KR20200082062A - Method for inferring life pattern and changed factor based on blood test result - Google Patents

Abstract

According to an embodiment of the present invention, a method of estimating a lifestyle pattern and a varying factor based on a blood test result, which is a method of estimating the lifestyle pattern of an examinee by obtaining lab data in accordance with a blood test of the examinee from a terminal, includes the following steps of: obtaining first lab data in accordance with a first blood test; obtaining second lab data in accordance with a second blood test after a first period; inputting the first lab data and the second lab data into a deep learning neural network by lifestyle pattern item, and obtaining first period lifestyle pattern information outputted from the deep learning neural network by lifestyle pattern item; and displaying the first period lifestyle pattern information. The lifestyle pattern information includes lifestyle information about at least one item among drug compliance, drinking habits, eating habits, and amount of the examinee′s experience.

Description

Method of estimating life patterns and change factors based on blood test results {METHOD FOR INFERRING LIFE PATTERN AND CHANGED FACTOR BASED ON BLOOD TEST RESULT}

The present invention relates to a method and apparatus for estimating life patterns and change factors based on blood test results.

More specifically, it relates to a method and apparatus for estimating the life pattern of the subject for a specific period of time based on a plurality of blood test results for the subject.

Blood test is a test of various components of blood for the purpose of diagnosis treatment and prognosis of diseases.

The blood test consists of collecting blood from a subject using a syringe and analyzing the collected blood using a blood analysis mechanism to obtain blood information.

However, the conventional blood analysis mechanism provides only the result information for each blood test that has been performed in one shot, and general testers use the blood test result information to find out the presence or absence of a disease.

However, skilled doctors can understand not only the presence or absence of disease, but also the lifestyle of the examinee through blood test result information.

For example, experienced doctors compare previous blood test results to current blood test results, and determine lifestyle information, such as the degree of adherence to drugs or the amount of alcohol intake, between test periods by experience, and the accuracy is quite high. Afterwards, it can be reflected in treatment or treatment to provide high-quality treatment and treatment.

However, even an experienced doctor may be difficult to judge accurately depending on the situation because the judgment is subjective, and inexperienced doctors cannot make this judgment at all, and there is a problem that the level of medical treatment may differ according to the skill level of doctors.

Conventional blood analysis mechanisms provide only information on diseases according to blood test results, and have difficulty in performing analysis on the subject's lifestyle change information and/or state change information, and the wrong life causing various diseases There is a limitation in estimating habits, and a solution to this is needed.

10-1054732 B

The present invention is based on blood test results capable of estimating a life pattern of a test subject for a specific period based on a plurality of blood test lab data for a test subject using deep learning in order to estimate objective life pattern information through a blood test. To provide a method and apparatus for estimating life patterns.

However, the technical problems to be achieved by the present invention and embodiments of the present invention are not limited to the technical problems as described above, and other technical problems may exist.

A method and apparatus for estimating life patterns and change factors based on blood test results according to an embodiment of the present invention is a method for estimating the life pattern of the test subject by obtaining lab data according to the blood test of the test subject at the terminal, the first blood test Obtaining first wrap data according to; Obtaining second lab data according to a second blood test after the first period; Inputting the first lap data and the second lap data into a deep learning neural network for each life pattern item, and obtaining first period life pattern information output from the deep learning neural network for each life pattern item; And displaying the life pattern information of the first period, and the life pattern information includes lifestyle information on at least one item of the drug compliance, drinking habits, eating habits, and exercise amount of the examinee.

In this case, in the obtaining of the first period life pattern information, the first lab data and the second lab data are input to the drug compliance neural network, and the subject during the first period output from the drug compliance neural network And obtaining a drug compliance evaluation index, which is an index indicating the degree of compliance with medication use according to the medication guidance for.

In addition, in the obtaining of the first period life pattern information, the first lap data and the second lap data are input to the drinking habit estimation deep learning neural network, and the drinking output of the drinking habit estimation deep learning neural network is output. And obtaining an alcohol-related evaluation index proportional to the amount of alcohol and the number of alcoholic beverages of the testee during one period.

The displaying of the first period life pattern information may include outputting the first period life pattern information through medical consultation supplementary content, and the medical consultation supplementary content includes specific items of the lifestyle pattern information. The method further includes displaying blood test blood component change information associated with the blood test in association with a specific item of the life pattern information.

The method and apparatus for estimating life patterns and change factors based on blood test results according to an embodiment of the present invention include: obtaining third lab data according to a third blood test after a second period, and the second lab data and the Obtaining second period life pattern information based on third lap data, obtaining third period life pattern information based on the first lap data and the third lap data, and the first to third And displaying life pattern information of the period.

A method and apparatus for estimating a life pattern based on blood test results according to an embodiment of the present invention, by acquiring a plurality of blood test lab data for a test subject and estimating the test subject life pattern for a specific period based on the acquired lab data In addition, it provides an objective and accurate life pattern estimation information for a specific period of time, and has the advantage of performing a high-quality medical examination and treatment.

In addition, the method and apparatus for estimating life patterns based on blood test results according to an embodiment of the present invention may obtain more accurate life pattern information by mutually correcting life pattern information obtained through a plurality of estimated lap data.

In addition, the method and apparatus for estimating life patterns based on blood test results according to an embodiment of the present invention are automated and fast and efficient blood test results by estimating life patterns of subjects through artificial intelligence data processing through a learned deep learning neural network. It is possible to provide a service based on estimation of life patterns.

In addition, the method and apparatus for estimating life patterns based on blood test results according to an embodiment of the present invention effectively generate a training data set by learning a deep learning neural network based on medical consultation content generated based on blood test results You can, and you can train deep learning neural networks based on more reliable data.

In addition, the method and apparatus for estimating life patterns based on blood test results according to an embodiment of the present invention display and provide the estimated life pattern information in various ways, so that the user can easily and intuitively receive life pattern information based on blood test results. Can be confirmed.

However, the effects obtainable in the present invention are not limited to the above-mentioned effects, and other effects not mentioned can be clearly understood from the following description.

