JP7266807B1 - Lifestyle prediction device, lifestyle prediction system, lifestyle prediction method, lifestyle prediction program, and recording medium - Google Patents

Abstract

The lifestyle prediction device includes a brain state image data acquisition unit, a brain volume-related data extraction unit, and a lifestyle index value data derivation unit. The correlation data is based on the first relational expression data representing the relationship between the brain volume data and the lifestyle index value data, the lifestyle index value data actually measured at a first point in time in the past, and the first relational expression data. Predicted data obtained by predicting lifestyle index data from data related to brain volume at the first time point, and first difference data that is the difference between the data, the lifestyle index value data actually measured at the first time point, and the data from the first time point It includes second difference data that is the difference from the lifestyle index value data actually measured at a later second time point, and second relational expression data representing the relationship between the first difference data and the second difference data. The lifestyle index value data deriving unit derives the future user's lifestyle index value data from the brain volume data based on the second relational expression data.

Description

The present disclosure relates to a lifestyle prediction device, a lifestyle prediction system, a lifestyle prediction method, a lifestyle prediction program, and a recording medium.

A system for preventing dementia has been disclosed (see Patent Document 1, for example). According to Patent Document 1, subjects are grouped with respect to the progression of dementia, problems are determined based on the group to which the subject belongs, and the determined problem is presented to the subject.

JP 2015-180933 A

2. Description of the Related Art Recently, various physical examinations and measurements are performed in health checkups and the like. If it is possible to accurately derive data relating to indicators of future lifestyle habits using such test results and measurement results, it is possible to accurately predict lifestyle habits. This is preferable because effective preventive measures can be taken so as not to suffer from lifestyle-related diseases in the future, and efficient improvement of lifestyle habits can be promoted.

Accordingly, one object of the present disclosure is to provide a lifestyle prediction device that can accurately predict future lifestyle habits.

A lifestyle prediction device according to the present disclosure is a lifestyle prediction device for predicting a future lifestyle of a user. The lifestyle prediction device includes a brain state image data acquisition unit that acquires brain state image data that is image data related to the form of the user's brain or image data related to the function of the user's brain, and the brain state image data acquired by the brain state image data acquisition unit. A brain volume-related data extraction unit for extracting data related to the user's brain volume from the user's brain state image data; and a lifestyle index value data deriving unit for deriving lifestyle index value data of the future user based on correlation data between and lifestyle index value data, which is data relating to lifestyle habits. Correlation data includes first relational expression data representing the relationship between brain volume data and lifestyle index data created based on past brain volume data and lifestyle index value data actually measured in the past. , the difference between the lifestyle index value data actually measured at the first time point in the past and the prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first time point based on the data of the first relational expression. first difference data, second difference data that is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at the second time point after the first time point; and second relational expression data representing a relationship between the difference data and the second difference data. The lifestyle index value data deriving unit derives the future user's lifestyle index value data from the brain volume data based on the second relational expression data.

According to the lifestyle prediction device, it is possible to accurately predict a future lifestyle.

FIG. 1 is a diagram schematically showing the appearance of a lifestyle prediction system according to Embodiment 1. FIG. FIG. 2 is a block diagram showing the configuration of the lifestyle prediction system shown in FIG. FIG. 3 is a conceptual diagram showing detailed data included in the correlation data. Fig. 4 shows the relationship between data on the ratio of total brain gray matter volume to intracranial volume, which is data on past brain volume, and systolic blood pressure data, which is one of the lifestyle index values actually measured in the past. is a graph showing FIG. 5 is a graph showing the relationship between the measured SBP value data at the first time point and the SBP value data predicted using the first relational expression from the data on the volume ratio of the total brain gray matter at the first time point. be. FIG. 6 is a graph showing the relationship between first difference data and second difference data. FIG. 7 is a flowchart showing a schematic process of deriving future user's lifestyle index value data using the lifestyle prediction system according to the first embodiment. FIG. 8 shows Log 10 data of the ratio of the total brain gray matter volume to the intracranial volume, which is data on the past brain volume, and systolic blood pressure (SBP), which is one of the lifestyle index values actually measured in the past. Value) is a graph showing the relationship with the data. FIG. 9 is a graph showing the relationship between the actually measured SBP value data at the first time point in this case and the SBP value data predicted from the Log 10 data of the ratio of the total brain gray matter volume at the first time point. FIG. 10 is a graph showing the relationship between the first difference data and the second difference data in this case.

