JP4499503B2 - Onset risk calculation system and program - Google Patents

Description

  The present invention relates to a system and the like for calculating a risk of developing a disease using health checkup data stored and stored in a database.

In recent years, with the advent of an aging society, public interest in health management has increased, and various systems have been devised for the purpose of health promotion and health management.
For example, the data from the past and present of the subject's health examination is indexed as the degree of health based on the comparison between the health standard model and various disease models, and the past and future of the individual's health status are displayed visually. There is also a system (see Patent Document 1).
Japanese Patent Laid-Open No. 2002-63278 (page 5, FIG. 8)

When calculating the incidence of disease using the health checkup data of the evaluator, the conventional system as described above predicts the future health based on the increasing / decreasing tendency of the value (health) indexed by the screening data. is doing.
However, it is known that the degree of obesity in the 30s to 40s is greatly involved in the onset of diabetes, for example, in the case of such a disease in which the health condition in a certain past period greatly affects the probability of later onset However, an accurate onset probability cannot be obtained by simply predicting the future onset probability based on the tendency to increase or decrease in health.
In addition, some diseases such as lifestyle-related diseases can reduce the risk of onset depending on the lifestyle improvement efforts of the person, and such efforts to improve the health condition of the person greatly affect the probability of later onset. With regard to diseases as well, an accurate onset probability cannot be obtained by simply predicting the future onset probability based on the tendency to increase or decrease in health.

The present invention has been made in view of the above circumstances, and provides an onset risk calculation system and the like for more accurately calculating the onset probability of a disease using stored and stored personal health checkup data. Objective.
Another object of the present invention is to provide an onset risk calculation system and the like suitable for calculating the onset probability of a disease whose health condition in a certain past period greatly affects the later onset probability.
Another object of the present invention is to provide an onset risk calculation system suitable for calculating the onset probability of a disease whose actual health improvement results greatly affect the later onset probability.

In order to achieve the above object, an onset risk calculation system according to the first aspect of the present invention includes:
And the evaluator ID which can specify the evaluator, response and medical examination data showing the results of medical examination of該被evaluator at multiple time points, and date of birth of該被evaluator, to each said object evaluator A database to store and store,
Of the disease ID that can identify the disease to be evaluated and the factors described in the health check data, when calculating the probability of developing the disease to be evaluated from the time when the health check is performed to a predetermined time Storage means for storing factor information for specifying a factor used for the weighting function for acquiring a weight value for each age in association with each disease;
In response to designation of the evaluated person and the disease to be evaluated by inputting the evaluated person ID and the disease ID, the plurality of points in time corresponding to the evaluated person ID reads and said and said health examination data birth from the database, reading means for reading out said factor information and the function weighting corresponding to the disease ID from the storage means,
A model formula having a factor designated by the factor information read by the reading unit as a parameter is generated, and the age of the evaluated person at the plurality of time points at which the health examination was performed from the date of birth is respectively determined. For each health check data at the plurality of time points, by specifying the factor indicated by the factor information in the health check data in the model formula , the predetermined health check data is obtained at each age at which the health check was performed. An onset probability calculating means for calculating the probability that the evaluated person will develop the disease to be evaluated by the time point ,
The sum of values obtained by multiplying the onset probability of the disease at each age calculated by the onset probability calculating means by the weight value for each age acquired from the weighting function read by the reading means, respectively. by calculating the a risk calculating means for determining the risk of the evaluation of diseases of the object evaluator,
It is characterized by providing.

A program according to the second aspect of the present invention is
Corresponds to each person to be evaluated, the person to be evaluated who can identify the person to be evaluated, the health examination data indicating the result of the health examination of the person to be evaluated at a plurality of time points, and the date of birth of the person to be evaluated. A database for accumulating and storing the disease, a disease ID that can identify the disease to be evaluated, and a disease described in the health check data from the time when the health check is performed to a predetermined time Storage means for storing factor information for specifying a factor used in calculating the probability of developing the disease and a weighting function for acquiring a weight value for each age in association with each disease Computer
In response to designation of the evaluated person and the disease to be evaluated by inputting the evaluated person ID and the disease ID, the plurality of points in time corresponding to the evaluated person ID the health examination data and reads the said date of birth from the database, reading means for reading the said factor information and the function weighting corresponding to the disease ID from the storage means,
A model formula having a factor designated by the factor information read by the reading unit as a parameter is generated, and the age of the evaluated person at the plurality of time points at which the health examination was performed from the date of birth is respectively determined. For each health check data at the plurality of time points, by specifying the factor indicated by the factor information in the health check data in the model formula , the predetermined health check data is obtained at each age at which the health check was performed. An onset probability calculating means for calculating the probability that the evaluated person will develop the disease to be evaluated by the time point ,
The sum of values obtained by multiplying the onset probability of the disease at each age calculated by the onset probability calculating means by the weight value for each age acquired from the weighting function read by the reading means, respectively. by calculating a risk calculating means for determining the risk of the evaluation of diseases of the object evaluator,
To function as.