1 is a conceptual diagram showing a system for estimating life patterns based on blood test results according to an embodiment of the present invention.
2 is an example of a terminal appearance according to an embodiment of the present invention.
3 is a view for explaining a life pattern estimation server according to an embodiment of the present invention.
4 is a flowchart illustrating a method for estimating life patterns based on blood test results according to an embodiment of the present invention.
5 is an example of lab data according to an embodiment of the present invention.
6 is an example of outputting a living pattern information according to an embodiment of the present invention as a graphic image through medical consultation assistance content.
7 is an example of displaying blood test lab data associated with each item of life pattern information and life pattern information according to an embodiment of the present invention through medical consultation supplementary content.
8 is a view for explaining a method of displaying life pattern information based on first to third blood tests in various ways according to an embodiment of the present invention.
9 is a flowchart illustrating a method of constructing a deep learning neural network for each living pattern item according to an embodiment of the present invention.

The present invention can be applied to various transformations and can have various embodiments, and specific embodiments will be illustrated in the drawings and described in detail in the detailed description. Effects and features of the present invention and methods for achieving them will be clarified with reference to embodiments described below in detail together with the drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various forms. In the following examples, terms such as first and second are not used in a limiting sense, but for the purpose of distinguishing one component from other components. In addition, a singular expression includes a plural expression unless the context clearly indicates otherwise. In addition, terms such as include or have means that a feature or component described in the specification exists, and does not preclude the possibility of adding one or more other features or components in advance. In addition, in the drawings, the size of components may be exaggerated or reduced for convenience of description. For example, since the size and thickness of each component shown in the drawings are arbitrarily shown for convenience of description, the present invention is not necessarily limited to what is shown.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. When describing with reference to the drawings, identical or corresponding components will be denoted by the same reference numerals and redundant description thereof will be omitted. .

1 is a conceptual diagram showing a system for estimating life patterns based on blood test results according to an embodiment of the present invention.

Referring to FIG. 1, the life pattern estimation system based on blood test results may include a terminal 100, a life pattern estimation server 200, and a medical consultation assistance service providing server 300.

Here, each component of FIG. 1 may be connected through a network. The network refers to a connection structure capable of exchanging information between each node, such as the terminal 100, the life pattern estimation server 200, and the medical consultation assistant service providing server 300. In one example of such a network, 3GPP ( 3rd Generation Partnership Project (LTE) network, Long Term Evolution (LTE) network, World Interoperability for Microwave Access (WIMAX) network, Internet, Local Area Network (LAN), Wireless Local Area Network (LAN), Wide Area (WAN) Network, PAN (Personal Area Network), Bluetooth (Bluetooth) network, satellite broadcasting network, analog broadcasting network, DMB (Digital Multimedia Broadcasting) network, and the like.

-Terminal

First, in the embodiment of the present invention, the terminal 100 is a smart terminal, a digital broadcasting terminal, a mobile phone, PDA (personal digital assistants), PMP (PMP) that is a portable terminal in which a program for performing a blood test-based life pattern estimation service is installed. portable multimedia player), navigation, a tablet PC (tablet PC), a wearable device (wearable device) and a smart glass (smart glass).

In addition, the terminal 100 provides a blood test-based life pattern estimation service based on wired/wireless communication, such as a fixed terminal, a desktop PC, a laptop computer, and a personal computer such as an ultrabook. It may further include a device with a program for installation.

In addition, in the embodiment of the present invention, the terminal 100 may acquire and store rap data according to the blood test of the testee, and obtain and output life pattern information of the testee based on the acquired lab data. .

In detail, the terminal 100 may acquire a plurality of lap data according to the blood test of the examinee, and may acquire the life pattern information by transmitting the obtained plurality of lap data to the life pattern estimation server 200, The acquired life pattern information may be transmitted to the medical consultation assistant service providing server 300 to obtain and output medical consultation assistant content.

Further, the terminal 100 may provide a questionnaire capable of grasping the stress level of the testee in order to more closely grasp the stress level occurring in the life of the testee.

In detail, the terminal 100 may receive and output the medical consultation assistant content that can analyze the stress level of the examinee through the questionnaire in the form of a question and answer from the medical consultation assistant service providing server 300.

In addition, the terminal 100 may acquire the subject stress questionnaire information that can help to grasp the stress information of the subject through the questionnaire.

In addition, the terminal 100 transmits the obtained subject stress questionnaire information to the life pattern estimation server 200 and/or the medical counseling assistance service providing server 300 to effectively acquire and utilize the life pattern information of the examinee in the future. I can assist.

Hereinafter, each component constituting the terminal 100 will be described in detail with reference to the accompanying drawings.

2 is an example of the appearance of the terminal 100 according to an embodiment of the present invention.

Referring to FIG. 2, the terminal 100 includes a communication unit 110, an input unit 120, a display unit 130, a touch screen 135, a camera 140, a storage unit 150, and a microphone ( 160), a speaker 170, and a control unit 180.

First, the communication unit 110 may transmit and receive various data and/or information for providing a blood test-based life pattern estimation service.

In an embodiment, the communication unit 110 communicates with the terminal 100 of another user, the life pattern estimation server 200 and/or the medical consultation assistance service providing server 300, and is associated with a blood test-based life pattern estimation service. Data (eg, life pattern information, etc.) can be transmitted and received.

In addition, the communication unit 110 may assist in printing the offline pattern by transmitting the life pattern information to an external printer.

The communication unit 110 includes technical standards or communication methods for mobile communication (eg, Global System for Mobile Communication (GSM), Code Division Multi Access (CDMA), High Speed Downlink Packet Access (HSDPA), HSUPA ( High speed uplink packet access (LTE), Long Term Evolution (LTE), Long Term Evolution-Advanced (LTE-A), etc. Send and receive signals.

Next, the input unit 120 may detect a user input related to a blood test-based life pattern estimation service.

In an embodiment, the input unit 120 may detect a user's input through a counseling input interface provided in a medical consultation process.

Next, the display 130 may output a graphic image of various information related to a blood test-based life pattern estimation service.

The display unit 130 includes a liquid crystal display (LCD), a thin film transistor-liquid crystal display (TFT LCD), an organic light-emitting diode (OLED), and a flexible display. (flexible display), a three-dimensional display (3D display), an electronic ink display (e-ink display) may include at least one.

