KR20190117829A - System for monitoring medical data - Google Patents

Abstract

According to an embodiment of the present invention, a system for monitoring medical data not only can predict various abnormal symptoms of the body but also can obtain data analysis and artificial intelligence learning data in distributed surroundings using a blockchain network.

Description

Medical data monitoring system {SYSTEM FOR MONITORING MEDICAL DATA}

The present invention relates to a medical data monitoring system. Specifically, the medical data is implemented to not only predict abnormal signs of various bodies, but also to acquire data analysis and artificial intelligence learning data in a distributed environment using a blockchain network. It relates to a monitoring system.

The platforms served in the biometric data measurement market using wearable devices are diverse and come in many forms. However, these various platforms have the following limitations in common.

1. Wearable devices only perform measurement of the biosignal.

2. When analyzing data, only the data measured or collected by the device is used as an analysis factor.

3. Physical limitations exist for accumulating large amounts of data measured from the body.

4. Measurement of biometric data using artificial intelligence algorithms already implemented internally in the device, but it cannot perform a series of processes such as learning artificial intelligence with measured data once.

5. The AI algorithm inside the hardware device cannot learn / modify without updating the firmware physically or manually.

6. General AI learning uses static big data.

7. General AI learns from data collected on a centralized infrastructure.

8. The way in which medical data can be obtained by unspecified individuals or groups in the medical market is very narrow and entry barriers are very high.

 The above-mentioned problems are not only problems of wearable devices but also limitations of the system of intrinsic hardware. Therefore, there is a demand for a method for overcoming and solving the above-mentioned problems in terms of software.

Republic of Korea Patent Publication No. 10-0737382

Medical data monitoring system according to an embodiment of the present invention aims to solve the above problems.

It is possible to suggest a concrete way to overcome the storage of data, which is a physical limitation, and to monitor medical data combined with artificial intelligence as an application for predicting various abnormal symptoms that may occur in the future, not just artificial analysis. The purpose is to provide a system.

Furthermore, by applying the blockchain network to the medical data monitoring system, it is possible to transmit and receive data safely even when the network in use is unstable, and provide medical data that can provide flawless and reliable medical data to the research institute, etc. without tampering. Its purpose is to provide a monitoring system.

The problems of the present invention are not limited to those mentioned above, and other problems not mentioned above will be clearly understood by those skilled in the art from the following description.

Medical data monitoring system according to an embodiment of the present invention includes a first gateway for receiving in real time the first data including the biometric information of the first user and the environment information; A preprocessor configured to preprocess the first data received from the first gateway to generate second data; Calculate and continuously update the correlation between the biometric information, the environmental information, and the health state of the first user by storing the second data and performing machine learning using the accumulated second data as an input variable. A machine learning module for generating third data which is current or future health state information of the first user based on the second data currently input and the correlation; And a second gateway configured to receive second data and third data from the preprocessor and the machine learning unit, respectively, and the health report of the first user generated based on the second data and the third data. In order to generate and output information to a second user, the second gateway transmits the second data and the third data to a second terminal, and the biometric information of the first user is mounted on the body of the first user. And the environment information is set by a first terminal receiving the biometric information in association with the wearable device or obtained from the wearable device.

The machine learning module may include: a third gateway receiving the second data in real time from the preprocessor; A first storage unit accumulating and storing second data received in real time from the third gateway to generate big data; A machine learning unit configured to derive a correlation between the biometric information, the environment information, and the health state of the first user by performing machine learning using the second data accumulated in the first storage unit as an input variable; A determination unit to determine a health state of the first user or to predict a future health state of the first user based on the correlation derived by the machine learning unit; And a health state notification unit receiving the third data which is a determination result of the determination unit and transmitting the received third data to the second gateway.

The machine learning module may further include a second storage unit accumulating and storing the third data.

The firmware installed in the wearable device or the first terminal may be updated by an algorithm that reflects the correlation generated and updated in the machine learning module.

The wearable device, the first terminal, the first gateway, a preprocessor, a machine learning module, a second gateway and a blockchain network formed to connect at least two or more of the second terminal, the wearable device, the first terminal, When an error occurs in a network environment (internet, http) for communication of at least two of a first gateway, a preprocessor, a machine learning module, a second gateway, and a second terminal, the communication is performed through the blockchain network. It is desirable to allow the process to proceed.