[Description of Embodiments of the Present Disclosure]
First, the embodiments of the present disclosure are listed and described. A lifestyle prediction device according to the present disclosure is a lifestyle prediction device for predicting a future lifestyle of a user. The lifestyle prediction device includes a brain state image data acquisition unit that acquires brain state image data that is image data related to the form of the user's brain or image data related to the function of the user's brain, and the brain state image data acquired by the brain state image data acquisition unit. A brain volume-related data extraction unit for extracting data related to the user's brain volume from the user's brain state image data; and a lifestyle index value data deriving unit for deriving lifestyle index value data of the future user based on correlation data between and lifestyle index value data, which is data relating to lifestyle habits. Correlation data includes first relational expression data representing the relationship between brain volume data and lifestyle index data created based on past brain volume data and lifestyle index value data actually measured in the past. , the difference between the lifestyle index value data actually measured at the first time point in the past and the prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first time point based on the data of the first relational expression. first difference data, second difference data that is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at the second time point after the first time point; and second relational expression data representing a relationship between the difference data and the second difference data. The lifestyle index value data deriving unit derives the future user's lifestyle index value data from the brain volume data based on the second relational expression data.

In order to prevent the occurrence of lifestyle-related diseases caused by lifestyle habits in the future and to improve lifestyle habits efficiently, it is necessary to gather as much lifestyle index data as possible, such as future systolic blood pressure data and smoking amount data. Accurate prediction is desirable. Therefore, the present inventors considered as follows. First, the lifestyle index value data was examined, and it was found that there is a correlation between the lifestyle index value data of the user and the brain state image data, which is image data regarding the form of the user's brain or image data regarding the function of the brain. . Furthermore, instead of simply deriving future lifestyle index value data from image data on brain morphology and image data on brain function, it is based on actually measured lifestyle index value data and brain state image data before measurement. If the relationship with the predicted lifestyle index value data can be taken into consideration and the predicted lifestyle index value data can be corrected, the future user's lifestyle index value data can be derived more accurately and accurate lifestyle habits can be obtained. I thought it was possible to predict the

Therefore, the present inventors have made intensive studies to find out about the relationship between the user's brain state image data and future lifestyle index value data. Correlation between deviation from habit index value data (first deviation) and deviation between lifestyle index value data actually measured in the past and lifestyle index value data actually measured after that time (second deviation) It is thought that future lifestyle index value data can be derived more accurately if correction based on this correlation is performed on predicted lifestyle index value data, and the configuration of the present invention is conceived. Arrived.

According to the lifestyle prediction device according to the present disclosure, the lifestyle index value data derivation unit obtains the brain volume data and the lifestyle habit data from the user's brain volume data extracted by the brain volume related data extraction unit. Future user's lifestyle index value data is derived based on correlation data with lifestyle index value data. The correlation data is the first relational expression representing the relationship between the brain volume data created based on the past brain volume data and the lifestyle index value data actually measured in the past, and the lifestyle index value data. difference between the data, the lifestyle index value data actually measured at the first point in time in the past, and the prediction data obtained by predicting the lifestyle index value data from the brain volume data at the first point in time based on the data of the first relational expression and the second difference data, which is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at the second time point after the first time point, and second relational expression data representing the relationship between the first difference data and the second difference data. By doing so, the user's brain state image data can be used to determine the future lifestyle, reflecting the correction considering the first difference data corresponding to the first deviation and the second difference data corresponding to the second deviation. Habit indicator value data can be derived. Therefore, according to the said lifestyle prediction apparatus, a future lifestyle can be correctly predicted.

In the lifestyle prediction device described above, the image data relating to the morphology of the brain may include at least one of MRI (Magnetic Resonance Imaging) image data of the brain and X-ray CT (Computed Tomography) image data of the brain. The image data on brain function may include at least one of brain PET (Positron Emission Tomography) image data and brain SPECT (Single Photon Emission Computed Tomography) image data. By using such image data on brain morphology or image data on brain function as brain state image data, it is possible to accurately predict future lifestyle habits more reliably.

In the lifestyle prediction device described above, the data on brain volume may include data on the ratio of the volume of total brain gray matter to the intracranial volume. By using such data related to the volume ratio of total brain gray matter, future lifestyle index value data of the user can be more accurately derived, and future lifestyle can be predicted.