Also, the input unit 120 and the display unit 130 may be combined to be implemented as a touch screen 135.

Next, the camera 140 may acquire a medical consultation-related image by photographing a course of treatment and medical consultation, and the terminal 100 that obtained the image may output the obtained image through the medical consultation supplementary content. have.

In addition, the camera 140 may acquire an image by photographing a direction side which is disposed on the front or/and rear of the terminal 100 and is disposed outside the terminal 100 to perform a treatment process from an external viewpoint. You can also shoot.

When the camera 140 is disposed outside the terminal 100, the camera 140 may transmit a medical procedure image photographed by the controller 180 through the communication unit 110.

In addition, the camera 140 may include an image sensor and an image processing module.

In detail, the camera 140 may process a still image or a moving image obtained by an image sensor (for example, CMOS or CCD).

In addition, the camera 140 may process a still image or a video acquired through an image sensor using an image processing module to extract necessary information, and transmit the extracted information to the controller 180.

Next, the storage unit 150 may store any one or more of various application programs, data, and instructions that provide a blood test-based life pattern estimation service according to an embodiment of the present invention.

In an embodiment, the storage unit 150 may store and manage a plurality of lap data, life pattern information, and/or medical consultation assistance content.

The storage unit 150 may be various storage devices such as ROM, RAM, EPROM, flash drive, hard drive, etc., and web storage performing a storage function of the storage unit 150 on the Internet. ).

Next, the microphone 160 may detect a voice input of an examinee and/or a doctor in a medical consultation process, and may acquire counseling content recording information based on the detected voice.

At this time, the terminal 100 that has obtained the counseling content recording information may output the acquired recording information through the medical consultation assistant content.

Next, the speaker 170 may output audio information related to a blood test-based life pattern estimation service.

In an embodiment, the speaker 170 may output and provide recording information of consultation contents included in the medical consultation auxiliary content.

Finally, the controller 180 can control the overall operation of each component described above in order to provide a blood test-based life pattern estimation service.

The controller 180 includes application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, It may be implemented using at least one of micro-controllers, microprocessors, and electrical units for performing other functions.

However, the components illustrated in FIG. 2 are not essential for implementing the terminal 100, and thus the terminal 100 described in this specification may have more or fewer components than those listed above. have.

-Life pattern estimation server

Meanwhile, in the embodiment of the present invention, the life pattern estimation server 200 may receive lab data according to the blood test result of the test subject from the terminal 100, and based on the received lab data, the life pattern information of the test subject Or/and output a change factor.

Here, the lab data refers to experimental data obtained by examining blood collected from a subject. Here, the lab data includes blood elements Fasting glucose, HbA1c (glycosylated hemoglobin), BUN (Blood Urea Nitrogen), Creatinine (Creatinine), Total Bilirubin (Total Bilirubin), AST (GOT), and ALT. (GPT), ALP (Alkaline Phosphatase), γ-GTP (gammajitipi), Total Cholesterol (total cholesterol), Triglyceride (triglyceride), HDL-Cholesterol (high density lipoprotein cholesterol), LDL-Cholesterol (low density lipoprotein cholesterol) and CPK may include at least one numerical value or percentage information. In addition, the lifestyle pattern information refers to lifestyle score information for at least one of drug compliance, drinking habits, eating habits, and exercise, which are indicators of the degree to which the test subject adhered to the drug use map. Further, the life pattern information may further include weight change information and/or stress information of the examinee. In general, information on changes in body weight can be an important indicator to comprehensively check whether proper lifestyle has been maintained.

In addition, the change factor means a change factor that matches a blood factor that needs to be reviewed by changing a predetermined percentage or more compared to a previous blood test. More specifically, the change factor means a factor that the blood element may change when there are blood elements that increase/decrease over a predetermined value or increase/decrease over a predetermined percentage for each blood element.

For example, if the blood sugar level rises sharply, it may be possible to estimate acute disease occurrence or cancer such as pancreatic cancer, steroid use, Chinese medicine use, hypo pituitary function, or liver function decrease. Conversely, if the blood sugar is rapidly lowered, it may be possible to estimate a hypoplastic function or a decrease in liver function as a change factor.

Although the above-mentioned life pattern information and change factors are very fluid in the amount of change along the timeline, the subject may not be able to grasp the change or may be inconvenient to comment to the examiner, and question or measure each time at each treatment Edo can be cumbersome and difficult.

In order to obtain life pattern information that can solve this, in the embodiment of the present invention, the life pattern estimation server 200 may output the life pattern information by inputting the received lap data into a deep learning neural network for each life pattern item. have.

In addition, the life pattern estimation server 200 may acquire the weight and/or stress score information of the examinee based on the output life pattern information or/and lab data through a deep learning neural network.

In detail, the life pattern estimation server 200 may input glycated hemoglobin, fasting blood sugar, and average heart rate into a deep learning neural network to estimate changes in stress index and sleep quality within a specific period.

In addition, the life pattern estimation server 200 may input the liver value and the fasting blood glucose into the deep learning neural network to estimate the patient's weight change during a specific period.

The weight information and the stress information thus obtained may be further included in the corresponding life pattern information to be used as supplementary medical consultation content.

Meanwhile, referring to FIG. 3, the life pattern estimation server 200 may include a data transmission/reception unit 210, a data processing unit 220, and a database 230.

First, the data transmission/reception unit 210 may exchange various data for providing a blood pattern-based life pattern estimation service with the terminal 100 and/or an external server through a network.

Next, the data processing unit 220 may perform a series of data processing to provide a blood test-based life pattern estimation service.

In particular, the data processing unit 220 may perform deep learning based on blood tests in conjunction with a deep learning neural network for each life pattern item.

At this time, according to the embodiment, such a deep learning neural network for each life pattern item is installed directly on the life pattern estimation server 200 or is a separate device from the life pattern estimation server 200 based on blood tests from the terminal 100. Deep learning can be performed by receiving information.

Hereinafter, in the embodiment of the present invention, a deep learning neural network for each life pattern item is directly installed in the life pattern estimation server 200 and is described based on an embodiment in which deep learning is performed.