Preferably, at least one of the first data, the second data and the third data of the first user may be transmitted through the blockchain network after obtaining the consent of the first user.

And a third user server capable of acquiring at least one of first data, second data, and third data of the first user through the blockchain network, wherein the third user is the first user. Or acquire at least one of the first data, the second data, and the third data based on the completion of a predetermined contract (programmed with a smart contract) with the blockchain network platform builder, and the details of the completed contract. Is preferably recorded on the blockchain network.

It is preferable to further include; a third storage unit for storing the first data.

When the medical data monitoring system according to an embodiment of the present invention is introduced, the medical service can be provided to the user, and since all data are managed by the cloud server, the user can maintain and manage his / her physical data. It has the effect of being able to have the authority to do so.

In addition, the medical data monitoring system according to an embodiment of the present invention is built to provide a transparent and valuable medical data market by block-chaining individual medical data to a user or a research institute through a P2P transaction. It can work.

The effects of the present invention are not limited to those mentioned above, and other effects that are not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a schematic diagram of a medical data monitoring system according to an embodiment of the present invention.
2 is a diagram for explaining data transmission and reception using a blockchain network in a medical data monitoring system according to an embodiment of the present invention.

DETAILED DESCRIPTION OF EMBODIMENTS Preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, and the same or similar components will be given the same reference numerals regardless of the reference numerals and redundant description thereof will be omitted.

In addition, in describing the present invention, when it is determined that the detailed description of the related known technology may obscure the gist of the present invention, the detailed description thereof will be omitted. In addition, it should be noted that the accompanying drawings are only for easily understanding the spirit of the present invention and should not be construed as limiting the spirit of the present invention by the accompanying drawings.

Hereinafter, a medical data monitoring system according to an embodiment of the present invention will be described with reference to FIGS. 1 and 2.

Medical data monitoring system according to an embodiment of the present invention, as shown in Figure 1 wearable device 110, the first terminal 120, the first gateway 200, the preprocessor 300, the machine learning module ( 400), a second gateway 500, and a second terminal 600.

The wearable device 110 is mounted on the body of the first user A, that is, a user who needs to obtain biometric information. For example, the wearable device 110 may include a wrist, a chest (heart), a finger, ankle, and the like of the first user A. It is preferably configured to be mounted on the shoulder or the like.

In this case, the user may be a human, but may be an animal such as a horse, a dog, or a cow.

The biometric information of the first user A detected by the wearable device 110 may include heart rate data, heart rate, body temperature, skin conductivity, movement, and electrocardiogram of the first user A. FIG.

The first terminal 120 is configured to communicate with the wearable device 110 and is connected to the wearable device 110 to receive biometric information from the wearable device 110 in real time.

Here, the first terminal 120 may be a smart phone, a tablet PC, etc. that the first user A carries, or may be a desktop PC or a laptop PC.

In particular, the first terminal 120 performs a function of acquiring the surrounding environment information of the first user A, wherein the environment information is the location of the first user A, which is country, weather, location, temperature, Humidity, fine dust concentration, altitude, carbon dioxide concentration, and the like.

The first gateway 200 receives the first data including the biometric information and the environmental information of the first user A from the first terminal 120 in real time and transmits the data to the back-end. do.

That is, the medical data monitoring system according to an embodiment of the present invention takes into consideration not only the biometric information of the first user A, but also the current environment information of the first user A in consideration of the current state of the first user A. You will be able to analyze and predict your physical situation more accurately.

Meanwhile, while the wearable device 110 obtains the biometric information and the first terminal 120 obtains the environmental information, the wearable device 110 may acquire both the biometric information and the environmental information. In this case, the wearable device 110 may directly transmit the biometric information and the environmental information to the first gate 200 without having to provide a separate first terminal 120.

The preprocessor 300 performs a function of preprocessing the first data received from the first gateway 200 to generate second data.

As described above, since the first data corresponds to a kind of raw data that is not yet refined as data including biometric information and environmental information of the first user A, it is difficult to derive a meaningful result using only the first data.

Therefore, in order to perform a meaningful analysis using the corresponding data, the preprocessor 300 generates the second data by performing preprocessing of the first data, and the machine learning module 400 or the second gateway 500 to be described later. To pass.

On the other hand, the first data is the data before being purified, but considering that there is a need to be backed up in order to examine the error of the data in the future, the medical data monitoring system according to an embodiment of the present invention in real time to the first data It is preferably configured to further include a third storage unit 900 for storing.