In the lifestyle prediction device, the lifestyle index value data includes systolic blood pressure (SBP) value data, diastolic blood pressure (DBP) value data, smoking amount value data, and drinking amount value. At least one of data and BMI (Body Mass Index) value data may be included. Such lifestyle index values are closely related to lifestyle-related diseases that are likely to occur in the future. can.

In the lifestyle prediction device described above, at least one of the first relational expression data and the second relational expression data may include a linear relational expression. By doing so, at least one of the first relational expression data and the second relational expression can be simplified to easily derive the lifestyle index value data.

The lifestyle prediction device may further include a storage unit that stores at least one of the first difference data, the second difference data, the first relational expression data, and the second relational expression data. By doing so, the future lifestyle index value data can be efficiently derived using the data stored in the storage unit.

A lifestyle prediction system according to the present disclosure includes a server and is a lifestyle prediction system for predicting a user's future lifestyle. The server externally acquires brain state image data, which is image data related to the form of the user's brain or image data related to the function of the user's brain. a brain volume-related data extraction unit for extracting data related to the user's brain volume from the brain state image data of the brain volume-related data extraction unit; a lifestyle index value data deriving unit that derives the future user's lifestyle index value data based on the correlation data with the lifestyle index value data, which is data related to lifestyle habits. Correlation data includes first relational expression data representing the relationship between brain volume data and lifestyle index data created based on past brain volume data and lifestyle index value data actually measured in the past. , the difference between the lifestyle index value data actually measured at the first time point in the past and the prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first time point based on the data of the first relational expression. first difference data, second difference data that is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at the second time point after the first time point; and second relational expression data representing a relationship between the difference data and the second difference data. The lifestyle index value data deriving unit derives the future user's lifestyle index value data from the brain volume data based on the second relational expression data.

According to such a lifestyle prediction system, a future lifestyle can be accurately predicted.

A lifestyle prediction method according to the present disclosure is a lifestyle prediction method for predicting a future lifestyle of a user, and includes image data relating to the form of the user's brain or image data relating to the function of the user's brain. a step of obtaining image data; a step of extracting data about the volume of the user's brain from the obtained brain state image data of the user; and deriving lifestyle index value data of the future user based on correlation data with lifestyle index value data, which is data related to lifestyle habits. Correlation data includes first relational expression data representing the relationship between brain volume data and lifestyle index data created based on past brain volume data and lifestyle index value data actually measured in the past. , the difference between the lifestyle index value data actually measured at the first time point in the past and the prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first time point based on the data of the first relational expression. first difference data, second difference data that is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at the second time point after the first time point; and second relational expression data representing a relationship between the difference data and the second difference data. The step of deriving the lifestyle index value data of the future user derives the lifestyle index value data of the future user from the brain volume data based on the second relational expression data.

According to such a lifestyle prediction method, a future lifestyle can be accurately predicted.

A lifestyle prediction program according to the present disclosure is a program for a lifestyle prediction system that includes a server and is used in a lifestyle prediction system for predicting future lifestyle habits of a user, wherein the server is a form of the user's brain. a brain state image data acquisition unit that acquires brain state image data that is image data related to or image data related to the function of the user's brain; A brain volume-related data extraction unit for extracting data related to volume, and from the data related to brain volume of the user extracted by the brain volume-related data extraction unit, data related to brain volume and lifestyle index value data that is data related to lifestyle habits It functions as a lifestyle index value data deriving unit that derives the future lifestyle index value data of the user based on the correlation data with the . Correlation data includes first relational expression data representing the relationship between brain volume data and lifestyle index data created based on past brain volume data and lifestyle index value data actually measured in the past. , the difference between the lifestyle index value data actually measured at the first time point in the past and the prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first time point based on the data of the first relational expression. first difference data, second difference data that is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at the second time point after the first time point; and second relational expression data representing a relationship between the difference data and the second difference data. The lifestyle index value data deriving unit derives the future user's lifestyle index value data from the brain volume data based on the second relational expression data.

According to such a lifestyle prediction program, a future lifestyle can be accurately predicted.

