determined A checklist to be used in the analysis of the first neural network model can be created based on the disease or health condition.

In other words, if the server 100 confirms that the worker is a field worker through the worker identification information, the disease determined based on the worker's work information may be heart disease, so it creates a checklist with checkup items related to heart disease. can be created. In addition, the server 100 determines that if a worker is identified as an office worker through worker identification information, the disease determined based on the worker's work information may be a mental disease including stress, depression, or anxiety. You can create a checklist with checkup items related to .

The server 100 inputs electrocardiogram data into the first neural network model, generates a primary reading result regarding the disease or health condition determined based on the worker's work information, and provides a user interface for expert in-depth reading through the expert terminal 300. ) is provided.

Additionally, the server 100 may provide different numbers of diseases or health conditions that are targets for ECG reading depending on the level of the product used for the worker's health care service. At this time, the server 100 divides the product grades into premium, economy, and basic, and as the grade of the product goes from lower to higher, the price of use and the number of diseases or health conditions that are targets for electrocardiogram reading increase. You can set it.

For example, for workers using premium products, which are the highest and highest price levels, the server 100 can provide analysis services for 24 diseases or health conditions, and for workers using economy products, which are the lowest and lowest price levels. For workers, the server 100 can provide analysis services for 10 diseases and health conditions.

The server 100 may modify and change the level of the product in various ways as the number of companies or workers using workers' health care services increases/decreases or the service environment changes.

The expert terminal 300 may be an electrocardiogram reading center capable of collaborating with external experts and providing expert in-depth reading services, or a terminal providing diagnostic services by medical experts. This expert terminal 300 generates expert reading information based on in-depth analysis of ECG data and provides it to the server 100.

Accordingly, the server 100 may generate ECG reading result data including at least one of the first reading result by the first neural network model or the secondary reading result including expert reading information by the expert terminal 300. .

Referring to FIG. 3, before the worker starts work, ECG data generated using the ECG meter 200 is transmitted to the server 100 (S311), and the server 100 uses the pre-trained first neural network model. This generates a primary reading result on the worker's disease or health condition (S312). In addition, the server 100 receives the expert reading information through the expert terminal 300 and generates a secondary reading result including the expert reading information (S316).

Since the ECG analysis process using the first neural network model can output ECG reading information at a faster reading speed than the ECG analysis process using the expert terminal 300, the server 100 outputs the first reading result using the first neural network model. After providing first, the secondary reading result through the expert terminal 300 can be provided later. Alternatively, the server 100 may integrate the first read results with the second read results and provide them.

Meanwhile, the server 100 determines the disease or health condition of the worker that is the target of ECG reading based on the worker's work information including at least one of job type, work intensity, or working hours, and determines the worker's disease or health condition thus determined. Based on the state, a checklist to be used in the analysis of the first neural network model can be created.

For example, if the worker is a field worker, the server 100 focuses on heart disease that can lead to a serious accident if it occurs suddenly, and the disease determined based on the worker's work information is heart disease. It can be. Conversely, when the server 100 is an office worker, the disease determined based on the worker's work information may be a mental disease including stress, depression, or anxiety. In this way, the server 100 can read the electrocardiogram by targeting diseases or health-related problems frequently identified in specific jobs, and through this, can efficiently process a large amount of data.

At this time, the first neural network model can determine the degree of balance of the autonomic nervous system through time or frequency analysis of heart rate variation (HRV) obtained through an electrocardiogram, and determines stress-related factors through the balance and activity of the autonomic nervous system. Algorithms that can predict and evaluate the risk of developing a disease can be used. At this time, stress-related diseases include nervous system diseases such as tension headache, migraine, and stroke, cardiovascular diseases such as myocardial infarction, angina, arrhythmia, tachycardia, and bradycardia, mental diseases such as depression, anxiety, insomnia, and panic disorder, functional gastrointestinal disorders, and irritable bowel. This may include syndrome, digestive system problems such as decreased appetite, diabetes, thyroid disease, obesity, fatigue, and decreased energy.