  According to the present invention, it is possible to calculate the onset probability of a disease more accurately using the personal health checkup data.

Hereinafter, an onset risk calculation system according to an embodiment of the present invention will be described with reference to the drawings.
A configuration example of the onset risk calculation system according to the embodiment of the present invention is shown in FIG. The onset risk calculation system includes, for example, an evaluated person information DB (database) 1, a risk-related DB 2, a management server 3, and a client terminal 4, and these devices can be connected via a network 10. It is configured. The network 10 includes various networks such as a LAN (local area network) and the Internet.

  The evaluated person information DB 1 stores a basic information table and a screening result table. The basic information table stores basic information such as an evaluated person ID, name, date of birth, and sex as shown in FIG. Further, in the examination result table, for example, data (health examination data) indicating the result of the health examination received by the person to be evaluated is stored for each person to be evaluated every year when the examination is performed. For example, as shown in FIG. 3, the health checkup data includes an evaluated person ID, physical measurements such as height, weight, BMI, and body fat percentage, history, smoking / drinking, visual acuity, blood pressure, blood test, urine Data on health conditions such as examinations are registered for each date of health examination.

  The risk-related DB 2 stores a disease name table, a factor / coefficient table, and a weight table. The disease name table associates disease IDs with disease names. In the factor / coefficient table, for example, as shown in FIG. 4, the factor information indicating the factor name and coefficient of each factor (parameter) used in the risk function for calculating the disease risk (disease probability) is shown for each disease ( Stored for each disease ID). In the weight table, a weighting function set in advance for each disease according to a disease characteristic or the like is set in association with the disease ID.

  As shown in FIG. 5, for example, the management server 3 includes a computer including a control unit 31, a storage unit 32, an input unit 33, a display unit 34, and a communication control unit 35.

  The control unit 31 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. The control unit 31 reads and executes an operation program or the like stored in advance in the storage unit 32. The input unit 311 and the onset risk calculation unit 312 are logically realized.

  The data input unit 311 accepts input of basic information, health checkup data, and the like of each evaluated person from the input unit 33 and the client terminal 4, and registers various input data in the evaluated person information DB1.

The onset risk calculation unit 312 accepts input for designating the evaluated person and the disease to be evaluated from the client terminal 4, and reads the health check data and basic information of the specified evaluated person from the evaluated person information DB 1. The factor information and the weighting function corresponding to the designated disease are read from the risk-related DB 2. Then, the read factor information is applied to a risk model formula (see Equation 1) preset in the storage unit 32 or the like, thereby generating a risk model formula for the disease to be evaluated. Note that S _t (i) is a model formula created based on a survival time analysis method such as a cox proportional hazard model or a logistic model.

(Equation 1)
If the disease onset probability from time point i (= 1, 2,..., N) to time point t (t> i) is S _t (i),

The onset risk calculation unit 312 calculates the onset probability (temporary onset probability) at each time point by substituting the health diagnosis data at each time point of the evaluated person into the generated risk model formula for the disease to be evaluated. To do. Note that the value of t may be received from the input unit 33 or the client terminal 4, or a value set in advance in the storage unit 32 or the like may be used. Next, the onset risk calculation unit 312 calculates the sum of each value weighted using the weighting function for the temporary onset probability S _t (i) at each time point according to the calculation formula shown in (Expression 2). The final onset probability (true onset risk) _St is output to the requesting client terminal 4. In the weighting function, the sum of the weight values at each time point is 1.

(Equation 2)

  An example of the weighting function used is shown in (Equation 3). The weighting function is selected according to the characteristics of each disease and set in the risk-related DB 2.

(Equation 3)

  In addition, the output format of a calculation result is arbitrary, For example, you may output the value of the temporary onset risk and true onset risk of each time point with numerical data and / or graph data.

