KR20180058159A - Method and apparatus for predicting probability of the outbreak of a disease - Google Patents

Abstract

The present invention relates to a method and an apparatus for predicting occurrence of a disease. According to an embodiment of the present, the method for predicting occurrence of a disease comprises the steps of: receiving original data including a plurality of items from at least one external database; generating processed data representing one medical treatment or one medical examination as one event in accordance with a predetermined criterion on the basis of the original data; inputting the processed data into a disease occurrence prediction model; and calculating a disease occurrence probability for at least one disease using the disease occurrence prediction model. According to the present invention, by representing health-related data having different forms as one event, it is possible to input various data to a disease occurrence prediction model.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a method for predicting the onset of disease,

The present invention relates to a method and apparatus for predicting disease outbreaks, and more particularly, to a method and apparatus for predicting disease outbreaks in which probability of disease outbreak is calculated using received health-related data and a disease outbreak prediction model.

Recently, the incidence of diseases caused by increased consumption of instant food or fast food which is harmful to the body, lack of activity, and excessive work are greatly increasing. Particularly, the onset of cardiovascular diseases such as hypertension, ischemic heart disease, coronary artery disease and arteriosclerosis is increasing rapidly.

Accordingly, disease risk assessment is used to prevent and manage cardiovascular disease. A variety of clinical decision-making tools are used to assess disease risk. For example, the Framingham risk score is used. The Framingham risk score is an index for assessing the risk of cardiovascular disease through various cardiovascular risk factors such as sex, age, systolic blood pressure, smoking, diabetes, total cholesterol and HDL cholesterol . However, because of the high risk of recurrence in patients with a history of cardiovascular disease, the Farmingham risk score, which does not take into account the past history, is limited in measuring disease risk. Since the Farmingham risk score is developed in a foreign country, there is a need to correct for the Korean population according to the average disease incidence rate and risk exposure level in Korea. There is currently a risk - adjusted risk assessment tool for Koreans, but the criteria for selection of high - risk groups are lacking and they are not widely used clinically because they do not play a major role in high - risk screening.

[Related Technical Literature]

Periodontal disease prediction system and method for predicting periodontal disease using the same (Japanese Patent Laid-Open No. 10-2016-0083502)

Currently, in the medical industry, only one element is used to predict the onset of a disease, or only statistically based on a plurality of elements, and there is a limit to extract essential elements by filtering a plurality of elements. Therefore, by using the medical data of Koreans, it is possible to have a much higher accuracy if the elements extracted through machine learning are considered in a multi-dimensional form based on a plurality of elements included in the medical data, and furthermore, The onset prediction model can be implemented.

Disclosure of Invention Technical Problem [8] The present invention provides a method and apparatus for predicting a disease outbreak, which can input various data into a disease onset prediction model by representing health related data having different forms as one event.

Another problem to be solved by the present invention is to provide a method and apparatus for predicting disease outbreak that can improve the accuracy of the disease outbreak probability by inputting various received health related data into a disease predicting model.

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

According to an aspect of the present invention, there is provided a disease predicting method comprising: receiving original data including a plurality of items from at least one external database; A step of generating machining data representing one medical examination or one medical examination as one event, inputting the machining data into the disease occurrence prediction model, and calculating the probability of occurrence of the disease for at least one disease .

According to another feature of the invention, the disease is at least one of cardiovascular disease, gastric cancer, liver cancer, colon cancer, lung cancer, breast cancer, prostate cancer, dementia or diabetes, and the disease onset prediction model can be separately constructed for each disease. According to another aspect of the present invention, the step of receiving the original data includes receiving at least one of sociological data, medical record data including at least one medical examination, and health examination data including at least one medical examination Lt; / RTI >

According to still another aspect of the present invention, the step of generating processing data includes a step of, when a plurality of original data exists for one medical care day, integrating the original data into one event for one medical care day .

According to another aspect of the present invention, an event may include data on an administration drug classification code and a dosage amount.

According to another aspect of the present invention, the disease incidence prediction method may further include filtering an item related to the disease outbreak among a plurality of items.

According to another feature of the invention, there may be at least 50 items associated with the disease outbreak.

According to another aspect of the present invention, there is provided a method for generating processed data, comprising the steps of: determining whether a missing event in an event exists; if a missing event is present, Generating at least one, and inputting at least one of a representative value, an average value, or an interpolated value to the missing event.

According to another aspect of the present invention, the step of generating processed data includes the steps of: determining whether there is data missing in a plurality of items included in the event; if there is missing data, , Generating an average value or an interpolated value, and inputting at least one of the representative value, the average value, or the interpolated value to the missing data.

According to another aspect of the present invention, the step of generating the processing data includes the step of calculating the distribution based on the frequency of the length of the event, and the step of generating the processing data so as to include only the event corresponding to the predetermined threshold value in the distribution And the threshold value may be a length for an event located in the 95% region from left to right with respect to the center of the distribution.