A storage medium according to the present disclosure includes a server, is used in a lifestyle prediction system for predicting future lifestyle habits of a user, and is a computer-readable recording medium, wherein the server relates to the morphology of the user's brain. a brain state image data acquisition unit for acquiring image data or brain state image data that is image data relating to the functions of the user's brain; a brain volume-related data extraction unit that extracts data related to the brain volume, and from the data related to the user's brain volume extracted by the brain volume-related data extraction unit, data related to brain volume and lifestyle index value data that is data related to lifestyle habits Based on the correlation data, it functions as a lifestyle index value data deriving unit that derives the future lifestyle index value data of the user. Correlation data includes first relational expression data representing the relationship between brain volume data and lifestyle index data created based on past brain volume data and lifestyle index value data actually measured in the past. , the difference between the lifestyle index value data actually measured at the first time point in the past and the prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first time point based on the data of the first relational expression. first difference data, second difference data that is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at the second time point after the first time point; and second relational expression data representing a relationship between the difference data and the second difference data. The lifestyle index value data deriving unit derives the future user's lifestyle index value data from the brain volume data based on the second relational expression data.

According to such a storage medium, it is possible to accurately predict future lifestyle habits.

[Details of the embodiment of the present invention]
Next, an embodiment of a lifestyle prediction system including the lifestyle prediction device of the present disclosure will be described with reference to the drawings. In the following drawings, the same reference numerals are given to the same or corresponding parts, and the description thereof will not be repeated.

(Embodiment 1)
The configuration of the lifestyle prediction system according to Embodiment 1 of the present disclosure will be described. FIG. 1 is a diagram schematically showing the appearance of a lifestyle prediction system according to Embodiment 1. FIG. FIG. 2 is a block diagram showing the configuration of the lifestyle prediction system shown in FIG.

1 and 2, lifestyle prediction system 11 in Embodiment 1 includes server 13 and terminal device 14 connectable to server 13 via network 12 . The network 12 may be the Internet, or an internal network in a closed environment such as an intranet built for the company. Network 12 may be wired or wireless.

First, the configuration of the terminal device 14 will be described. In this embodiment, the terminal device 14 is, for example, a portable terminal device 14 such as a smart phone. Of course, it may be a tablet PC, a notebook PC, or a stationary desktop PC. The terminal device 14 serves as an interface between a terminal device network interface unit 31 for connecting to the network 12, a terminal device control unit 32 for controlling the terminal device 14 itself, a terminal device memory 33 for storing data, and a user. and a touch panel 15 . The terminal device 14 inputs data and information to the server 13 and outputs data and information transmitted from the server 13 via the touch panel 15 .

Next, the configuration of the server 13 will be described. The server 13 functions as a lifestyle prediction device. The server 13 includes a server network interface section 21 for connecting to the network 12, a server control section 22 for controlling the server 13 itself, and a server hard disk 23 as a storage section for storing data. Connected to the server 13 are a display for displaying data, and a keyboard and mouse for inputting data (both not shown). The server hard disk 23 stores first difference data, second difference data, first relational expression data, and second relational expression data, which will be described later.

Next, a specific configuration of the server control unit 22 will be described. The server control unit 22 includes a brain state image data acquisition unit 41 , a brain volume-related data extraction unit 42 , and a lifestyle index value data derivation unit 43 . The brain state image data acquisition unit 41 acquires brain state image data, which is image data relating to the morphology of the user's brain or image data relating to the function of the user's brain. The brain volume-related data extraction unit 42 extracts data related to the volume of the user's brain from the user's brain state image data acquired by the brain state image data acquisition unit 41 . The lifestyle index value data deriving unit 43 calculates the correlation between the brain volume-related data and the lifestyle index value data, which is data related to lifestyle habits, from the data related to the user's brain volume extracted by the brain volume-related data extraction unit 42. Future user's lifestyle index value data is derived based on the relationship data. These configurations will be described in detail later.

In this embodiment, the brain state image data acquired by the brain state image data acquisition unit 41 is MRI image data of the brain, which is image data relating to the morphology of the brain. Further, in the present embodiment, the brain volume-related data extracted by the brain volume-related data extraction unit 42 is data related to the ratio of the total brain gray matter volume to the intracranial volume.

Here, the configuration of correlation data will be described. FIG. 3 is a conceptual diagram showing detailed data included in the correlation data. Referring to FIG. 3 , correlation data 51 includes first relational expression data 52 , first difference data 53 , second difference data 54 and second relational expression data 55 .