The server 100 may include in the first or second reading results a prediction value of at least one of a determination level indicating the worker's health status, presence of a disease, or a disease probability score related to the disease.

The server 100 simultaneously performs the ECG reading process by the first neural network model and the ECG reading process by the expert terminal, and displays the first reading result by the first neural network model and the secondary reading result by the expert terminal at regular time intervals. By providing this, you can increase the reliability of the ECG analysis results.

Meanwhile, the server 100 performs an ECG reading process using the first neural network model, but can simultaneously perform an ECG reading process using an expert terminal only under certain conditions. That is, the server 100 can generate a secondary reading result through the expert terminal 300 only when the predicted value included in the primary reading result is greater than or equal to a preset threshold. For example, if the prediction value of the first neural network model predicts the probability score of cardiac arrest within 24 hours above a preset threshold, the prediction value of the first neural network model predicts the probability score of occurrence of a specific disease above the preset threshold value. In this case, when the first neural network model predicts that the first neural network model has a disease for one or more diseases among multiple diseases, or when the first neural network model has an index value indicating health for mental diseases such as stress, depression, or anxiety, a specific criterion is specified. If it is predicted to fall short of , a secondary reading result including expert reading information through expert (or medical staff) analysis can be additionally performed.

Meanwhile, the server 100 provides a secondary reading information including expert reading information through expert (or medical staff) analysis only when the predicted value included in the current primary reading result differs by more than a predetermined percentage from the predicted value included in the past primary reading result. The reading results can be further processed.

At this time, the judgment level included in the first reading result is divided into a normal level indicating that the worker's health is 'normal' or an abnormal level indicating that the worker's health is outside the normal range, in accordance with the output of the first neural network model. The judgment level included in the secondary reading result can be classified as a normal level, abnormal level, or dangerous level depending on the degree of deterioration of the worker's health condition. At this time, the risk level may indicate an 'emergency' level where the worker's health condition requires immediate treatment.

The server 100 can determine whether to continue with the task assigned to the worker, replace it with another task, or recommend the worker to receive medical treatment or treatment according to this judgment level. In addition, the server 100 can inform the worker of the decided matters and manage in real time whether the worker maintains an appropriate health condition according to the job. In addition, the server 100 can provide a working environment in which workers can prevent health problems that may occur while performing work and take quick action by guiding the workers to the decision.

Specifically, when the determination level of the first reading result is a normal level (see S313), the server 100 generates ECG reading result data including the worker's work availability status (S314).

However, when the determination level of the first reading result is an abnormal level, the server 100 generates ECG reading result data including first guide information indicating the worker's exclusion from work (S315).

When the server 100 receives expert reading information from the expert terminal 300, it generates a secondary reading result including the expert reading information (S316). At this time, if the determination level included in the secondary reading result is a normal level (S317), the server 100 confirms the ECG reading result data including the worker's work availability status (see S314).

However, if the determination level included in the secondary reading result is an abnormal level (S318), the server 100 generates ECG reading result data including second guide information that instructs the worker to prohibit work and recommends hospital treatment. Do it (S319).

If the judgment level included in the secondary reading result is a risk level (S321), the server 100 generates ECG reading result data including third guide information prohibiting the worker from working and instructing immediate hospital treatment (S321 ).

The electrocardiogram measuring device 200 not only performs electrocardiogram measurement, but can also embed a second neural network model implemented in an on-device artificial intelligence method. In the second embodiment of the present disclosure, the health of workers is maintained using the electrocardiogram meter 200, server 100, or expert terminal 300 with a built-in second neural network model implemented in an on-device artificial intelligence method. Provides a method of providing management services.

Referring to FIG. 4, the electrocardiogram measuring device 200 determines the disease or health condition that is the target of the worker's electrocardiogram reading based on the worker's work information including at least one of job type, work intensity, or working hours (S411 ), using a pre-trained second neural network model, a primary reading result for a disease or health condition determined based on ECG data can be generated (S412).