The storage unit 32 is composed of, for example, a hard disk device or the like, and stores an operation program to be executed by the control unit 31 and various data necessary for processing (for example, a risk model formula, a true onset risk calculation formula, etc.) and the like. Is done.
The input unit 33 is composed of an input device such as a keyboard and a pointing device, for example, and transmits input data to the control unit 31.
The display unit 34 is constituted by a display device such as a display device, for example, and outputs various data according to instructions from the control unit 31.
The communication control unit 35 controls communication with other devices via the network 10 in accordance with instructions from the control unit 31.

For example, as shown in FIG. 6, the client terminal 4 includes a computer including a control unit 41, a storage unit 42, a display unit 43, an input unit 44, a communication control unit 45, and the like. The client terminal 4 may include a mobile terminal such as a mobile phone or a PDA, for example.
The control unit 41 requests the management server 3 to calculate the risk of developing the disease for the person to be evaluated by reading and executing an operation program or the like stored in advance in the storage unit 42. Specifically, the control unit 41 receives input of information specifying the evaluated person and the disease to be evaluated, such as a disease ID and an evaluated person ID, from the input unit 44, and calculates input data and an onset risk. Is transmitted to the management server 3 via the network 10. Then, data on the calculation result of the onset risk is received from the management server 3 and displayed on the display unit 43.
For example, the control unit 41 receives input of a disease ID and health checkup data, and transmits a registration request for input data and health checkup data to the management server 3 via the network 10.

The storage unit 42 is composed of, for example, a hard disk device or the like, and stores an operation program to be executed by the control unit 41 and various data necessary for processing.
The display unit 43 is constituted by a display device such as a display device, for example, and outputs various data according to instructions from the control unit 41.
The input unit 44 is composed of an input device such as a keyboard and a pointing device, for example, and transmits input data to the control unit 41.
The communication control unit 45 controls communication with other devices via the network 10 in accordance with instructions from the control unit 41.

Next, the onset risk calculation process by the management server 3 will be described with reference to the flowchart of FIG.
For example, the control unit 41 of the client terminal 4 is input in response to input of the evaluated person ID of the evaluated person A, the disease ID of the disease B, and the calculation instruction for the onset risk from the input unit 44. The evaluation subject ID of the evaluation subject A and the disease ID of the disease B are transmitted to the management server 3 together with the calculation request for the onset risk.
When the onset risk calculation unit 312 of the management server 3 receives, from the client terminal 4, the to-be-evaluated person A's ID, the disease ID of the disease B, and the onset risk calculation request (step S1), The health examination data and basic information for n time points registered in association with the evaluator A's evaluator ID are read from the evaluator information DB1 (step S2), and are associated with the disease ID of disease B. The factor information and the weighting function registered in the above are read from the risk-related DB 2 (step S3).

Next, the onset risk calculation unit 312 generates a risk model formula for disease B by applying the factor information of disease B to a preset risk model formula (see Equation 1), and the risk model formula and health Using each diagnosis data, for each time point i (= 1, 2,..., N) at which a health check is performed, an onset probability S _t (i) from time point i to time point t (> i) is calculated. (Step S4).

  Next, the onset risk calculation unit 312 applies the weighting function of the disease B and the calculation result of step S4 to a true expression risk calculation formula (see Equation 2) set in advance, and from the time point n. The true onset probability until time t is calculated (step SS5), and the calculation result is transmitted to the client terminal 4 that is the transmission source of the calculation request (step S6).

  For example, the case where the onset risk of the disease A whose onset probability in the 30s to 40s greatly affects the later onset rate is calculated for the evaluated person B will be specifically described. In this case, the onset risk calculation unit 312 accepts designation of the disease ID of the disease A and the evaluator ID of the evaluator B from the user terminal 4, and the health diagnosis data (e.g. , 40-year-old, 50-year-old, 60-year-old, and 70-year-old health check data) and basic information, disease A factor information, weighting function, and other data. The weighting function of disease A is set in advance such that, for example, the weight value of the onset rate in the 30s to 40s is larger than the weight values at other time points before and after. And the onset risk calculation part 312 produces | generates the risk model formula for the disease A, For example, it shows about each time point as shown to FIG. A temporary onset probability is calculated, and a weighting value at each time point as shown in FIG. 8B is calculated using a weighting function set in advance for the disease A, for example. As shown in the figure, the calculated weight value emphasizes the onset probability in the 30s to 40s. Then, the onset risk calculation unit 312 calculates the sum of the onset probability at each time point and the weight value according to the true onset risk calculation formula, for example, as shown in FIG. Is calculated. In this way, as shown in FIG. 9, the true risk of onset with an emphasis on the onset probability in the 30s to 40s is calculated.