According to still another aspect of the present invention, the step of generating processing data includes the steps of calculating an average and a standard deviation of data of a plurality of items included in an event, z-score, and inputting z-scores to the data of a plurality of items.

According to another aspect of the present invention, the step of generating the processing data may include extracting each unit corresponding to a plurality of items, and converting each unit into a unit defined in the processing data .

According to still another aspect of the present invention, the step of generating the processing data may include generating the processing data so as to include only a part of the data of the plurality of items.

According to still another aspect of the present invention, the step of calculating the disease occurrence probability may be a step of calculating at least one of the probability of occurrence of the disease or the probability of occurrence according to the type of the disease.

According to another aspect of the present invention, the method for predicting the onset of a disease may further include calculating a body age or an expected life using a disease onset prediction model.

According to an aspect of the present invention, there is provided an apparatus for predicting the onset of disease according to an embodiment of the present invention. The apparatus includes a communication unit configured to receive original data including a plurality of items from at least one external database, And a storage configured to store original data and processed data, the processor configured to input the processed data to the disease onset prediction model , And is configured to calculate a disease occurrence probability for at least one disease using a disease occurrence prediction model.

According to another aspect of the present invention, the communication unit may be configured to receive at least one of sociological data, medical record data including at least one medical examination, and health examination data including at least one medical examination.

According to another aspect of the present invention, a processor is configured to determine whether a missing event in an event is present, to generate at least one of a representative value, an average value, or an interpolated value for a missing event if a missing event is present, To input at least one of a representative value, an average value, or an interpolation value to a missing event.

According to another aspect of the present invention, a processor determines whether there is data missing from a plurality of items included in an event, and when there is missing data, the processor extracts a representative value, an average value, Generate at least one, and input at least one of the representative value, the average value, or the interpolation value to the missing data.

According to another aspect of the present invention, a processor is configured to calculate a distribution based on a frequency of a length for an event, and to generate processed data to include only events corresponding to a predetermined threshold value in the distribution, , And the length for the event located in the 95% region from left to right with respect to the center of the distribution.

According to another aspect of the present invention, a processor calculates an average and a standard deviation of data of a plurality of items included in an event, converts data of a plurality of items into a z-score using an average and a standard deviation , And to input the z-score to the data of a plurality of items.

According to another aspect of the present invention, the processor may be configured to extract each unit corresponding to a plurality of items, and to convert each unit into a unit defined in the processing data.

The details of other embodiments are included in the detailed description and drawings.

The present invention has the effect of providing a method and apparatus for predicting disease outbreak that can input various data into a disease onset prediction model by representing health related data having different forms as one event.

The present invention has an effect of providing a method and apparatus for predicting a disease outbreak that can increase the accuracy of a disease occurrence probability by variously processing received health data and inputting it into a disease onset prediction model.

The effects according to the present invention are not limited by the contents exemplified above, and more various effects are included in the specification.

1 is a schematic diagram for explaining a method for predicting a disease occurrence probability according to an embodiment of the present invention.
2 is a block diagram showing a schematic configuration of an apparatus for predicting the onset of diseases according to an embodiment of the present invention.
3 is a flowchart illustrating a procedure for calculating a disease occurrence probability according to a disease occurrence prediction method according to an embodiment of the present invention.
4A and 4B are schematic diagrams showing a process data table integrated into one event for one medical care day according to an embodiment of the present invention.
5A and 5B are schematic diagrams illustrating a processed data table that is input by calculating a missing event according to an embodiment of the present invention.
FIGS. 6A and 6B are schematic diagrams illustrating a processed data table obtained by calculating missing data according to an embodiment of the present invention.
FIGS. 7A and 7B are schematic diagrams showing a processed data table input by normalizing values of a plurality of items according to an embodiment of the present invention. FIG.
8A and 8B are schematic diagrams showing a processed data table in which values of a plurality of items are converted into defined units according to an embodiment of the present invention.
FIG. 9 illustrates a screen for providing a disease incidence probability according to an embodiment of the present invention.
FIGS. 10A and 10B show a screen for providing health findings and suitability for insurance.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention, and the manner of achieving them, will be apparent from and elucidated with reference to the embodiments described hereinafter in conjunction with the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims.

The shapes, sizes, ratios, angles, numbers, and the like disclosed in the drawings for describing the embodiments of the present invention are illustrative, and thus the present invention is not limited thereto. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. Where the terms 'comprises', 'having', 'done', and the like are used herein, other parts may be added as long as '~ only' is not used. Unless the context clearly dictates otherwise, including the plural unless the context clearly dictates otherwise.

In interpreting the constituent elements, it is construed to include the error range even if there is no separate description.

Although the first, second, etc. are used to describe various components, these components are not limited by these terms. These terms are used only to distinguish one component from another. Therefore, the first component mentioned below may be the second component within the technical spirit of the present invention.

Like reference numerals refer to like elements throughout the specification unless otherwise specified.