Returning again, the data processing unit 220 of the life pattern estimation server 200 includes a main processor that controls all units according to an embodiment, and a plurality of graphic processors that process a large amount of computation required when driving a deep learning neural network ( Graphics Processing Unit (GPU).

In addition, the data processing unit 220 reads the deep learning neural network driving program for each living pattern item built to perform deep learning from the database 230 and performs the following according to the constructed deep learning neural network system for each living pattern item. You can perform the deep learning you describe.

In detail, here, the deep learning neural network for each living pattern item may include a drug compliance deep learning neural network, a drinking habit estimation deep learning neural network, a diet estimation deep learning neural network, an exercise estimation deep learning neural network, and a change factor estimation neural network. Can.

Here, the drug compliance deep learning neural network, based on the received lab data, can determine whether the test subject adhered well to the drug use in accordance with the guided drug use guidelines, and output the result of the determination as an evaluation index.

At this time, the evaluation index is one means for indicating the life pattern information, and in other embodiments, the life pattern information may be expressed through life pattern types or texts classified through certain criteria other than the evaluation index.

For example, the drug compliance deep learning neural network may grasp the degree of compliance of the test subject's medication use map and display the result in an arbitrary value between 1-100.

In detail, the drug compliance deep learning neural network outputs a high value of 1 to 100, and determines that the medication map has not been complied with when the testee determines that the drug use was well observed according to the medication map. You can print out a low number.

That is, the drug compliance deep learning neural network may input a plurality of lab data and output a drug compliance evaluation index, which is an index indicating the degree of drug compliance according to a medication map during a period between lab data.

In addition, the deep learning neural network for estimating drinking habits can estimate information on the amount of drinking, the number of drinking, etc. of the examinee during a specific period based on the received lab data, and output the estimated drinking-related evaluation index.

That is, the drinking habit estimation deep learning neural network may output a drinking-related evaluation index proportional to the amount and frequency of drinking by the testee during a period between lab data.

For example, the deep learning neural network for estimating drinking habits may determine whether or not the testee maintains proper drinking habits, and display the result in an arbitrary value between 1-100.

In addition, the dietary estimation deep learning neural network estimates information such as blood sugar level and various nutrient levels of the test subject based on the received lab data, so that the test subject has performed proper eating habits to maintain a predetermined nutritional balance for a specific period of time. Can be estimated.

In addition, the deep learning neural network for estimating the eating habits can determine whether the testee performed proper eating habits and output the result as an eating habits evaluation index.

For example, the deep learning neural network for estimating eating habits may determine whether the testee maintains proper eating habits, and display the result in an arbitrary value between 1-100.

In addition, the exercise estimation estimating deep learning neural network may estimate the exercise amount of the examinee during the period between the lap data based on the received lap data, and output an exercise evaluation index that is an index proportional to the estimated exercise amount.

For example, the deep-learning neural network for estimating the amount of exercise may grasp the degree of the amount of exercise of the examinee and display the result as an arbitrary value between 1-100.

That is, when the exercise amount estimation deep learning neural network inputs the received lap data, the exercise amount evaluation index, which is a value calculated in proportion to the exercise amount of the examinee during the period between the lap data, may be output.

Finally, the change factor estimation deep learning neural network may output a change factor changed by a predetermined value or more by estimating a main change factor between the wrap data based on the received wrap data.

In detail, the change factor estimating deep learning neural network may compare the past lap data and the current lap data to detect/reduce the review blood elements that increase/decrease by a predetermined value or more or increase/decrease by a predetermined percentage or more for each blood element.

In other words, the deep learning neural network for estimating change factors differs between blood elements with high volatility and blood elements with low volatility, so blood elements with high volatility are detected as reviewed blood elements when there is a large change, and blood elements with small volatility Can be detected as a review blood element even when there are relatively small changes.

In addition, the deep learning neural network for estimating change factors can be extracted by extracting the change factors matched with the reviewed blood elements from a table or by learning by deep learning.

For example, the change factor estimation deep learning neural network can be extracted as a review blood element when HbA1c is relatively higher than a predetermined level (eg, fasting blood sugar 91 mg/dL, HbA1c 7.0%) compared to fasting blood glucose. In addition, the deep learning neural network for estimating change factors may output a carbohydrate meal, excess sugar intake, or a decrease in exercise amount, which is matched with a change factor to change the blood elements in the table.

In addition, the deep learning neural network for estimating change factors can be extracted as a reviewed blood element when HbA1c is relatively lower than a predetermined value (eg, 91 mg/dL of fasting blood sugar, 7.0% of HbA1c) compared to fasting blood glucose. In addition, the deep learning neural network for estimating the change factor may output fatty liver, excessive drinking or sleep deterioration as a change factor when the protein and fat are excessively consumed compared to carbohydrates when the above-mentioned blood element is extracted.

In addition, the deep learning neural network for estimating change factors, when fasting blood glucose and glycated hemoglobin increases significantly above a predetermined level, may stop taking diabetes drugs, exclude starting steroids, and output specific cancer occurrences as change factors. .

In this case, the method of learning the deep learning neural network for each life pattern item to perform each function will be described later in the detailed description of the deep learning neural network construction method for each life pattern item described below.

Subsequently, the data processing unit 220 that has obtained the output data for each life pattern item may combine the evaluation indexes for each life pattern item output from the deep learning neural network for each life pattern item and generate the life pattern information.

In addition, the data processing unit 220, application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers ( controllers), micro-controllers, microprocessors, and other electrical units for performing other functions.

Finally, the database 230 may store various data related to a blood test-based life pattern estimation service, and may include a deep learning neural network driving program built to perform deep learning.

The database 230 may be various storage devices such as ROM, RAM, EPROM, flash drive, hard drive, etc., and is a web storage that performs a storage function of the database 230 on the Internet. It might be.

-Medical consultation support service providing server

Next, the medical consultation assistance service providing server 300 according to an embodiment of the present invention, the life pattern information generated based on the blood test lab data of the testee from the terminal 100 and/or the life pattern estimation server 200 Can receive.

In addition, the medical consultation assistant service providing server 300 may provide medical consultation assistant content capable of outputting a graphic image based on the received life pattern information.

