KR102329995B1 - Method for calculating and predicting fine dust influence degree of user - Google Patents

Abstract

A method for calculating and predicting the impact of fine dust on a user is provided. According to an aspect of the present invention, a method for calculating and predicting a user's degree of influence of fine dust includes: acquiring a concentration of fine dust corresponding to a user's current location; acquiring PPG data of a user corresponding to the fine dust concentration by using a PPG (photoplethysmic pulse wave) sensor; and synchronizing the fine dust concentration and the PPG data, and calculating user's state information based on the synchronized fine dust concentration and PPG data.

Description

How to calculate and predict the user's fine dust impact

The present invention relates to a method of calculating and predicting the influence of fine dust, and more particularly, to a method of calculating and predicting the influence of fine dust at a user's current location in real time by machine learning the user's state information for each fine dust .

Dust refers to particulate matter that floats or is blown down in the atmosphere. Although the sources of fine dust are natural, in general, soot produced when burning fossil fuels such as coal and oil in boilers and power plants, automobile exhaust gas, construction There are flying dust generated in the field, raw materials in powder form in factories, powder components in the subsidiary material handling process, and smoke from incinerators. The fine dust generated in this way floats in the atmosphere and is deposited on roads, etc. and dispersed again.

Such dust is classified into total suspended particles (TSP) with a particle size of 50 μm or less and particulate matter (PM) with a very small particle size according to the particle size. Fine dust is divided into fine dust with a diameter of less than 10 μm (PM10) and ultrafine particles with a diameter of less than 2.5 μm (hereinafter PM2.5).

If PM10 is about 1/5 to 1/7 smaller than the diameter of a human hair (50~70㎛), PM2.5 is so small that it is only about 1/20~1/30 of a human hair. Since fine dust is so small that it is invisible to the naked eye, it stays in the air and then enters the lungs through the respiratory tract or moves into the body through blood vessels, thereby adversely affecting health.

The World Health Organization (WHO) has been presenting air quality guidelines for fine dust (PM10, PM2.5) since 1987, and in 2013, the International Agency for Research on Cancer (IARC) under the World Health Organization (WHO) established fine dust as carcinogenic to humans. This confirmed group 1 carcinogen (Group 1) has been designated, and the risks to health such as respiratory diseases, cardiovascular diseases, and asthma caused by fine dust are being studied.

Recently, as environmental pollution has become more serious, the frequency of such fine dust alerts is high, which seriously threatens the health of modern people. It is difficult to confirm the fine dust concentration information for the location.

Due to this, despite the low concentration of fine dust, there is a problem of having to go outside with anxiety about health. occurred.

For example, when moving to a destination, it is difficult to sense the risk of fine dust because it is difficult to see how much one is exposed to a bad atmospheric environment, and for this reason, most people neglect their health care in outdoor activities. There is a problem you are doing.

(Patent Document 0001) KR 2017-0136297

The problem to be solved by the present invention is to calculate the fine dust influence by using the fine dust concentration by location of the user and the user's PPG data, and to provide a guide according to the fine dust influence level to the user in real time at the current location. will do

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

According to an aspect of the present invention for solving the above problems, a method for calculating and predicting an influence of a user's fine dust includes: acquiring a fine dust concentration corresponding to a user's current location; acquiring PPG data of a user corresponding to the fine dust concentration by using a PPG (photoplethysmic pulse wave) sensor; and synchronizing the fine dust concentration and the PPG data, and calculating user's state information based on the synchronized fine dust concentration and PPG data.

Other specific details of the invention are included in the detailed description and drawings.

According to the present invention as described above, it has various effects as follows.

The present invention can calculate the user's degree of influence of fine dust by using state information calculated based on PPG data for each fine dust concentration.

In addition, the present invention can provide the user with the degree of influence of fine dust in real time even by acquiring only the concentration of fine dust at the user's current location.

In addition, the present invention can reduce the user's risk due to fine dust by guiding the appropriate fine dust influence level for each user.