  Next, for example, a case will be specifically described in which the risk of developing disease C, for which an effort to improve the health condition at a certain point of time greatly affects the subsequent incidence, is calculated for the evaluated person D. In this case, the onset risk calculation unit 312 accepts designation of the disease ID of the disease C and the evaluated person ID of the evaluated person D from the user terminal 4, and the medical examination data of the evaluated person D (for example, 30 years old) , 40-year-old, 50-year-old, 60-year-old, 70-year-old health diagnosis data) and basic information, disease C factor information, weighting function, and other data. Note that the weighting function of the disease C is such that, for example, the weight value of the onset rate at a time point (for example, 40 years old) on which the improvement of the health condition greatly affects the disease is larger than the weight value at other time points before and after. Functions are preset. And the onset risk calculation part 312 produces | generates the risk model type | formula for the disease C, and uses the produced | generated risk model type | formula and the medical examination data of the patient D, for example about each time point as shown to FIG. 10 (A). A temporary onset probability is calculated, and a weight value at each time point as shown in FIG. 10B is calculated using a weighting function set in advance for the disease C, for example. As shown in the figure, the calculated weight value gives priority to the onset probability at the time (for example, 40 years old) when the improvement result of the health condition greatly affects the disease. Then, the onset risk calculation unit 312 calculates the sum of the onset probability and the weight value at each time point according to the true onset risk calculation formula, for example, as shown in FIG. Is calculated. In this way, as shown in FIG. 11, the true onset risk is calculated with an emphasis on the onset probability that has decreased due to improvement efforts at a certain point in time.

  As described above, according to the present invention, weights are assigned to the onset probabilities calculated from past medical examination data and the like, and the total value thereof is set as a true onset risk. Diseases in which the health status in a certain period of time has a major effect on the probability of later onset, such as lifestyle-related diseases whose risk of onset increases or decreases due to changes, and efforts to improve the health status of the person greatly affect the probability of later onset The onset risk can be predicted more accurately for the disease.

とおき、

という条件の下、過去のサンプルデータ（例えば、健康診断データから求めたＳ_ｔ（ｉ）、実際の発症の有無等のデータ）を用いて、最近推定法等によって、β_ｉ（ｉ＝１，２，・・・，ｎ）を算出してもよい。そして、求めたβ_ｉ（ｉ＝１，２，・・・，ｎ）を各時点における重み値として各疾患毎（疾患ＩＤ毎）にテーブル等に格納して、リスク関連ＤＢ２に記憶してもよい。この場合、管理サーバ３の制御部３１は、評価対象として指定された疾患に対応するテーブルから各時点における重み値のデータを読み込み、各時点の仮の発症確率にそれぞれ重み値を掛けた値の総和を算出して、真の発症リスクを算出する。 The present invention can be variously modified and applied.
For example, as a weighting function, what predicts the future disease risk best using past data may be obtained and used using a statistical method, a learning algorithm, or the like.
For example,

Toki,

Under the conditions, β _i (i = 1, 1) using past sample data (for example, _St (i) obtained from health check data, actual occurrence of onset data, etc.) 2, ..., n) may be calculated. Then, the obtained β _i (i = 1, 2,..., N) may be stored in a table or the like for each disease (for each disease ID) as a weight value at each time point and stored in the risk-related DB 2. Good. In this case, the control unit 31 of the management server 3 reads the weight value data at each time point from the table corresponding to the disease designated as the evaluation target, and multiplies the temporary onset probability at each time point by the weight value. Calculate the sum and calculate the true risk of onset.

Note that the risk model equation (Equation 1) and the true expression risk calculation equation (Equation 2) may be described in the operation program executed by the control unit 31 of the management server 3, and are also referred to by the operation program. It may be data.
The risk model formula (Equation 1) may be a formula common to each disease, or may be provided for each disease.
Moreover, it is good also as a data structure that health checkup data contains each item of basic information.
In addition, the management server 3 may have a system configuration including at least one of the evaluated person information DB1 and the risk related DB2.

The system of the present invention can be realized using a normal computer system, not a dedicated system. For example, a program for executing the above operation is stored in a computer-readable recording medium (FD, CD-ROM, DVD, etc.) and distributed, and the program is installed in the computer to execute the above processing. The management server 3, the client terminal 4 and the like may be configured. Alternatively, it may be stored in a disk device of a server device on a network such as the Internet and downloaded to a computer, for example.
In addition, when the OS realizes the above functions by sharing the OS or jointly with the OS and the application, etc., only the part other than the OS may be stored and distributed in the medium, or may be downloaded to the computer. Good.