The machine learning module 400 stores the second data after receiving the second data from the preprocessor 300, and performs machine learning using the second data as an input variable.

That is, the machine learning module 400 calculates and continuously updates the correlation between the biometric information and the environmental information of the first user A and the health state of the first user A, and displays the second data and Based on the correlation, third data, which is current or future health state information of the first user A, is generated, which will be described later.

The second gateway 500 receives the second data or the third data from at least one of the preprocessor 300 and the machine learning unit 400 and delivers the second data or the third data to the front-end side of the second user B side. It performs the function.

The second terminal 600 generates and outputs health report information for the first user A based on at least one of the second data and the third data received from the second gateway 500, and thus the second user B. Perform the function provided to).

Here, the second terminal 600 may be a smart phone or a tablet PC carried by the second user B, or a laptop PC or a desktop PC of the second user B.

That is, the second terminal 600 is a second user (application) of the report format that can grasp the information related to the health state of the first user (A) at a glance based on the data received from the second gateway 500. B), for example, to the attending physician or family of the first user A.

In addition, in the case of an emergency, in consideration of the urgent need of the second user B to urgently notify the current state of the first user A, whether or not the above-mentioned health report information has been arrived or summarized is provided. It is preferable to notify the second user B through SMS, MMS) or mail delivery.

Hereinafter, the machine learning module 400 included in the medical data monitoring system according to an embodiment of the present invention will be described in more detail.

As shown in FIG. 1, the machine learning module 400 includes a third gateway 410, a first storage unit 420, a machine learning unit 430, a determination unit 440, a health state notification unit 450, and It may be configured to include a second storage unit 460.

The third gateway 410 performs a function of receiving the second data in real time from the preprocessor 300.

The first storage unit 420 accumulates and stores data received in real time from the third gateway 410 to generate big data.

That is, as the second data is stored in the first storage unit 420, the amount of data stored in the first storage unit 420 becomes larger, resulting in the biometric information and the environmental information of the first user A. Big data can be obtained.

The machine learning unit 430 performs machine learning using the second data accumulated in the first storage unit 420 as an input variable to perform biometric and environmental information of the first user A, and the first user A. Its function is to derive correlations between health conditions.

Herein, the health state of the first user A may include, for example, the possibility of arrhythmia disease, and the types of such arrhythmia diseases include, specifically, no contraction (deficit contraction), ventricular fibrillation, and ventricular extraventricular contraction (early ventricular Contraction), ventricular tachycardia, biphasic vein (or delayed pulse-a symptom of rapidly developing two-pulse pulses), tricuspid vein (or tripulsatile pulse), east and west veins, atrial fibrillation, atrial fibrillation, and ventricular tachycardia (ventricular stomach) Fast mac), and the like.

That is, the machine learning unit 430 learns artificial intelligence using big data stored in the first storage unit as an input variable. Specifically, the machine learning unit 430 uses a deep learning technique, a field of machine learning, to accurately correlate the artificial intelligence. Do the learning so that can be derived.

In particular, in the correlation described above, the input may be defined as a biometric information and environmental information of the first user A, and the output may be defined as a function composed of the health state of the first user A. After all, the machine learning unit 430 calculates the weights of the plurality of inputs in the function through deep learning through learning.

Learning performed in the machine learning unit 430 may be performed by supervised learning or unsupervised learning, but in the case of a medical data monitoring system according to an embodiment of the present invention, It is preferable to carry out.

In addition, as an artificial intelligence network model used for such learning, various models such as Recurrent Neural Network (RNN), Deep Neural Network (DNN), and Dynamic Recurrent Neural Network (DRNN) may be used.

The determination unit 440 may determine a current health state of the first user A or predict a future health state of the first user A based on the correlation derived by the machine learning unit 430. Perform.

In particular, the determination unit 440 is not substantially separated from the machine learning unit 430, and is a neural network in which deep learning is performed in the machine learning unit 430.

As a result, the functions of the machine learning unit 430 and the determination unit 440 can be largely divided into two types. First, the second data received from the preprocessor 300 in real time is analyzed and compared to be generated later. The biometric data of the first user A may be predicted, and secondly, the precision of the correlation may be further improved by utilizing the second data received in real time from the preprocessor 300 as data for learning.

In other words, as long as the second data is continuously secured, comparison and learning are continuously performed, and thus, the prediction accuracy of artificial intelligence is increased and refined.