In addition, the present invention can calculate whether or not cardiopulmonary disease according to the concentration of fine dust by additionally using the cardiopulmonary function test results.

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

1 is a block diagram illustrating an apparatus for predicting the influence of fine dust according to an embodiment of the present invention.
2 is a flowchart for explaining a method of calculating and predicting the influence of fine dust according to an embodiment of the present invention.
3 is a flowchart for explaining a method of calculating and predicting the degree of influence of fine dust according to another embodiment of the present invention.
4 is an exemplary signal waveform of an ECG signal and a PPG signal according to another embodiment of the present invention.
5 is a block diagram for explaining a method of calculating and predicting the degree of influence of fine dust according to another embodiment of the present invention.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and those of ordinary skill in the art to which the present invention pertains. It is provided to fully understand the scope of the present invention to those skilled in the art, and the present invention is only defined by the scope of the claims.

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. As used herein, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components. Like reference numerals refer to like elements throughout, and "and/or" includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various elements, these elements are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first component mentioned below may be the second component within the spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used herein will have the meaning commonly understood by those of ordinary skill in the art to which this invention belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless specifically defined explicitly.

Spatially relative terms "below", "beneath", "lower", "above", "upper", etc. It can be used to easily describe the correlation between a component and other components. A spatially relative term should be understood as a term that includes different directions of components during use or operation in addition to the directions shown in the drawings. For example, when a component shown in the drawing is turned over, a component described as “beneath” or “beneath” of another component may be placed “above” of the other component. can Accordingly, the exemplary term “below” may include both directions below and above. Components may also be oriented in other orientations, and thus spatially relative terms may be interpreted according to orientation.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating an apparatus for predicting the influence of fine dust according to an embodiment of the present invention.

Referring to FIG. 1 , the apparatus 100 for predicting the influence of fine dust according to an embodiment of the present invention may obtain the concentration of fine dust at the user's current location and predict the influence of fine dust in real time. For example, the degree of influence of fine dust may refer to a general term for health abnormalities such as cardiopulmonary disease of users affected by fine dust. In addition, in the present invention, the word "user's risk" is also used interchangeably. The user's risk level may mean a probability of the user's health abnormality (eg, cardiopulmonary disease).

For example, the fine dust influence prediction apparatus 100 may acquire the fine dust concentration and PPG data at the user's current location to obtain the user's state information for each fine dust, and obtain the state information for each fine dust. After repeatedly accumulating the learning data, a fine dust influence prediction model can be created, and the user's risk can be output by inputting the fine dust concentration into the fine dust influence prediction model. In addition, as will be described later, the accuracy of the fine dust influence prediction model can be further improved by additionally considering the user's electrocardiogram data.

In an embodiment, the apparatus 100 for predicting the influence of fine dust may have a dedicated program configured to set the method of predicting the influence of fine dust may be installed. For example, the apparatus 100 for predicting the impact of fine dust includes a PPG sensor 110 capable of acquiring PPG data, an ECG sensor 120 capable of acquiring ECG data, and a data processing unit 130 that pre-processes the acquired data. ), a state information calculating unit 140 that calculates the user's state information based on PPG data and fine dust concentration, and a deep learning unit 150 that performs machine learning using learning data and a deep neural network and outputs the user's risk level ), a user risk determination unit 160 that determines whether a user has an abnormality in health, a database ( 170), and may include a display 180 for displaying a guide message.

In an embodiment, the external device 200 may be a medical institution server or a medical device capable of acquiring the user's cardiopulmonary function test result data and transmitting the data to the apparatus 100 for predicting the effect of fine dust.

For example, the PPG sensor 110 may be implemented while being separated from the main body of the smart watch and mounted on the watch strap. In this case, when the user wears the smart watch on the wrist, the PPG sensor mounted on the watch strap is in close contact with the user's wrist, thereby acquiring the user's photoplethysmogram by detecting the pulse on the wrist.