A detailed description of the medical consultation assistant content provided by the medical consultation assistant service providing server 300 will be described later in the detailed description of the deep learning neural network construction method for each life pattern item described below.

In addition, the medical consultation assistant service providing server 300 may provide a medical consultation assistant content search function capable of searching for various medical consultation assistant contents, and life pattern information received through the medical consultation assistant contents selected through the search Can output

The medical consultation assistance service providing server 300 may include a data communication unit, a memory, and a processor.

At this time, the data communication unit may exchange data with the terminal 100 and/or an external server to provide a blood test-based life pattern estimation service through a network.

In addition, the memory may store various data related to a blood test based life pattern estimation service.

For example, the memory may store and manage received blood test lab data, medical consultation auxiliary content, and the like.

In addition, the processor may control the overall operation of each of the above-described components in order to effectively provide the medical pattern-based life pattern estimation service by generating auxiliary contents for medical consultation.

These processors include application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers ( It may be implemented using at least one of micro-controllers, microprocessors, and electrical units for performing other functions.

-Life pattern estimation method based on blood test results

Hereinafter, a method of estimating a subject's life pattern based on blood test results will be described in detail with reference to the accompanying drawings.

4 is a flowchart illustrating a method for estimating life patterns based on blood test results according to an embodiment of the present invention.

Referring to FIG. 4, first, the terminal 100 may acquire and store first lab data according to the first blood test of the testee. (S101)

In detail, the terminal 100 may acquire and store first lab data that is various blood-related information derived from the blood of the testee through the first blood test.

In this case, referring to FIG. 5, the lab data may be information including at least one of blood name, test result, result determination, measurement unit, and reference value information.

Next, the terminal 100 may acquire and store the second lab data according to the second blood test of the testee after the first period has elapsed. (S103)

That is, the terminal 100 may acquire and store the first lap data and the second lap data measured over a first period, which is an arbitrary period.

In addition, the terminal 100 may acquire the life pattern information of the first period by inputting the obtained first lab data and second lab data into a deep learning neural network for each life pattern item. (S105, S107)

In detail, the terminal 100 transmits the acquired first and second lap data to the life pattern estimation server 200 to input the first lap data and the second lap data into a deep learning neural network for each life pattern item. Can.

Further, the terminal 100 may receive the life pattern information for the first period output from the deep learning neural network for each life pattern item receiving the first to second lap data from the life pattern estimation server 200.

For example, the terminal 100 may receive the drug use compliance evaluation index of the first period, which is output for the input of the first to second lab data, from the drug compliance deep learning neural network as life pattern information.

In addition, the terminal 100 may receive the drinking-related evaluation index of the first period, which is the output for the input of the first to second rap data, from the deep learning neural network for drinking habits as life pattern information.

Also, the terminal 100 may receive the first and second lab data from the dietary estimation deep learning neural network and receive the dietary evaluation index of the first period output as life pattern information.

In addition, the terminal 100 may receive the exercise amount evaluation index of the first period output by inputting the first to second lap data from the exercise estimation estimating deep learning neural network as life pattern information.

That is, the terminal 100 may receive life pattern information including an evaluation index for each life pattern item output from each deep learning neural network for each life pattern item from the life pattern estimation server 200.

In addition, the terminal 100 lives the weight and/or stress score information of the examinee who can be obtained based on lifestyle score information and lap data output from at least one item of drug compliance, drinking habits, eating habits, and exercise amount. It can be received by further including in the pattern information.

That is, the terminal 100 receives the life pattern information further including the weight and/or stress score information of the testee, and thus the drug compliance, the drinking habits, the eating habits, and/or the exercise amount, the fasting blood sugar, the glycated hemoglobin level, and/or Or, it can assist in making a more comprehensive estimation of the quality of sleep.

In summary, the terminal 100 may input various lab data obtained through blood tests into a deep learning neural network to estimate various changes in lifestyle habits of users, and provide it as a medical consultation supplementary content, to Can be utilized.

That is, the terminal 100 provides the lifestyle change between the previous treatment and the current treatment in a numerical or graph that is an accurate and intuitive indicator, so that the doctor can accurately catch the lifestyle change that the patient cannot understand or mention. Auxiliary, and through these data, doctors can recommend medications that fit the patient's lifestyle or provide medication guidance and guide the correct lifestyle, thereby improving the quality of care.

As described above, the terminal 100 obtains information on the subject's life pattern estimated by artificial intelligence data processing through a deep learning neural network learned with big data, thereby providing an automated, fast, and objective blood test result-based life pattern estimation service. Can provide.

In addition, the terminal 100 may input change data to the change factor estimation neural network to obtain change factors generated in the examinee during the first period.

In detail, the terminal 100 increases or decreases a predetermined value or more among blood elements examined in the lab data through the change factor estimation neural network, or when a difference of a predetermined value or more occurs in comparison with a plurality of blood elements, the Blood elements can be output as reviewed blood elements.

That is, when the lab data is input to the change factor estimation neural network, it can be learned based on the change in blood elements during the first period to output the reviewed blood elements.

For example, in the neural network for estimating the change factor, when HbA1c is relatively higher than a predetermined value compared to fasting blood glucose, when HbA1c is relatively lower than a predetermined value than fasting blood glucose, fasting blood glucose and glycated hemoglobin are relatively If it increases significantly above a preset value, Cr does not fluctuate significantly, but when it is higher than before BUN, when Cr continuously rises, when AST, ALT is 20 or more, when the AST/ALT ratio is less than 1, AST, ALT It can be output as a review blood factor when it is higher than the previous test value, when the AST/ALT ratio is 1 or more, when ALT, r-GTP rises above normal, or when HDL-Cholesterol is maintained below 40.

In addition, the terminal 100 may extract at least one change factor matched to the review blood element and provide it to the user.

Here, the change factors may include eating patterns, sleep disorders, exercise, acute disease occurrence, cancer occurrence, gastrointestinal disease, excessive drinking, weight change, fatty liver, acute infection, or biliary tract disease.

That is, the terminal 100 extracts a review blood element by inputting lab data for the first blood test and the second blood test into a change factor estimation neural network, and searches for a change factor matching the review blood element, It is possible to estimate and inform the subject of the change factors that may need to be assessed for a period of time.