The health state notification unit 450 receives the third data, which is the determination result of the determination unit 440, from the determination unit 440 and transmits the third data to the second gateway 500.

For example, when the determination unit 440 inputs the second data of the first user A to the correlation derived from the machine learning unit 430, the heart abnormality of the first user A, such as arrhythmia, is found. If it is determined that the symptom is predicted, the health state notification unit 450 transmits the third data, which is the result of the determination, to the second gateway 500 to the second user B such as the attending physician. You will be informed of signs of abnormalities.

In addition, since the above-described third data is also preferably stored separately from the information and the history, the machine learning module 400 of the medical data monitoring system according to an embodiment of the present invention may accumulate and store the third data. It would be further desirable to have a reservoir 460.

Furthermore, the medical data monitoring system according to an embodiment of the present invention is an artificial intelligence installed in the above-described wearable device 110 or the first terminal 120 using an algorithm generated by learning in the machined module 400. You can also update the algorithm.

That is, the firmware of the wearable device 110 or the first terminal 120 is updated. As the body data of the user increases and becomes more sophisticated, the accuracy and efficiency of detecting or measuring the biometric data or the environmental data of the user is increased. It can be increased.

Meanwhile, in the medical data monitoring system according to an embodiment of the present invention, the wearable device 110, the first terminal 120, the first gateway 200, and the preprocessor 300 are illustrated in FIGS. 1 and 2. The apparatus may further include a blockchain network 700 formed to connect at least two of the machine learning module 400, the second gateway 500, and the second terminal 600, and thereby, the first user A. Medical data will be blockchained to enhance ownership of medical data and to build a P2P and P2B trading platform for medical data.

The blockchaining of the medical data of the first user A proceeds only with the prior consent of the first user A, and the medical data of the first user A, which is a block by first blockchaining, becomes a block. Node data is used to construct and maintain a chain network, and the first node is called a genesis node.

Each such node data contains purified data, that is, second or third data, collected or measured during a predetermined reference time in the blockchain platform, and the minimum unit is called a block.

Medical data collected or measured later is continuously blocked, and gradually added within the node to which it belongs.

As a rule, the medical data split in units of blocks based on a predetermined criterion as described above is granted only to the first user A who is a user who provides the information.

However, in the medical data monitoring system according to an exemplary embodiment of the present invention, the medical data of the first user A may be provided to medical companies such as research institutes, hospitals, and companies, and the institutions or companies may utilize medical data of unspecified individuals. It may be configured to perform research and development.

To this end, the medical data monitoring system according to an embodiment of the present invention is configured to acquire at least one of the first data, the second data, and the third data of the first user A through the blockchain network 700. A third user's server 800 may be further included, wherein the third user means the research institute, hospital, or company described above.

At this time, the block-based block set in advance in the blockchain network 700 is treated as one unit, and research institutes or medical companies pay a block or network of goods to the blockchain network platform builder for each block. 1 may be configured to obtain the medical data of the user (A).

The amount or goods paid by the research institute or medical company is returned to the individual who provided the medical data, and the transaction between the research institute and the medical company and the blockchain network platform builder or with the first user. Will be recorded on the blockchain network.

Through this, a research institution or a medical company can obtain reliable medical data that cannot be tampered with, and thus can be used for research and business purposes, and the first user A can manage his / her physical abnormalities. At the same time, you can use your medical data in many different ways.

In addition, the blockchain network 700 in the medical data monitoring system according to an embodiment of the present invention has a data communication route such that communication can be performed through a blockchain network when an abnormality occurs in a generally used network environment. By dualizing a route, it is possible to achieve stability of data securing through data communication.

That is, when the network infrastructure is not smooth or the physical distance between the server and the first terminal 120 or the server and the wearable device 110 may be inconsistent and the concurrency may be reduced, the blockchain network is required. 700 may be utilized.

In particular, when HTTP communication is not smooth and data is not transmitted to the back-end smoothly, it can be applied to AI learning later by making blockchain of the data being measured or collected.

For example, as shown in FIG. 2, when the machine learning module 400 and the server of the first user A are physically separated from each other, the server between the server of the first user and the machine learning module 400 is separated from each other. If the HTTP communication is not smooth and data cannot be transmitted to the back-end smoothly, the data currently being measured / collected is blockchained.