For example, the electrocardiogram sensor 120 includes an electrocardiogram electrode and a reference electrode (or a ground electrode), and can detect an electrocardiogram signal. In some cases, there may be two ECG electrodes. The electrocardiogram sensor may be mounted on the watch strap of a smart watch. Of course, unlike this, the electrocardiogram sensor 120 may be omitted, and the fine dust influence prediction apparatus 100 may receive the user's electrocardiogram data from the external device 200 .

For example, the apparatus 100 for predicting the impact of fine dust may audibly warn the user through a speaker that the risk is high according to the fine dust.

Also, the external device 200 may be a fine dust measuring server or a fine dust concentration measuring device capable of transmitting the fine dust concentration of the user's current location. The fine dust concentration meter is to measure the concentration of air pollutants such as fine dust, VOC (Volatile Organic Compound) and temperature/humidity contained in the air. At least one possible sensor (not shown) may be provided. When the external device 200 is a fine dust concentration meter, the measured concentration information may be transmitted to the user's fine dust influence prediction device through a wireless communication network. Of course, the user's fine dust effect prediction device 100 may be connected to the external device 200 through wireless communication such as Bluetooth, RFID, ZigBee, and WIFI to be interoperable.

In an embodiment, the apparatus 100 for predicting the effect of fine dust may be a wearable device. Of course, the fine dust concentration can be received in real time according to the user's current location, and any electronic device that is easy to carry can constitute the apparatus 100 for predicting the effect of fine dust. That is, for example, the apparatus 100 for predicting the effect of fine dust is a smartphone, a tablet personal computer, a mobile phone, a video phone, an e-book reader, and a PDA. It may include at least one of a personal digital assistant, a portable multimedia player (PMP), an MP3 player, a mobile medical device, and a camera.

2 is a flowchart for explaining a method of calculating and predicting the influence of fine dust according to an embodiment of the present invention.

Referring to FIG. 2 , in an embodiment, in operation 21 , the data processing unit 130 may acquire the fine dust concentration corresponding to the user's current location. For example, the data processing unit 130 may receive the fine dust concentration corresponding to the user's current location from the external device 200 through the communication module and the GPS module.

In an embodiment, in operation 22 , the PPG sensor 110 may acquire the user's PPG data corresponding to the fine dust concentration using a PPG (photoplethysmography) sensor. For example, the photoplethysmogram (PPG) uses the periodic change in blood flow in blood vessels as the heart beats. measurement to estimate the heart rate.

For example, the PPG sensor 110 may control to measure the bio-signal every predetermined time interval (eg, 15 minutes) to collect the measured bio-signal as learning data. Alternatively, the PPG sensor 110 may collect biosignals whenever the user's location changes.

In an embodiment, in operation 23 , the state information calculating unit 140 may synchronize the fine dust concentration with the PPG data. Here, synchronization may mean controlling the fine dust concentration and PPG data acquired based on the user's current location as data corresponding to each other. That is, when the fine dust concentration is a specific value, the corresponding PPG data may be matched.

In one embodiment, in operation 24, the state information calculator 150 may calculate the user's state information based on the synchronized fine dust concentration and PPG data. For example, the user's heart rate, body temperature, stress level, and the degree of abnormality or distortion compared to a normal PPG waveform may be analyzed through the PPG data, and the analysis result may be user status information. Accordingly, the apparatus 100 for predicting the effect of fine dust according to the present invention may store the user's state information for each fine dust concentration. For example, when a specific fine dust concentration is present, the user's heart rate may be stored as the user's state information.

In an embodiment, in operation 25, the deep learning unit 150 repeatedly performs the steps of obtaining the fine dust concentration, obtaining the PPG data, and calculating the user's state information, thereby increasing the fine dust concentration, PPG data and status information can be continuously accumulated and stored. That is, it cannot be concluded that the user always has the same state information at the corresponding fine dust concentration only by acquiring the data once or twice. It is possible to continuously accumulate and store fine dust concentration, PPG data, and status information corresponding to

In an embodiment, in operation 26, the deep learning unit 150 may input the fine dust concentration into the deep neural network, and use the user's state information corresponding to the fine dust concentration as a correct answer value through the deep neural network. The user's risk level can be output, and a fine dust impact prediction model can be created and continuously updated by repeatedly machine learning the input and output actions to the deep neural network.