Subsequently, the terminal 100 acquiring the life pattern information of the first period may display the acquired life pattern information or/and change factors of the first period. (S109) When displaying the change factor, the reviewed blood element may also be displayed.

In detail, referring to FIG. 6, the terminal 100 may output the life pattern information of the first period as a graphic image through the medical consultation assistant content.

In more detail, when the medical consultation assistance service providing server 300 receives the life pattern information of the first period from the life pattern estimation server 200, the first life pattern information may be displayed as various images, texts, and/or graphs. It is possible to generate medical consultation supplementary content that is displayed in various formats.

In addition, the terminal 100 may receive and output medical consultation auxiliary content representing life pattern information and medical consultation auxiliary content visualizing lab data from the medical consultation auxiliary service providing server 300.

As described above, the terminal 100 can easily and intuitively check the life pattern information by providing the life pattern information in the first period as a graphic image through the medical consultation assistant content.

In the above, it has been described that the terminal 100 transmits the life pattern information to the medical consultation assistant service providing server 300 to obtain and output the medical consultation assistant content, but the life pattern estimation server 200 and the medical consultation assistant service provider server Various embodiments may also be possible, such as the 300 being interlocked to generate the medical consultation auxiliary content and transmit it to the terminal 100.

In addition, referring to FIG. 7, the terminal 100 may display the life pattern information of the first period in combination with blood test blood information associated with each life pattern item of the first period.

In detail, the terminal 100 matches the blood of the blood to be tested for each item of drug compliance, drinking habits, eating habits, and/or exercise quantity of the first period to each item, and provides medical consultation supplementary content and life pattern information representing lab data. The displayed medical consultation auxiliary content can be combined and displayed.

In more detail, the blood component of blood, which is generally detected through a blood test, may be associated with each item of life pattern information.

For example, blood urea nitrogen (BUN, Blood Urea Nitrogen) contains information that reflects metabolites of proteins, and when blood urea nitrogen levels are high, kidney function disorders, gastrointestinal bleeding, etc. may be suspected. That is, the blood urea nitrogen may be blood test lab data associated with food habit items among items of life pattern information.

Accordingly, the terminal 100 may generate and provide medical counseling supplementary content that combines blood element nitrogen numerical changes of the first and second lab data with the food habit evaluation index among the life pattern information, and thus provides the combined medical counseling. The auxiliary content can improve the reliability of the estimated life pattern information.

As another example, gamma zitipi (γ-GTP) contains information indicating an indicator of liver and biliary tract function, and may indicate that there is jaundice or a high amount of drinking when it is higher than a normal level. In other words, the gamma zippi may be blood test lab data associated with a drinking habit item among each item of life pattern information.

Therefore, it is possible to generate and provide medical consultation supplementary content combining the gamma-graphicity value change of the first and second lab data and the drinking-related evaluation index among the life pattern information, and the combined medical consultation supplementary content is estimated. Reliability of life pattern information can be improved.

That is, the terminal 100 considers the association between blood test blood components and life pattern information provided by the blood test lab data, and provides information on the blood component changes to be matched for each item of the first life pattern information. It can be displayed together.

In this case, blood to be tested that are matched and displayed for each item of the first life pattern information may be displayed and provided in a graph format through the medical consultation assistant content.

As described above, the terminal 100 displays information on changes in blood components to be tested for each item of the life pattern information of the first period together, so that the life pattern information of the first period and the life pattern information of the first period are It is possible to output information on blood to be tested, which may be the cause of the derivation, and to provide life pattern information based on more detailed blood test results.

Meanwhile, the terminal 100 may acquire and store the third lab data according to the third blood test after the second period has elapsed. (S111)

That is, the terminal 100 may acquire and store the first wrap data or the second wrap data and the measured third wrap data over an arbitrary period.

In this way, the terminal 100 measures the first lap data, the second lap data, the third lap data of the examinee measured over an arbitrary period of time. , Based on the n-th lab data, it is possible to obtain and provide life pattern information for a period having a number of different cases.

Subsequently, the terminal 100 may acquire and store the life pattern information of the second period by inputting the obtained third lab data and the second lab data into the deep learning neural network for each life pattern item. (S113, S115)

In detail, the terminal 100 may transmit the acquired second to third lap data to the life pattern estimation server 200 to be input to the deep learning neural network for each life pattern item.

In addition, the terminal 100 may receive and store life pattern information for the second period output from the deep learning neural network for each life pattern item receiving the second to third lap data from the life pattern estimation server 200.

In addition, the terminal 100 may acquire and store the life pattern information of the third period by inputting the obtained third lab data and the first lab data into the deep learning neural network for each life pattern item. (S117, S119)

In detail, the terminal 100 may transmit the acquired first to third lap data to the life pattern estimation server 200 to be input to the deep learning neural network for each life pattern item.

In addition, the terminal 100 may receive and store the life pattern information for the third period output from the deep learning neural network for each life pattern item receiving the first to third lap data from the life pattern estimation server 200.

As described above, the terminal 100 obtaining the life pattern information of the first to third periods based on the first to third lap data may display the obtained life pattern information of the first to third periods in various ways. . (S121)

In this case, the terminal 100 may output the life pattern information of the first to third periods as a graphic image through the medical consultation assistant content generated from the medical consultation assistant service providing server 300.

In addition, referring to FIG. 8, the terminal 100 may display life pattern information according to the first to third blood tests 1) a method of displaying life pattern information in a third period T3, and 2) a first period (T1) ), a method for displaying life pattern information in the second period T2 and the third period T3, respectively, and 3) life pattern information in the first period T1 and life pattern information in the second period T2, respectively. After deriving and combining, it may be provided by outputting in any one or more of a method of performing a correction and displaying with the life pattern information of the third period T3.

For example, if the terminal 100 has a momentum evaluation index of 80 in the first period, an evaluation index of 90 in the second period, and an evaluation index of 75 in the third period, the terminal 100 first evaluates the exercise amount in the first period The index and the momentum evaluation index for the second period can be combined in a predetermined manner (eg, an average value).