It is to be understood that each of the features of the various embodiments of the present invention may be combined or combined with each other partially or entirely and technically various interlocking and driving is possible as will be appreciated by those skilled in the art, It may be possible to cooperate with each other in association.

1 to 8B, the probability of the onset of the disease is described based on the probability of occurrence of cardiovascular disease, but the present invention is not limited thereto. The probability of the onset of diabetes can also be predicted by virtually the same process.

1 is a schematic diagram for explaining a method for predicting a disease occurrence probability according to an embodiment of the present invention.

Referring to FIG. 1, the disease occurrence probability providing system 1000 is a system for calculating the disease occurrence probability 300 by inputting the processed data 100 into the disease occurrence prediction model 200.

The processing data 100 is data obtained by processing original data received from an external database and is processed to include one event by integrating original data according to a predetermined criterion. The processing data 100 includes at least one event. An event is defined as a medical-related activity associated with the probability of disease outbreak. Here, the disease may be cardiovascular disease, cancer, dementia or diabetes. For example, an event can be defined as a clinic, prescription, or health checkup at a hospital. One event may include the care and prescription of the same person. At this time, the number of the processed data 100 and the number of events included in the processed data 100 are not limited.

 The disease onset prediction model 200 is a model for computing the result of the input data and calculating the result. At this time, the input data is the processed data 100, and the resultant value may be the disease occurrence probability 300. The disease occurrence predicting model 200 can receive a plurality of the processed data 100 and calculate the disease occurrence probability 300 corresponding to each of the plurality of processed data 100. [ Furthermore, the disease occurrence prediction model 200 can calculate one disease occurrence probability 300 for a plurality of the processed data 100 by calculating a plurality of the processed data 100.

The disease incidence probability (300) is a value for the probability of disease incidence, and is calculated by the disease incidence prediction model (200). At this time, the disease occurrence probability 300 may be a plurality of disease occurrence probability 300 corresponding to each of the plurality of processed data 100 and a disease occurrence probability 300 corresponding to the plurality of processed data 100 have.

Hereinafter, FIG. 2 will be referred to for a more detailed description of the disease onset prediction method in the disease occurrence probability prediction apparatus 400 implementing the disease onset prediction model.

2 is a block diagram showing a schematic configuration of a disease occurrence probability prediction apparatus according to an embodiment of the present invention. Will be described with reference to Fig. 1 for convenience of explanation.

Referring to FIG. 2, the disease occurrence probability prediction apparatus 200 includes a communication unit 210, a processor 220, and a storage unit 230.

The communication unit 210 of the disease occurrence probability prediction apparatus 400 is configured to receive original data including a plurality of items from at least one external database. Here, the external data may be the data of the health examination cohort database of the health insurance corporation, and the medical examination database of the medical institution. The health check-up cohort database contains data on the health insurance and medical claimant's entire medical treatment and treatment history, infectious disease history, and prescription history. In addition, the communication unit 210 can provide the calculated disease occurrence probability to the medical institution, the insurance company, and the individual.

The processor 420 of the disease occurrence probability predicting apparatus 400 is configured to generate processing data indicating one treatment or one health examination as one event in accordance with a predetermined criterion based on the original data. At this time, the processor 420 generates the processing data to increase the accuracy of the disease occurrence probability to be calculated. In particular, processor 420 may generate a missing event if there is a missing event of a plurality of events, or may generate missing data if there is missing data in an item included in the event have. Further, the processor 420 calculates the distribution based on the frequency of the length for the event, and generates the processed data so as to include only the event corresponding to the predetermined threshold value in the distribution. In this case, the threshold value is the length of the event located in the 95% region from the left to the right with respect to the center of the distribution. In addition, the processor 420 extracts each unit corresponding to a plurality of items, and converts each unit into a unit defined in the processed data. Further, the processor 420 inputs the processed data to the disease onset prediction model, and calculates the disease onset probability using the disease onset prediction model.

The storage unit 430 of the disease occurrence probability prediction apparatus 400 stores the received data and the generated data. More specifically, the storage unit 430 stores the processed data generated based on the original data and the original data received from the external database, and further stores the calculated disease occurrence probability.

Hereinafter, FIG. 3 will be referred to for a more detailed description of the disease onset prediction method in the disease occurrence probability predicting apparatus 400. FIG.

3 is a flowchart illustrating a procedure for calculating a disease occurrence probability according to a disease occurrence prediction method according to an embodiment of the present invention. For convenience of explanation, the components will be described with reference to FIG. 1 and FIG. 2 and reference numerals.

The communication unit 410 of the disease occurrence probability prediction apparatus 400 receives original data including a plurality of items from at least one external database (S310).