That is, the fine dust impact prediction model of the present invention can output the user's risk level in real time when only the fine dust concentration at the user's current location is input. To this end, PPG data for each user's fine dust concentration and corresponding user status information can be continuously collected as learning data, and the fine dust impact prediction model can be continuously created and updated by machine learning.

For example, as described above, the user's risk level may be whether the user's health is abnormal according to the fine dust concentration, and may have a probability value. For example, the probability value corresponding to the user's risk may have a value in the range of 0 to 1, if the probability value is close to 0, it may be determined that the user is in a normal state, and if it is close to 1, it is considered that there is a health problem. can judge Here, the health condition may be, for example, cardiopulmonary disease.

For example, the deep neural network may be a CNN or an RNN. In a recurrent neural network (RNN), one or more feature values (vectors) are input, and the output values of the hidden layer or the output layer for the previously input feature values are also input to the new input feature values. CNN (Convolutional Neural Network) receives two or more dimensional vectors or matrices as input, and convolutional or pooling steps are performed before the general ANN step. In the convolution step, a multidimensional matrix with weights and a convolution operation are performed. In the pooling step, there are max pooling or mean pooling methods.

In an embodiment, in operation 27 , the user risk determination unit 120 may display an outing guide message to the user through the display 180 based on the user's risk level. For example, the outing guide message may include text such as "Please refrain from going out because the current concentration of fine dust is high and may be dangerous to your health."

In an embodiment, the user's risk level may be assigned a color for each risk level in advance, and the user risk determination unit 120 may display a color designated according to the risk level through the display 180 .

3 is a flowchart for explaining a method of calculating and predicting the degree of influence of fine dust according to another embodiment of the present invention. 4 is an exemplary signal waveform of an ECG signal and a PPG signal according to another embodiment of the present invention. 5 is a block diagram for explaining a method of calculating and predicting the degree of influence of fine dust according to another embodiment of the present invention.

The operations of FIG. 3 may be an embodiment of increasing the accuracy of the fine dust effect by additionally considering the electrocardiogram data and cardiopulmonary function test result data. Accordingly, content overlapping with FIG. 2 will be omitted for simplicity of description.

3 to 5 , in an embodiment, in operation 31 , the data processing unit 130 may acquire the user's ECG data. For example, the electrocardiogram (ECG) signal is an important signal used for judging diseases of the heart system. is a signal that can be identified. The data processing unit 130 may obtain ECG data by performing A/D conversion of the ECG signal.

In an embodiment, in operation 32 , the state information calculating unit 140 may synchronize the electrocardiogram data, the fine dust concentration, and the PPG data.

For example, as shown in FIG. 4 , a PPG signal for measuring a heart rhythm or a heartbeat has a contour of a specific waveform and is different from the contour of a waveform of an ECG signal. ECG measures the electrical activity of the heart and the ECG signal (curve 250) contains a prominent feature known as the QRS complex representing the heart's major pumping contraction. The R peak in the ECG signal is used by the heart rate algorithm to measure the time that occurs between pulsating pulses. The duration between each R peak is referred to as the RR interval.

On the other hand, PPG measures the pressurized pulse acting on the arteries in the body, which dilates the arteries slightly before returning to their former state. The PPG signal is an optical signal whose amplitude is directly proportional to the pulse pressure. The PPG signal (curve 252) includes quasi-periodic pulses with peaks and valleys that can be used to indicate the periodicity of the signal waveform and allow for an estimation of the heart rate. In particular, the duration between the peaks of two adjacent pulses or between the minimum points of two adjacent pulses is referred to as the Inter-beat Interval (IBI), which can be used as an indicator of heart rate. In some cases, the PPG signal exhibits overlapping notch. An overlap notch is a small downward bend observed in the downstroke of the arterial pressure waveform. This represents the intersection of superimposed primary and reflected pressure waves in the arterial tree.