Subsequently, the momentum evaluation index of the first to second periods (eg, 85) combined in a predetermined manner differs by at least a predetermined value (eg, 5) from the predetermined condition (eg, the momentum evaluation index of the third period). If it satisfies, etc.), the combined momentum evaluation index of the first to second periods is corrected according to a predetermined method (e.g., average value) as the third period momentum evaluation index to obtain a final momentum evaluation index (e.g., 80). Can be calculated.

As described above, the terminal 100 may display and provide the estimated life pattern information in various ways, so that the user can easily and intuitively check the life pattern information based on the blood test results from various viewpoints.

-How to build a deep learning neural network for each life pattern item

Hereinafter, a method of constructing a deep learning neural network for each living pattern item will be described in detail with reference to the attached drawing, focusing on the living pattern estimation server 200.

9 is a flowchart illustrating a method of constructing a deep learning neural network for each living pattern item according to an embodiment of the present invention.

Referring to FIG. 9, the life pattern estimation server 200 may first obtain the medical consultation content generated based on a blood test from the terminal 100. (S201)

In this case, the medical consultation content may be information generated by adding a record related to medical consultation input by a doctor on the medical consultation auxiliary content.

Here, the medical consultation auxiliary content is information that can be formed through a combination of at least one of medical related images, videos, animations, and text, and output through the terminal 100 to assist with medical consultation. EMR, Electronic Medical Record (EHR), Electronic Health Record (EHR), image to assist consultation provided by the medical consultation assistant service providing server 300 or blood test information provided from the terminal 100 (ie, lab Data).

In more detail, the electronic medical record is a computerized patient chart written on an existing paper, and the personal information of the patient (ie, the subject), past medical history, diagnosis record, treatment contents, medication history record, medications taken, surgery It may be data information including at least one of records, admission records, and outpatient care information.

In addition, the electronic health record is medical-related data in a digital format measured by a device (Divice) for acquiring data, and may include biometric information obtained from the device by a patient (ie, an examinee) wearing the device.

For example, the bioinformation is information obtained by measuring the body of a patient (ie, a subject) such as blood sugar, blood pressure, body temperature, or electrocardiogram, or a patient (ie, a subject), such as insulin input information, asthma inhalant input information, pain information, etc. Input or may include input information for pain.

That is, the medical consultation assistant content may include at least one or more lap data according to an electronic medical record, electronic health record, and/or blood test of the testee as described above in the embodiment of the present invention.

In detail, the medical consultation assistant content provided by the medical consultation assistant service providing server 300 may include audiovisual materials for assisting medical consultation, and in the embodiment, the audiovisual materials include blood test result output content, disease diagnosis content , Treatment method content, drug consultation content, treatment cost content, insurance information content, signature content, other content, and medical consultation supplementary content incorporating the above-mentioned contents.

For example, the medical consultation assistant content may include at least one image or information among blood test-based lab data, body organ images, disease information, treatment images, treatment information, drug medication methods, drug information, and insurance information. In addition, various information used to create medical consultation content may also be included as an aid.

Such various medical consultation assistant contents may be generated by the medical consultation assistant service providing server 300 and transmitted to the terminal 100, and the medical consultation according to the doctor's selection at the terminal 100 receiving the medical consultation assistant content It is displayed to be utilized, and the terminal 100 may provide a plurality of selected medical consultation supplementary contents to be easily arranged for use by a doctor.

In addition, in order to generate medical consultation contents by receiving medical consultation contents of a doctor on the medical consultation auxiliary contents, the terminal 100 may provide a consultation input interface.

In detail, the consultation input interface may include a graphic user interface that allows a doctor to input a handwriting, voice, editing, and/or image on an image of the displayed medical consultation auxiliary content.

In an embodiment, the consultation input interface may provide a handwriting input interface that detects a doctor's handwriting input on the image of the medical consultation assistant content.

In addition, in the embodiment of the present invention, the counseling input interface is based on blood test information by a high-level expert who satisfies a predetermined condition (for example, a specialist having a predetermined career or more) on the medical consultation assistant content. It is possible to provide an input interface for the skilled person to input.

In detail, the counseling input interface may provide an input interface of a skilled person who receives information on a subject's life pattern for a certain period of time estimated by a skilled person based on the blood test information of the testee output through the medical consultation auxiliary content. have.

That is, the terminal 100 may receive and store information on the life pattern estimated by the skilled person through the consultation input interface with respect to the blood test information of the testee provided through the medical consultation auxiliary content.

In addition, the terminal 100, the information generated by adding the medical consultation content (for example, life pattern information estimated by a skilled person based on blood test information, etc.) input through the consultation input interface on the aforementioned medical consultation auxiliary content In-person medical consultation content can be created.

In addition, the terminal 100 may output a generated medical consultation content and serve to assist in medical consultation, and may transmit the generated medical consultation content to the life pattern estimation server 200.

As described above, the terminal 100 according to an embodiment of the present invention can provide reliable feedback based on objective data by providing a blood test-based life pattern estimation service by utilizing medical consultation content. It can effectively assist the learning of deep learning neural networks.

Subsequently, the life pattern estimation server 200 that acquires the medical consultation content generated from the terminal 100 may extract life pattern related information from the medical consultation contents of the obtained medical consultation content. (S203)

At this time, the life pattern related information refers to life pattern information based on the blood test information of the testee existing on the obtained medical consultation content, that is, various information related to any one or more of the test subject's drug compliance, drinking habits, eating habits, and exercise items it means.

The information related to the living pattern may be extracted from a keyword recognition method through an optical character reading device (OCR) on medical consultation content, an automatic extraction method based on a predetermined keyword, and/or information input through an expert opinion interface, and the like. , It may be extracted by manually selecting the relevant medical personnel (eg, a doctor).

Subsequently, the life pattern estimation server 200 extracting the life pattern related information may generate a training data set by matching the extracted life pattern related information with the corresponding blood test information (ie, lab data). (S205)

At this time, the life pattern estimation server 200 may generate a training data set by automatically matching the life pattern related information and at least one or more corresponding blood test information, and the related medical personnel (for example, a doctor) manually related to the life pattern. A training data set may be generated based on information matching the information with at least one corresponding blood test information.