Specifically, the communication unit 410 receives at least one of sociological data, medical record data including at least one medical examination, and health examination data including at least one medical examination. Here, the sociological data is the health insurance qualification information of the health insurance subscribers and the medical benefit beneficiaries. It includes demographic information such as sex, age, residence area, death date, death related information including the cause of death, Economic status and other information, including types of health care, such as whether health care payments are made, and income quintiles and disability registration information. In addition, the medical record data refers to medical use history and medical fee occurrence details on the medical care benefit cost statement. The medical record data includes the details of medical treatment such as medical institution use information, medical care cost, medical treatment subject, medical illness information, medical examination, treatment, operation, other behavioral benefit details, and therapeutic materials. Table 1 shows the characteristics of the original data and field names in the external database.

Furthermore, the original data is used only in data of less than 80 years old, which does not have a history of disease or cancer, in the health examination cohort database of external databases. Because it receives a variety of original data, the problem of low localization, cultural characteristics, and environmental factors that are different from each other depending on the age is less accurate. There are advantages to be able to supplement.

Subsequently, the processor 420 generates processing data indicating one medical examination or one medical examination as one event in accordance with a predetermined criterion based on the original data (S320).

Specifically, the processor 420 constructs a plurality of items included in the original data as one event based on one medical examination or one medical examination, and generates processed data according to a predetermined criterion. For example, the processor 420 may classify items such as a personal serial number, a drug classification code, and a medication dose into one event by classifying the items according to the day of the day of the day of medical treatment, that is, one medical examination or one medical examination And generates processing data in accordance with a predetermined criterion. One event includes data on the dosing drug classification code and dosing dose. At this time, the processor 420 filters items related to the disease outbreak among a plurality of items included in the original data. For example, the processor 420 may filter items corresponding to a drug classifier code and a drug dosage amount associated with the disease. At this time, there are at least 50 items associated with the onset of the disease.

Further, in another embodiment, if there is a plurality of source data for one medical care day, the processor 420 may combine the source data into one event for one medical care day. For example, when there is a plurality of dosing dosage codes corresponding to each of the plurality of dosing drug classification codes and a plurality of dosing drug classification codes at a single medical care day, the processor 420 calculates a plurality of dosing drug classification codes and dosage amounts It can be integrated into one event corresponding to one medical care day.

Meanwhile, in another embodiment, the processor 420 determines whether there is a missing event among a plurality of events. If a missing event is present, the processor 420 generates at least one of a representative value, an average value, or an interpolated value for the missing event, and inputs at least one of the representative value, the average value, or the interpolated value to the event. For example, the processor 420 may perform a health checkup corresponding to 2003, 2005, or 2009, i.e., if there are three events, the event corresponding to 2004, 2006, 2007, and 2008 It is determined that the event is a missed event. Thus, the processor 420 generates at least one of a representative value, an average value, or an interpolated value for events for 2004, 2006, 2007, and 2008. Specifically, the processor 420 determines at least one of the representative values, average values, or interpolated values for the age, BMI, and blood pressure using the items included in the events of 2003, 2005, and 2009, One can be created. Then, the processor 420 inputs at least one of the generated representative value, the average value, or the interpolated value to the age, BMI, blood pressure item of the event corresponding to 2004, 2006, 2007, and 2008.

In various embodiments, the processor 420 determines whether there is data missing in the item included in the event. If missing data is present, the processor 420 generates at least one of a representative value, an average value, or an interpolated value for the missing data. For example, if it is determined that the data for the key in the 2006 event among the items included in the event of the patient in 2004, 2005 and 2006 has been lost, the processor 420 determines the key of the event in 2004 and 2005 And generates at least one of a representative value, an average value, and an interpolation value. Subsequently, the processor 420 inputs at least one of the generated representative value, the average value, or the interpolated value in the item for the key of the event of 2004 and 2005.

Meanwhile, in various embodiments, the processor 420 calculates the distribution based on the frequency of the length for the event, and generates the processed data to include only events corresponding to a predetermined threshold value in the distribution. In this case, the threshold value is the length of the event located in the 95% region from the left to the right with respect to the center of the distribution. When the number of events is large and the distribution of the event length is high, the accuracy with respect to time increases. As the accuracy of the time increases, the number of events may need to be adjusted according to the date distribution, since the scale of the processed data becomes larger and affects the probability of disease onset.

Further, in another embodiment, the processor 420 calculates the mean and standard deviation for the data of the plurality of items included in the event. Then, the processor 420 converts the data of a plurality of items into z-scores using the calculated average and standard deviation, and inputs the z-scores into the data of the plurality of items. By converting the data of a plurality of items included in the event into z-scores and inputting them, the processor 420 can normalize data for each item.

In another embodiment, the processor 420 extracts each unit corresponding to a plurality of items. For example, processor 420 extracts m and kg, which are units of key and weight. Subsequently, the processor 420 converts each unit into a unit defined in the process data. For example, if the units defined in the machining data are ft and lb, the processor 420 converts the units corresponding to the key and weight items from m to ft, kg to lb. That is, the processor 420 may convert the units corresponding to a plurality of items, thereby unifying the units in different cases for one item.