Accordingly, the state information calculating unit 140 may match the electrocardiogram data and the PPG data having the above characteristics with a specific fine dust concentration.

In an embodiment, in operation 33, the state information calculating unit 140 may calculate the user's state information based on the synchronized electrocardiogram data, the fine dust concentration, and the PPG data. For example, a user's heart rate, body temperature, stress level, and abnormality or distortion compared to a normal waveform may be analyzed through PPG data and electrocardiogram data, and the analysis result may be user status information. In particular, by combining the two data, the user's state information may include whether the user's cardiopulmonary disease has increased accuracy.

In one embodiment, in operation 34, the deep learning unit 150 repeatedly performs the steps of obtaining electrocardiogram data, obtaining fine dust concentration, obtaining PPG data, and calculating user's state information. By doing so, ECG data, fine dust concentration, PPG data, and status information can be continuously accumulated and stored.

In an embodiment, in operation 35, the deep learning unit 150 may input the fine dust concentration into the deep neural network, and use the user's state information corresponding to the fine dust concentration as a correct answer value through the deep neural network. The user's risk level can be output. That is, the fine dust impact prediction model of the present invention can output the user's risk level in real time when only the fine dust concentration at the user's current location is input, and can increase the risk prediction probability by considering the electrocardiogram data. To this end, PPG data and electrocardiogram data for each fine dust concentration of the user and the user's state information can be continuously collected as learning data, and the fine dust impact prediction model can be continuously generated and updated by machine learning.

For example, as described above, the user's risk level may be whether the user's health is abnormal according to the fine dust concentration, and may have a probability value. For example, the probability value corresponding to the user's risk may have a value in the range of 0 to 1, if the probability value is close to 0, it may be determined that the user is in a normal state, and if it is close to 1, it is considered that there is a health problem. can judge Here, the health condition may be, for example, cardiopulmonary disease.

In an embodiment, in operation 36, the user risk determination unit 130 may acquire the user's cardiopulmonary function test result data, and in operation 37, the user risk determiner 120 determines the cardiopulmonary function test result data and the user's Based on the risk level, the user's likelihood of developing cardiopulmonary disease can be predicted. That is, according to the present invention, by additionally acquiring the user's cardiopulmonary function test result data from the external device 200 , it is possible to further increase the accuracy of the user's possibility of developing cardiopulmonary disease.

According to an aspect of the present invention, a method for calculating and predicting a user's degree of influence of fine dust includes: acquiring a concentration of fine dust corresponding to a user's current location; acquiring PPG data of a user corresponding to the fine dust concentration by using a PPG (photoplethysmic pulse wave) sensor; and synchronizing the fine dust concentration and the PPG data, and calculating user's state information based on the synchronized fine dust concentration and PPG data.

According to various embodiments, the fine dust concentration, the PPG data, and the status information are continuously maintained by repeatedly performing the steps of obtaining the fine dust concentration, obtaining the PPG data, and calculating the user's state information. It may further include the step of accumulating and storing.

According to various embodiments, inputting the fine dust concentration into a deep neural network; and outputting the user's risk level through the deep neural network using the user's state information corresponding to the fine dust concentration as a correct answer value.

According to various embodiments, the method may further include continuously updating the fine dust impact prediction model by repeatedly machine-learning the input to the deep neural network and the output to the deep neural network.

According to various embodiments, the method may further include displaying an outing guide message to the user through a display based on the user's risk level.

According to various embodiments of the present disclosure, the method may include: acquiring data of the cardiopulmonary function test result of the user; and predicting the possibility of occurrence of cardiopulmonary disease of the user based on the cardiopulmonary function test result data and the user's risk level.

According to various embodiments, the method may further include acquiring the user's ECG data.