Next, the life pattern estimation server 200 that generated the training data set may classify the generated training data set according to the item characteristics of the life pattern related information. (S207)

In detail, the life pattern estimating server 200 sets the training data for each item according to the information of the life pattern related information item of the training data set, that is, the information of the test subject's drug compliance, drinking habits, eating habits, and/or exercise amount items. Can be classified separately.

At this time, the life pattern estimation server 200 may automatically perform classification according to the characteristics of the life pattern-related information items of the training data set, and the related medical personnel (eg, doctors) manually set the training data sets according to the item characteristics. It is also possible to classify the training data set based on the classified information.

The life pattern estimation server 200 that classifies the training data set as described above may train a deep learning neural network for each life pattern item based on the classified training data set. (S209)

In detail, the life pattern estimation server 200 may match and train the training data set classified by item characteristics and the deep learning neural network for each life pattern item by matching the related items.

In an embodiment, the life pattern estimation server 200 may match a training data set categorized as a drug compliance item with a drug compliance deep learning neural network, and learn a drug compliance deep learning neural network with the corresponding drug compliance training data set. I can do it.

In the same way, the life pattern estimation server 200 may match the drinking habit training data set and the drinking habit estimation deep learning neural network to train the deep learning neural network, and the eating habit training data set and the eating habit estimation deep learning neural network. By matching, the deep learning neural network can be trained, and the deep learning neural network can be trained by matching the momentum training data set with the momentum estimation deep learning neural network.

As described above, the life pattern estimation server 200 generates a training data set based on medical consultation content generated based on blood test information, and the generated training data set is applied to the items of the deep learning neural network for each life pattern item. By appropriately classifying and performing deep learning neural network learning, it is possible to more effectively perform learning of a deep learning neural network for each life pattern item based on a training data set and improve reliability.

As described above, the method and apparatus for estimating a life pattern based on blood test results according to an embodiment of the present invention acquires a plurality of blood test lab data for the test subject, and determines the test subject's life pattern for a specific period based on the acquired lab data. By estimating, there is an advantage in that it is possible to perform a medical examination by estimating the life pattern of the examinee for various periods.

In addition, the method and apparatus for estimating life patterns based on blood test results according to an embodiment of the present invention provide a function of mutually correcting the life patterns of test subjects estimated for various periods, thereby providing a life pattern estimation service based on blood test results. This has the effect of improving accuracy.

In addition, the method and apparatus for estimating life patterns based on blood test results according to an embodiment of the present invention are automated and fast and efficient blood test results by estimating life patterns of subjects through artificial intelligence data processing through a learned deep learning neural network. It is possible to provide a service based on estimation of life patterns.

In addition, the method and apparatus for estimating life patterns based on blood test results according to an embodiment of the present invention effectively generate a training data set by learning a deep learning neural network based on medical consultation content generated based on blood test results You can, and you can train deep learning neural networks based on more reliable data.

In addition, the method and apparatus for estimating life patterns based on blood test results according to an embodiment of the present invention display and provide the estimated life pattern information in various ways, so that the user can easily and intuitively receive life pattern information based on blood test results. Can be confirmed.

In addition, the embodiments according to the present invention described above may be implemented in the form of program instructions that can be executed through various computer components and can be recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded on the computer-readable recording medium may be specially designed and configured for the present invention or may be known and available to those skilled in the computer software field. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magneto-optical media such as floptical disks. medium), and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device can be changed to one or more software modules to perform the processing according to the present invention, and vice versa.

The specific implementations described in the present invention are exemplary embodiments, and do not limit the scope of the present invention in any way. For brevity of the specification, descriptions of conventional electronic configurations, control systems, software, and other functional aspects of the systems may be omitted. In addition, the connection or connection members of the lines between the components shown in the drawings are illustrative examples of functional connections and/or physical or circuit connections, and in the actual device, alternative or additional various functional connections, physical Connections, or circuit connections. In addition, unless specifically mentioned, such as “essential”, “importantly”, etc., it may not be a necessary component for application of the present invention.

In addition, the detailed description of the present invention has been described with reference to preferred embodiments of the present invention, but those skilled in the art or those of ordinary skill in the art will appreciate the spirit of the present invention as set forth in the claims below. And it will be understood that various modifications and changes can be made to the present invention without departing from the technical field. Therefore, the technical scope of the present invention should not be limited to the contents described in the detailed description of the specification, but should be defined by the claims.

Claims (5)

As a method of estimating the life pattern of the test subject by acquiring lap data according to the blood test of the test subject at the terminal,
Obtaining first lab data according to the first blood test;
Obtaining second lab data according to a second blood test after the first period;
Inputting the first lap data and the second lap data into a deep learning neural network for each life pattern item, and obtaining first period life pattern information output from the deep learning neural network for each life pattern item; And
And displaying the first period life pattern information,
The life pattern information includes life habit information on at least one item of the drug compliance, drinking habits, eating habits, exercise weight, weight, and stress of the test subject
A method of estimating life patterns and change factors based on blood test results. According to claim 1,
The obtaining of the first period life pattern information may include:
Drugs that are input to the drug compliance neural network by inputting the first lab data and the second lab data, and are indicators that indicate the degree of drug use compliance according to a medication map for the subject during the first period output from the drug compliance neural network Comprising the steps of obtaining a compliance evaluation index
A method of estimating life patterns and change factors based on blood test results. According to claim 1,
The obtaining of the first period life pattern information may include:
The first lap data and the second lap data are input to the deep learning neural network for drinking habit estimation, and drinking in proportion to the drinking amount and the number of drinking during the first period output from the deep learning neural network for drinking habit estimation And obtaining a relevant evaluation index.
A method of estimating life patterns and change factors based on blood test results. The method of claim 3,
In the displaying of the first period life pattern information,
And outputting the first period life pattern information through the medical consultation assistant content,
The medical consultation assistance content,
The method further comprises displaying blood test blood component change information associated with the blood test associated with a specific item of the life pattern information by combining it with a specific item of the life pattern information.
A method of estimating life patterns and change factors based on blood test results. According to claim 1,
The method further includes extracting a review blood element based on the first lab data and the second lab data, and outputting a change factor matched to the review blood element.
A method of estimating life patterns and change factors based on blood test results.

Images