Subsequently, the processor 420 inputs the processed data to the disease onset prediction model (S330).

At this time, the processor 420 inputs the at least one processed data to the disease onset prediction model, which is an algorithm for calculating the disease onset probability. The processed data may include a plurality of events.

Next, the processor 420 calculates a disease occurrence probability using the disease occurrence prediction model (S340).

Here, the disease occurrence prediction model learns the input processing data by machine learning, and calculates the disease occurrence probability by applying the parameters determined as a result of the learning. At this time, the processor 420 may calculate a disease occurrence probability for each of a plurality of events included in the processed data, and may calculate a single disease occurrence probability integrated for a plurality of events included in the processed data . Furthermore, the processor 420 may also calculate the probability of onset according to the type of disease. In other words, the processor 420 may be used to determine the probability of suffering from hypertension, angina pectoris, myocardial infarction, stroke, stomach cancer, colon cancer, lung cancer, breast cancer, prostate cancer, dementia, diabetes or hypertension, angina pectoris, myocardial infarction, stroke, gastric cancer, , Breast cancer, prostate cancer, dementia, diabetes, and the like. A separate disease onset prediction model can be generated and used for each disease. A separate disease onset prediction model for each disease can be generated by an unrestricted manner and machine running. In addition, one disease incidence prediction model can be implemented to calculate multiple disease incidence probabilities. Further, the multiple disease disease prediction model can be implemented so as to calculate the probability of developing a disease. The probability of occurrence of the disease caused by the disease or the type of disease is provided to individuals, insurance companies, medical institutions, and health insurance corporations .

Accordingly, the disease occurrence probability predicting device 400 can calculate the disease occurrence probability with high accuracy based on the processed data considering various conditions by inputting the processed data obtained by processing the original data into the disease disease occurrence model.

4A and 4B illustrate a processed data table integrated into one event for one medical care day according to an embodiment of the present invention.

Referring to FIG. 4A, the original data table 510 includes a plurality of events for one treatment date 511, 512. For example, the original data table 510 includes two dosing goods classification codes 521 and dosage medication quantity 531 for the medical care date 511 corresponding to December 07, 2002. Therefore, the original data table 510 includes two rows corresponding to the medical care date 511 of December 07, 2002 in accordance with A043016 and A054502 drug classification code 521. At this time, the dosing medication dose 531 is included in the row corresponding to the medical care date 511 on December 07, 2002. Similarly, the original data table 510 includes two rows corresponding to the medical care date 512, which is December 21, 2002, in accordance with drug classification code 522 for A166503, A037008. At this time, the dosing medication dose 532 is also included in the row corresponding to the medical care date 512 (December 21, 2002).

Referring to FIG. 4B, the process data table 520 includes one event for one treatment date. For example, the processing data table 520 includes data on the date of care, that is, dosage medicines corresponding to each of the dosing drug classification codes, in one row. Specifically, the processing data table 520 includes an administration medication classification code 521 and an administration medication dosage 531 on a medical treatment date 511 of December 07, 2002, which is a medical treatment date. The processing data table 520 includes an administration medication classification code 522 and an administration medication dosage 532 on the medical examination date 512 of December 21, That is, the processing data table 520 includes a row for one event that combines a plurality of events corresponding to one medical care date.

Accordingly, the disease occurrence probability predicting device 400 integrates a plurality of original data for one medical care day to generate processed data as one event for one medical care date, so that a plurality of Features For example, the dosing drug classification code and dosage medication dosage can be expressed as one event.

FIGS. 5A and 5B are diagrams illustrating a processed data table that is input by calculating a missing event according to an embodiment of the present invention.

Referring to FIG. 5A, the original data table 610 includes events 611, 612, and 613 such as age, blood sugar, and BMI according to individual serial numbers. For example, the original data table 610 includes 2003 events 611, 2005 events 612, and 2009 events 613 of the same private serial number.

Referring to FIG. 5B, the machining data table 620 includes a missing event 621 generated based on the 2003 event 611, the 2005 event 612, and the 2009 event 613. For example, processing data 620 includes a missing event 621 corresponding to 2004, 2006, 2007, and 2008. At this time, the missing event 621 corresponding to 2004, 2006, 2007, and 2008 indicates the age, blood glucose, and BMI of the 2003 event 611, the 2005 event 612, and the 2009 event 613 An average value, or an interpolation value generated based on the representative value, the average value, or the interpolation value.

Accordingly, the disease occurrence probability prediction apparatus 400 generates at least one of a representative value, an average value, and an interpolation value with respect to a missed event to generate processed data, thereby expanding the data to be input to the disease occurrence prediction model, Can be increased.

FIGS. 6A and 6B show a processed data table that is calculated by calculating missing data according to an embodiment of the present invention.