According to various embodiments, synchronizing the electrocardiogram data, the fine dust concentration, and the PPG data, and calculating the user's state information based on the synchronized electrocardiogram data, the fine dust concentration, and the PPG data; can

According to various embodiments, by repeatedly performing the steps of obtaining the electrocardiogram data, obtaining the fine dust concentration, obtaining the PPG data, and calculating the user's state information, continuously accumulating and storing the fine dust concentration, the PPG data, and status information; inputting the fine dust concentration into a deep neural network; outputting the user's risk level through the deep neural network by using the user's state information corresponding to the fine dust concentration as a correct answer value; and continuously updating the fine dust impact prediction model by repeatedly performing machine learning of input and output to the deep neural network.

According to various embodiments, the user's fine dust influence calculation and prediction program may be stored in a medium in order to execute the method of any one of claims 1 to 9 by being combined with a computer that is hardware.

The steps of a method or algorithm described in relation to an embodiment of the present invention may be implemented directly in hardware, as a software module executed by hardware, or by a combination thereof. A software module may contain random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any type of computer-readable recording medium well known in the art to which the present invention pertains.

As mentioned above, although embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art to which the present invention pertains can realize that the present invention can be embodied in other specific forms without changing its technical spirit or essential features. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

100: Fine dust influence prediction device
110: PPG sensor
120: electrocardiogram sensor
130: data processing unit
140: state information calculation unit
150: deep learning unit
160: user risk determination unit
170: database
180: display

Claims (10)

In the method of calculating and predicting the effect of fine dust on the user,
obtaining a fine dust concentration corresponding to the user's current location;
acquiring PPG data of a user corresponding to the fine dust concentration by using a PPG (photoplethysmic pulse wave) sensor; and
Synchronizing the fine dust concentration and the PPG data and calculating the user's first state information based on the synchronized fine dust concentration and PPG data;
Calculating the first state information includes:
Calculating the user's first state information at a specific fine dust concentration through analysis of PPG data synchronized to a specific fine dust concentration,
The method is
Continuously accumulating the fine dust concentration, the PPG data, and the first state information by repeatedly performing the steps of obtaining the fine dust concentration, obtaining the PPG data, and calculating the first state information of the user to save it;
applying the fine dust concentration as an input value to the deep neural network, and applying the user's first state information when the fine dust concentration is the correct answer value to the deep neural network;
outputting a user's risk level when the fine dust concentration is based on the applied result; and
Further comprising; continuously updating the fine dust impact prediction model by repeatedly machine-learning the application step and the output step;
The first state information includes a user's heart rate, body temperature, stress level, and a degree of distortion compared to a normal PPG waveform,
The fine dust influence prediction model, when a specific fine dust concentration is input, the user's fine dust influence calculation and prediction method, characterized in that for predicting the risk of the user at the specific fine dust concentration. delete delete delete The method of claim 1, further comprising: displaying an outing guide message to the user through a display based on the user's risk level. The method of claim 1 , further comprising: acquiring the cardiopulmonary function test result data of the user; and
Predicting the possibility of occurrence of cardiopulmonary disease of the user based on the cardiopulmonary function test result data and the user's risk level; The method of claim 1, further comprising: acquiring the user's electrocardiogram data. The method of claim 7, further comprising: synchronizing the electrocardiogram data, the fine dust concentration, and the PPG data, and calculating the user's second state information based on the synchronized electrocardiogram data, the fine dust concentration, and the PPG data; A method of calculating and predicting the impact of fine dust on the user, characterized in that it includes. The electrocardiogram data according to claim 8, wherein the electrocardiogram data is obtained by repeatedly performing the obtaining of the electrocardiogram data, the obtaining of the fine dust concentration, the obtaining of the PPG data, and the calculating of the user's second state information. , continuously accumulating and storing the fine dust concentration, the PPG data, and second state information;
applying the fine dust concentration as an input value to the deep neural network, and applying second state information of a user corresponding to the fine dust concentration as a correct answer value to the deep neural network;
outputting a user's risk level when the fine dust concentration is based on the applied result;
Continuously updating the fine dust influence prediction model by repeatedly performing machine learning of the application step and the output step; In combination with a computer that is hardware, claim 1, claim 5 to claim 9, stored in the medium to execute any one of the method, the user's fine dust effect calculation and prediction program.

Images