It is a figure which shows the structural example of the onset risk calculation system of embodiment of this invention. It is a figure which shows an example of basic information. It is a figure which shows an example of medical examination data. It is a figure which shows an example of factor information. It is a figure which shows the structural example of a management server. It is a figure which shows the structural example of a client terminal. It is a flowchart for demonstrating onset risk calculation processing. It is a figure for demonstrating concretely the process in the case of calculating the onset risk of the disease for which the onset probability in a certain period has a large effect on the later onset rate. It is a figure for demonstrating concretely the process in the case of calculating the onset risk of the disease for which the onset probability in a certain period has a large effect on the later onset rate. It is a figure for demonstrating concretely the process in the case of calculating the onset risk of the disease for which the improvement efforts of the health condition at a certain time point have a large effect on the later onset rate. It is a figure for demonstrating concretely the process in the case of calculating the onset risk of the disease for which the improvement efforts of the health condition at a certain time point have a large effect on the later onset rate.

Explanation of symbols

1 Evaluator information DB
2 Risk-related DB
DESCRIPTION OF SYMBOLS 3 Management server 4 Client terminal 10 Network 31 Control part 311 Data input part 312 Incidence risk calculation part 32 Storage part 33 Input part 34 Display part 35 Communication control part 4 Client terminal 41 Control part 42 Storage part 43 Display part 44 Input part 45 Communication Control unit

Claims (2)

Corresponds to each person to be evaluated, the person to be evaluated who can identify the person to be evaluated, the health examination data indicating the result of the health examination of the person to be evaluated at a plurality of time points, and the date of birth of the person to be evaluated. A database to store and store,
Of the disease ID that can identify the disease to be evaluated and the factors described in the health check data, when calculating the probability of developing the disease to be evaluated from the time when the health check is performed to a predetermined time Storage means for storing factor information for specifying a factor used for the weighting function for acquiring a weight value for each age in association with each disease;
In response to designation of the evaluated person and the disease to be evaluated by inputting the evaluated person ID and the disease ID, the plurality of points in time corresponding to the evaluated person ID Reading means for reading the health check data and the date of birth from the database, and reading the factor information corresponding to the disease ID and the weighting function from the storage means;
A model formula having a factor designated by the factor information read by the reading unit as a parameter is generated, and the age of the evaluated person at the plurality of time points at which the health examination was performed from the date of birth is respectively determined. For each health check data at the plurality of time points, by specifying the factor indicated by the factor information in the health check data in the model formula, the predetermined health check data is obtained at each age at which the health check was performed. An onset probability calculating means for calculating the probability that the evaluated person will develop the disease to be evaluated by the time point,
The sum of values obtained by multiplying the onset probability of the disease at each age calculated by the onset probability calculating means by the weight value for each age acquired from the weighting function read by the reading means, respectively. An onset risk calculating means for obtaining an onset risk of the disease to be evaluated of the evaluation subject,
Onset risk calculation system that comprising: a. Corresponds to each person to be evaluated, the person to be evaluated who can identify the person to be evaluated, the health examination data indicating the result of the health examination of the person to be evaluated at a plurality of time points, and the date of birth of the person to be evaluated. A database for accumulating and storing the disease, a disease ID that can identify the disease to be evaluated, and a disease described in the health check data from the time when the health check is performed to a predetermined time Storage means for storing factor information for specifying a factor used in calculating the probability of developing the disease and a weighting function for acquiring a weight value for each age in association with each disease Computer
In response to designation of the evaluated person and the disease to be evaluated by inputting the evaluated person ID and the disease ID, the plurality of points in time corresponding to the evaluated person ID Reading means for reading the health check data and the date of birth from the database, and reading the factor information corresponding to the disease ID and the weighting function from the storage means;
A model formula having a factor designated by the factor information read by the reading unit as a parameter is generated, and the age of the evaluated person at the plurality of time points at which the health examination was performed from the date of birth is respectively determined. For each health check data at the plurality of time points, by specifying the factor indicated by the factor information in the health check data in the model formula, the predetermined health check data is obtained at each age at which the health check was performed. An onset probability calculating means for calculating the probability that the evaluated person will develop the disease to be evaluated by the time point,
The sum of values obtained by multiplying the onset probability of the disease at each age calculated by the onset probability calculating means by the weight value for each age acquired from the weighting function read by the reading means, respectively. An onset risk calculating means for obtaining an onset risk of the disease to be evaluated of the evaluated person by calculating
Program to function as.

Images