Referring to FIG. 6A, the original data table 710 includes data for a plurality of events according to one personal serial number. At this time, the plurality of events include a plurality of items, and the missing data 711 may exist in the data corresponding to the plurality of items. Therefore, the original data table 710 can receive the missing data 711 generated based on the data of a plurality of items according to one personal serial number. The missing data 711 is at least one of a representative value, an average value, and an interpolation value generated based on data of a plurality of items based on one personal serial number.

Referring to FIG. 6B, the processing data table 720 includes data for a plurality of events according to a plurality of personal serial numbers. At this time, the missing data 721 may exist in the data corresponding to the plurality of items included in the plurality of events. Therefore, the machining data table 720 can receive the missing data 721 generated based on the data of a plurality of items according to the plurality of personal serial numbers. That is, the machining data table 720 can receive the missing data 721 in at least one of a representative value, an average value, or an interpolation value generated based on the data of a plurality of others.

Accordingly, the disease occurrence probability predicting device 400 generates at least one of a representative value, an average value, and an interpolation value to the data missing based on the individual's data or the other's data, thereby generating the processed data, The data to be input can be expanded to increase the accuracy of the disease occurrence probability.

FIGS. 7A and 7B illustrate a processed data table in which values of a plurality of items are normalized and input according to an embodiment of the present invention.

Referring to FIG. 7A, the original data table 810 includes a plurality of events according to a personal serial number. At this time, a plurality of events include a plurality of items such as BMI, systolic blood pressure, and diastolic blood pressure, and a plurality of items are inputted as numerical values of different units. For example, BMI is entered in kg / m ^2, systolic and diastolic blood pressure values are given in mmHg.

Referring to Fig. 7B, the machining data table 820 includes numerical values converted into z-scores for a plurality of items. In this case, the values converted into z-scores are calculated from the mean and standard deviation of numerical values of different units. That is, the machining data table 820 may include z-score converted numerical values, which are the same values as numerical values of different units corresponding to a plurality of items, as one unit, in a plurality of items.

Accordingly, the disease occurrence probability predicting device 400 converts a plurality of items of different units into z-scores, thereby applying the same reference value to a plurality of items to more easily recognize items that affect the disease occurrence probability .

8A and 8B illustrate a processed data table in which values of a plurality of items are converted into defined units according to an embodiment of the present invention.

Referring to FIG. 8A, the original data table 910 includes a plurality of events according to a personal serial number. At this time, the plurality of events includes a plurality of items of the key, the weight, the current smoking period, the average daily smoking amount, and the once drinking amount. At this time, numerical values corresponding to one item can be input in different units. For example, the key is cm, ft, the weight is kg, lb, the current smoking period is 5 years, the year is 1 year, the current average daily smoking amount is half a unit, Can be input.

Referring to FIG. 8B, the machining data table 920 includes numerical values of the same unit in one item. For example, the machining data table 920 includes numerical values corresponding to the items of a key in cm, a weight in kg, a current smoking period in a year, a current average daily smoking amount in a unit of cow, do.

Accordingly, the disease occurrence probability predicting device 400 generates numerical values of different units in a single item with numerical values of the same unit, so that original disease data whose disease occurrence predicting model is composed of numerical values of different units are also input So that it is possible to calculate the probability of disease occurrence with high accuracy based on a variety of data.

FIG. 9 illustrates a screen for providing a disease incidence probability according to an embodiment of the present invention.

Referring to FIG. 9, the disease occurrence probability providing screen 1100 may include a disease occurrence probability item 1110, a disease occurrence probability item 1120, and a current user location item 1130.

Specifically, the disease occurrence probability providing screen 1100 provides a disease occurrence probability item 1110 for each year calculated based on past health examination data, past medical history item data, and past medical history data classified by time series. For example, the probability of disease outbreak screen 1100 may provide a probability of disease outbreak in 2017 corresponding to the past, 2016 corresponding to the present, and 2017 corresponding to the present. In addition, the disease occurrence probability providing screen 1100 provides a disease occurrence probability, that is, a disease occurrence probability item 1120, depending on the type of disease. For example, the probability of disease occurrence screen 1100 can be used to determine the probability of occurrence of cardiovascular diseases such as hypertension, angina pectoris and arteriosclerosis, the probability of developing cancer diseases such as gastric cancer, colorectal cancer and liver cancer, You can provide a percentage. In addition, the disease occurrence probability providing screen 1100 displays the probability that the current user belongs to the population based on the calculated disease occurrence probability, how many percent the disease occurrence probability, the percentile is the percentage, and the health status of the current user The converted score may provide a current user's location item 1130 for how many points there are. For example, the disease occurrence probability providing screen 1100 can provide 190 million, 80%, and 90 points out of the total population of 2.38 million, which calculates the probability of disease occurrence for the current user's location. Furthermore, the disease occurrence probability providing screen 1100 may provide the location of the user by year according to the disease occurrence probability.