That is, before generating a reading result using the first neural network model in the server 100, the first reading result may be generated using the second neural network model of the ECG meter 200. At this time, the second neural network model is a lighter model than the first neural network model, and may be a model that performs targeted inference related to the worker's job based on some indicators present in the electrocardiogram. For example, if the electrocardiogram monitor 200 is a kiosk-type device set up in a field work environment, the second neural network model may be a model that targets and infers heart diseases frequently identified in field workers. In this way, when the second neural network model is learned to target a specific disease or health condition, the computing resources required by the ECG meter 200 are sufficiently reduced compared to the server 100, so that a reading can be performed with fewer resources. can be built.

If the reliability of the first reading result using the second neural network model is below the preset standard value (S413), the server 100 inputs the electrocardiogram data into the first neural network model to determine the disease or health status based on the worker's work information. It can be generated as a result of a secondary reading (S414, S416). At this time, the server 100 may provide fourth guide information instructing the worker to wait for work while the secondary reading result is generated (S415).

Meanwhile, the server 100 detects if the predicted value included in the primary reading result is greater than or equal to a preset threshold or if the predicted value included in the current primary reading result differs from the predicted value included in the past primary reading result by a predetermined ratio or more. (S417), a user interface for expert in-depth reading may be provided to the expert terminal 300 to generate a third reading result (S418).

The server 100 may generate ECG reading result data using at least one of the first reading result, the second reading result, or the third reading result. As described in the first embodiment above, the server 100 determines the decision level (normal/abnormal) included in the first or second read result, and the decision level (normal/abnormal/critical) included in the third read result. ) may include first guide information, second guide information, or third guide information.

In the first and second embodiments described above, the server 100 provides at least one of heart health index information and wellness information for each worker based on the worker's ECG data and ECG reading result data stored for a preset period of time. Health management data including can be generated and stored in the database 150. At this time, the wellness information may include content for at least one of the physical domain, emotional domain, social domain, cognitive domain, spiritual domain, or occupational domain. For example, wellness information in the physical domain may include customized exercise prescriptions for workers, diet and health food recommendation services, and wellness information in the emotional domain may include music and YouTube that can relieve stress and reduce depression or anxiety. , may include mental health care services such as therapy.

The present disclosure can produce highly reliable ECG analysis results by performing artificial intelligence-based ECG reading and medical staff reading in parallel in several steps. In addition, through this disclosure, workers can easily monitor their health status or work progress by freely measuring their electrocardiogram before starting work or at any time and place of their choice. In addition, through this launch, employers can conveniently monitor the health status of workers, allowing them to preemptively respond to health issues that may arise for each job in the working environment. In addition, through this disclosure, hospitals can induce workers to visit the hospital as needed, creating a medical environment in which disease outbreaks and responses can be proactively performed.

The various embodiments of the present disclosure described above may be combined with additional embodiments and may be changed within the scope understandable to those skilled in the art in light of the above detailed description. The embodiments of the present disclosure should be understood in all respects as illustrative and not restrictive. For example, each component described as unitary may be implemented in a distributed manner, and similarly, components described as distributed may also be implemented in a combined form. Accordingly, all changes or modified forms derived from the meaning and scope of the claims of the present disclosure and their equivalent concepts should be construed as being included in the scope of the present disclosure.