And, among the nodes of the blockchain network 700, a blockchain node closest to the wearable device 110 or the first terminal 120 on the first user's side located at a remote location is designated, and the block data is stored in the blockchain network 700. Transfer and record).

Specifically, the first user A uses a protocol called a gossip protocol to generate a plurality of transactions on the blockchain network 700 before generating a transaction, that is, a transaction that records the obtained biometric and environmental information. After finding the node closest to the first user A among the nodes, the transaction is created and the corresponding data is shared with the nearest node.

The data shared to the nearest node is then equally shared to all nodes on the blockchain network 700.

On the contrary, even when the second user B inquires or requests data, when the node closest to the second user B is allocated using the aforementioned gossip protocol, the data is already on the blockchain network 700. Since the node is shared with all nodes, the second user B can immediately inquire or receive the corresponding data from the assigned nearest node.

Through the above process, the machine learning module 400 can refer to the biometric data and the environmental data sent by the user in an environment where communication is unstable through the node closest to the self.

By transmitting the data referenced from the blockchain node back to the gateway, it can be delivered to the existing back-end infrastructure, which enables the first and second users as well as the third and third users to be physically separated from each other. By eliminating the uncertainty of communication, existing services can be provided without discomfort.

As described above, the medical data monitoring system according to an embodiment of the present invention can build a decentralized platform service by utilizing the blockchain network 700 without relying on a general centralized server infrastructure scheme.

The embodiments described in the present specification and the accompanying drawings merely illustrate some of the technical ideas included in the present invention. Therefore, since the embodiments disclosed in the present specification are not intended to limit the technical spirit of the present invention but to explain, it is obvious that the scope of the technical spirit of the present invention is not limited by these embodiments. Modifications and specific embodiments that can be easily inferred by those skilled in the art within the scope of the technical idea included in the specification and drawings of the present invention are included in the scope of the present invention. It should be interpreted.

Claims (8)

A first gateway receiving real-time first data including biometric information of the first user and the environment information;
A preprocessor configured to preprocess the first data received from the first gateway to generate second data;
Calculate and continuously update the correlation between the biometric information, the environmental information, and the health state of the first user by storing the second data and performing machine learning using the accumulated second data as an input variable. A machine learning module for generating third data which is current or future health state information of the first user based on the second data currently input and the correlation; And
A second gateway receiving second data and third data from the preprocessor and the machine learning unit, respectively;
Including,
In order to generate and output health report information for the first user generated based on the second data and the third data and to provide it to the second user, the second gateway provides the second data and the third data. 2 to the terminal,
And the environment information is set by a first terminal receiving the biometric information in association with the wearable device or obtained from the wearable device.
The method of claim 1, wherein the machine learning module,
A third gateway receiving the second data in real time from the preprocessor;
A first storage unit accumulating and storing second data received in real time from the third gateway to generate big data;
A machine learning unit configured to perform machine learning using second data accumulated in the first storage unit as an input variable to derive a correlation between the biometric information, the environment information, and a health state of the first user;
A determination unit to determine a health state of the first user or to predict a future health state of the first user based on the correlation derived by the machine learning unit; And
A health state notification unit receiving third data which is a determination result of the determination unit and transmitting the received third data to the second gateway;
Medical data monitoring system comprising a.
The method of claim 2, wherein the machine learning module,
And a second storage unit for accumulating and storing the third data.
The method of claim 1,
The medical data monitoring system updating the firmware installed in the wearable device or the first terminal by reflecting the correlation generated and updated in the machine learning module.
The method of claim 1,
And a blockchain network configured to connect at least two or more of the wearable device, the first terminal, the first gateway, the preprocessor, the machine learning module, the second gateway, and the second terminal.
When an error occurs in a network environment for at least two communication among the wearable device, the first terminal, the first gateway, the preprocessor, the machine learning module, the second gateway, and the second terminal, the communication is performed through the blockchain network. Medical data monitoring system.
The method of claim 5,
And at least one of the first data, the second data, and the third data of the first user may be transmitted through the blockchain network after obtaining the consent of the first user.
The method of claim 5,
And a third user server capable of acquiring at least one of the first data, the second data, and the third data of the first user through the blockchain network.
The third user acquires at least one of the first data, the second data, and the third data based on a smart contract of a predetermined contract with the first user or the blockchain network platform builder;
A medical data monitoring system in which details of the contract concluded are recorded on the blockchain network.
The method of claim 1,
And a third storage unit for storing the first data.

Images