Accordingly, the onset prediction server 200 provides the user's disease occurrence probabilities for each disease type such as year, cardiovascular disease, cancer, dementia, and diabetes, and provides the position of the user according to the disease occurrence probability, To be able to recognize the onset information and to make it easier for insurers and medical institutions to make health findings.

FIGS. 10A and 10B show a screen for providing health findings and suitability for insurance.

Referring to FIG. 10A, the health finding providing screen 1200 may include a disease occurrence probability item 1210 and a health finding item 1220.

Specifically, the health finding providing screen 1200 provides a disease occurrence probability item 1210, which is a probability of occurrence according to each disease such as hypertension, arteriosclerosis, stroke, cerebrovascular disease, and the like. For example, the screen 1200 for providing health information includes a probability of developing hypertension of 70%, a probability of developing angina pectoris of 50%, a probability of developing atherosclerosis of 80%, a probability of developing gastric cancer of 20% The probability of developing cancer is 15%, the probability of developing liver cancer is 10%, the probability of developing dementia is 30%, and the probability of developing diabetes is 50%. In addition, the health finding providing screen 1200 can also provide elements for increasing the probability of disease occurrence. For example, the health finding providing screen 1200 may provide items for blood pressure, body fat, HDL cholesterol and LDL cholesterol, and numerical values for each item. At this time, different visual effects can be provided for the factors that increase the probability of the disease depending on the degree of influence on the probability of disease onset. In other words, the health finding providing screen 1200 displays a slanting line to the left in the factors that increase the probability of disease occurrence, a slanting line in the right direction on the factors having an average influence on the disease occurrence probability, A plurality of dot displays and the like can be provided to the elements. In addition, the health finding providing screen 1200 provides the health finding item 1220 determined based on the disease occurrence probability item 1210. Health findings are comments made by referring to the factors causing the disease and the probability of disease occurrence by disease. At this time, since the health findings are processed in a natural language, the health finding providing screen 1200 may be processed in a natural language to provide a judgment on the health status of the determined user. That is, the health finding providing screen 1200 may provide whether the health findings are positive or negative. In addition, the health finding providing screen 1200 provides a send button 1230 for transmitting health findings to the onset prediction server 200. Accordingly, when receiving the selection signal for the send button 1230, the health findings are transmitted to the onset prediction server 200.

Referring to FIG. 10B, the insurance adequacy providing screen 1200 may include a disease occurrence probability item 1210 and an insurance compliance item 1240. The insurance coverage suitability screen including the specific disease risk probability item 1210 is the same as that described with reference to FIG. 6A, and therefore, description thereof is omitted.

Specifically, the insurance adequacy providing screen 1200 provides an insurance compliance item 1240 determined based on health findings in the onset prediction probability server 200. The compliance eligibility item (1240) is a comment that includes whether the user is eligible to insure based on the health findings generated according to the determined disease outbreak probability. Furthermore, the insurance coverage suitability screen 1200 can also provide a numerical score for the adequacy of the insurance coverage.

Accordingly, the onset prediction server 200 provides a probability of disease occurrence according to a disease causing factor as well as a probability of occurrence of a disease, so that it is possible to determine the probability that a user will have an onset of a disease, To recognize the specific disease probability of how much it is. In addition, the onset prediction server 200 provides insurance suitability based on health findings so that the insurer can objectively determine whether or not the insurer's insurance is appropriate, thereby making it easier to calculate the profitability of the insured.

In this specification, each block or each step may represent a part of a module, segment or code that includes one or more executable instructions for executing the specified logical function (s). It should also be noted that in some alternative embodiments, the functions mentioned in the blocks or steps may occur out of order. For example, two blocks or steps shown in succession may in fact be performed substantially concurrently, or the blocks or steps may sometimes be performed in reverse order according to the corresponding function.

The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software module may reside in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, a register, a hard disk, a removable disk, a CD-ROM or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, which is capable of reading information from, and writing information to, the storage medium. Alternatively, the storage medium may be integral with the processor. The processor and the storage medium may reside within an application specific integrated circuit (ASIC). The ASIC may reside within the user terminal. Alternatively, the processor and the storage medium may reside as discrete components in a user terminal.

Although the embodiments of the present invention have been described in detail with reference to the accompanying drawings, it is to be understood that the present invention is not limited to those embodiments and various changes and modifications may be made without departing from the scope of the present invention. . Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. Therefore, it should be understood that the above-described embodiments are illustrative in all aspects and not restrictive. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

100: Process data
200: Predictive model of disease outbreak
300: probability of disease occurrence
400: Probability of disease occurrence prediction device
410:
420: processor
430:
510, 610, 710, 810, 910: original data table
511, 512: Date of medical examination
520, 620, 720, 820, 920:
521, 522: Drug classification code
531, 532: Dosage of medication
611, 612, 613: Event
621: Missing event
711, 721: Missing data
1000: Probability of disease occurrence providing system
1100: Probability of disease occurrence screen
1110: Probability of disease occurrence by year
1120: probability of disease occurrence item
1130: Current user's location entry
1200: Healthy Providing Screen
1210: Occurrence probability item by disease
1220: Healthy items
1230: Send button
1240: Insurance Eligibility Items