Claims

A system that provides health care services to workers based on electrocardiograms,

At least one electrocardiogram measuring device that measures the electrocardiogram of the worker and generates electrocardiogram data including the electrocardiogram and worker identification information; and

Comprising a server that generates ECG reading result data for managing the worker's disease or health condition based on the ECG data, using a pre-trained first neural network model,

system.
According to paragraph 1,

The server is,

Generating the ECG reading result data using the worker's work information along with the ECG data,

The work information of the above worker is:

Generated based on user input through the electrocardiogram meter,

Calculated by the server based on the worker identification information,

system.
According to paragraph 2,

The server is,

Based on the worker's work information including at least one of job type, work intensity, or working hours, determine the disease or health condition that is the target of the worker's electrocardiogram reading,

Based on the determined disease or health condition, generating a checklist to be used in the analysis of the first neural network model,

system.
According to paragraph 3,

If the above worker is a field worker,

The disease determined based on the worker's work information is heart disease,

If the above worker is an office worker,

The disease determined based on the worker's work information is a mental disease including stress, depression, or anxiety,

system.
According to paragraph 3,

The server is,

Input the electrocardiogram data into the first neural network model to generate a primary reading result regarding the determined disease or health condition,

A user interface for expert in-depth reading is provided to the expert terminal to generate a secondary reading result including the expert reading information,

The electrocardiogram reading result data is,

Containing at least one of the first read result or the second read result,

system.
According to clause 5,

The first read result or the second read result is,

Containing at least one predictive value among the judgment level indicating the health status of the worker, whether there is a disease, or a disease probability score related to the disease,

system.
According to clause 6,

The decision level included in the first reading result is,

Divided into normal level or abnormal level according to the output of the first neural network model,

The server is,

If the determination level included in the first reading result is an abnormal level,

Generating electrocardiogram reading result data including first guide information instructing the worker's exclusion from work,

system.
According to clause 6,

The decision level included in the secondary reading result is,

Depending on the degree of deterioration of the worker's health, it is classified into a normal level, abnormal level, or dangerous level,

The server is,

If the determination level included in the secondary reading result is an abnormal level,

Generating electrocardiogram reading result data including second guide information instructing the worker to refrain from working and recommending hospital treatment,

If the judgment level included in the secondary reading result is a risk level,

Generating electrocardiogram reading result data containing third guide information directing the worker's work prohibition and immediate hospital treatment,

system.
According to clause 6,

The server is,

If the predicted value included in the first reading result is greater than or equal to a preset threshold,

Providing a user interface for the expert in-depth reading to the expert terminal to generate secondary reading results,

system.
According to clause 6,

The server is,

If the predicted value included in the current primary reading result differs from the predicted value included in the past primary reading result by more than a certain percentage,

Providing a user interface for the expert in-depth reading to the expert terminal to generate secondary reading results,

system.
According to paragraph 2,

The electrocardiogram meter,

Based on the worker's work information including at least one of job type, work intensity, or working hours, determine the disease or health condition that is the target of the worker's electrocardiogram reading,

Generating a primary reading result for the determined disease or health condition based on the electrocardiogram data, using a pre-trained second neural network model,

system.
According to clause 11,

The server is,

If the reliability of the first reading result is below a preset standard value,

Inputting the electrocardiogram data into the first neural network model to generate a secondary reading result regarding the determined disease or health condition,

While the secondary reading result is generated, generating fourth guide information instructing the worker to wait for work,

system.
According to clause 12,

The server is,

If the predicted value included in the primary reading result is greater than or equal to a preset threshold, or if the predicted value included in the current primary reading result differs from the predicted value included in the past primary reading result by more than a predetermined ratio,

Providing a user interface for the expert in-depth reading to an expert terminal to generate a third reading result,

system.
According to paragraph 1,

The server is,

Generating health management data including at least one of heart health index information or wellness information for each worker based on the worker's ECG data and ECG reading result data stored for a preset period,

system.
According to paragraph 1,

The server is,

Depending on the level of the product used for the worker's health care service, the number of diseases or health diseases targeted by electrocardiogram reading is provided differently.

system.
As a method of providing health care services to workers based on electrocardiogram,

The electrocardiogram measuring device measures the electrocardiogram of the worker based on user input, and provides electrocardiogram data including the electrocardiogram and worker identification information to the server; and

The server uses a pre-trained first neural network model to generate ECG reading result data for managing the worker's disease or health condition based on the ECG data;

Including,

method.