Claims (22)

Receiving original data including a plurality of items from at least one external database;
Generating processing data indicating one medical examination or one medical examination as one event according to a predetermined criterion based on the original data;
Inputting the processed data to a disease onset prediction model; And
And calculating a disease incidence probability for at least one disease using the disease incidence prediction model. The method according to claim 1,
Wherein the disease is at least one of cardiovascular disease, stomach cancer, liver cancer, colon cancer, lung cancer, breast cancer, prostate cancer, dementia or diabetes, and the disease onset prediction model is separately constructed for each of the diseases. The method according to claim 1,
Wherein the step of receiving the original data comprises:
Sociological data, at least one of the medical care record data including the medical care, and at least one of health examination data including at least the one health care examination. The method according to claim 1,
Wherein the step of generating the processed data comprises:
When a plurality of the original data exists for one medical care day,
Further comprising integrating the original data into one event for the one care date. The method according to claim 1,
Wherein the one event includes data on an administration drug classification code and a dosage amount. The method according to claim 1,
Further comprising the step of filtering an item associated with the disease out of the plurality of items. The method according to claim 6,
The method of predicting the onset of a disease, wherein the item associated with the onset of the disease is at least 50 individuals. The method according to claim 1,
Wherein the step of generating the processed data comprises:
Determining whether a missing event is present in the event;
If the missing event is present,
Generating at least one of a representative value, an average value, or an interpolation value for the missing event; And
And inputting at least one of the representative value, the average value, or the interpolation value to the missing event. The method according to claim 1,
Wherein the step of generating the processed data comprises:
Determining whether there is data missing in the plurality of items included in the event;
If the missing data is present,
Generating at least one of a representative value, an average value, or an interpolation value for the missing data; And
And inputting at least one of the representative value, the average value, or the interpolation value to the missing data. The method according to claim 1,
Wherein the step of generating the processed data comprises:
Calculating a distribution based on the frequency of the length for the event; And
And generating the processed data so as to include only events corresponding to a predetermined threshold value in the distribution,
Wherein the threshold value is a threshold value,
Wherein the length of the event is located in a 95% region from left to right with respect to the center of the distribution. The method according to claim 1,
Wherein the step of generating the processed data comprises:
Calculating an average and a standard deviation of data of the plurality of items included in the event;
Converting the data of the plurality of items into a z-score using the average and standard deviation; And
And inputting the z-score to the data of the plurality of items. The method according to claim 1,
Wherein the step of generating the processed data comprises:
Extracting each unit corresponding to the plurality of items; And
And converting each of the units into units defined in the processed data. The method according to claim 1,
Wherein the step of generating the processed data comprises:
And generating the processed data so as to include only a part of data of the plurality of items of data. The method according to claim 1,
The step of calculating the disease occurrence probability comprises:
And calculating at least one of a probability of the disease occurring or a probability of occurrence of the disease according to the type of the disease. The method according to claim 1,
Further comprising the step of calculating a body age or an expected life span using the disease onset prediction model. A communication unit configured to receive original data including a plurality of items from at least one external database;
A processor configured to generate processing data indicating one medical examination or one medical examination as one event according to a predetermined criterion based on the original data; And
And a storage unit for storing the original data and the processed data,
The processor comprising:
Inputting the processed data into a disease onset prediction model,
Wherein the probability of disease occurrence for at least one disease is calculated using the disease occurrence prediction model. 17. The method of claim 16,
Wherein,
Sociological data, medical record data including at least the one medical examination, and health examination data including at least the one medical examination. 17. The method of claim 16,
The processor comprising:
Determines whether there is a missing event in the event,
If the missing event is present,
Generating at least one of a representative value, an average value, and an interpolation value for the missing event,
And to input at least one of the representative value, the average value, or the interpolation value to the missing event. 17. The method of claim 16,
The processor comprising:
Determining whether there is data missing in the plurality of items included in the event,
If the missing data is present,
Generating at least one of a representative value, an average value, and an interpolation value for the missing data,
And to input at least one of the representative value, the average value, or the interpolation value to the missing data. 17. The method of claim 16,
The processor comprising:
Calculating a distribution based on the frequency of the length of the event,
And generate the processed data so as to include only events corresponding to a predetermined threshold value in the distribution,
Wherein the threshold value is a threshold value,
Wherein the length of the event is located in the 95% region from left to right with respect to the center of the distribution. 17. The method of claim 16,
The processor comprising:
Calculating an average and a standard deviation of data of the plurality of items included in the event,
Converting the data of the plurality of items into a z-score using the average and standard deviation,
And inputting the z-score to the data of the plurality of items. 17. The method of claim 16,
The processor comprising:
Extracting each unit corresponding to the plurality of items,
And convert each of the units into a unit defined in the processed data.

Images