First, the first relational expression data 52 will be described. The first relational expression data 52 represents the relationship between the brain volume data and the lifestyle index value data created based on the past brain volume data and the lifestyle index value data actually measured in the past.

Figure 4 shows past data on the volume of the brain, which is the ratio of the total brain gray matter volume to the intracranial volume, and systolic blood pressure, which is one of the lifestyle index values actually measured in the past (hereinafter simply referred to as It is a graph showing the relationship with the data (sometimes referred to as "SBP value"). The graph shown in FIG. 4 shows the first relational expression. In FIG. 4, the horizontal axis indicates the volume ratio (%) of total brain gray matter, and the vertical axis indicates the measured SBP value (mmHg). The percentage of total brain gray matter volume has been extracted from brain MRI imaging data. That is, it is calculated from the ratio of the volume of the whole brain gray matter to the intracranial volume in the MRI image data.

Referring to FIG. 4, past total brain gray matter volume ratio and actually measured SBP value data have a correlation represented by a line _β1 with a slope and _e1 with an intercept. This correlation is a linear relational expression and is represented by Equation 1 below.

SBP value = whole brain gray matter × β ₁ + e ₁ (1)

In this embodiment, the data of this formula (1) is the first relational expression data 52 included in the correlation data 51 . Note that the p-value in the graph shown in FIG. 4 is, for example, 4.01×10 ⁻²⁴ . This value is sufficiently small and highly reliable.

Next, the first difference data 53 will be explained. The first difference data 53 is the lifestyle index value data actually measured at the first time point in the past, and the prediction of the lifestyle index value data from the brain volume data at the first time point based on the first relational expression data 52. It is the difference between data and . That is, the prediction data obtained by predicting the lifestyle index value data from the brain volume data at the first time point based on the first relational expression data 52 is calculated from the slope β1 and the intercept e1 using the above formula (1). It is the predicted data of the SBP value.

FIG. 5 shows the relationship between the actually measured SBP value data at the first time point and the SBP value data predicted using the first relational expression data 52 from the data on the volume ratio of the total brain gray matter at the first time point. graph. In FIG. 5, the horizontal axis indicates the predicted SBP value (mmHg), and the vertical axis indicates the actually measured SBP value (mmHg). The predicted SBP value is calculated using equation (1) above.

Referring to FIG. 5, there is a slight deviation (first deviation) between the actually measured SBP value data and the predicted SBP value data. This deviation is calculated as the first difference data. The first difference data is represented by the following formula (2).

First difference data = Predicted SBP value - Measured SBP value (2)

In this embodiment, the data calculated by this formula (2) is the first difference data 53 included in the correlation data 51 . Note that the p-value in the graph shown in FIG. 5 is, for example, 9.82×10 ⁻⁸ . This value is sufficiently small and highly reliable.

Next, the second difference data 54 will be explained. The second difference data 54 is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at the second time point after the first time point.

That is, the second difference data 54=measured SBP value (second point in time)−measured SBP value (first point in time). From this relational expression, it is derived that the actually measured SBP value (second point in time) is the actually measured SBP value (first point in time)+second difference data 54 .

Next, the second relational expression data 55 will be described. The second relational expression data 55 represents the relationship between the first difference data 53 and the second difference data 54 .

FIG. 6 is a graph showing the relationship between the first difference data 53 and the second difference data 54. As shown in FIG. The graph shown in FIG. 6 shows the second relational expression. In FIG. 6, the horizontal axis indicates the first difference data 53 (mmHg), and the vertical axis indicates the second difference data 54 (mmHg).

Referring to FIG. 6, first difference data 53 and second difference data 54 have a correlation whose slope is represented by line _β2 and whose intercept is represented by _e2 . This correlation is a linear relational expression and is represented by the following formula (3).

Second difference data=First difference data×β ₂ +e ₂ (3)

In this embodiment, the data of this formula (4) is the second relational expression data 55 included in the correlation data 51 . Note that the p-value in the graph shown in FIG. 6 is, for example, 4.86×10 ⁻⁹ . This value is sufficiently small and highly reliable.

Next, a case of predicting a user's future lifestyle using such a lifestyle prediction system 11 will be described. FIG. 7 is a flowchart showing a schematic process of deriving future user's lifestyle index value data using the lifestyle prediction system 11 according to the first embodiment.

Referring to FIG. 7, first, brain state image data acquisition unit 41 acquires brain state image data that is image data relating to the morphology of the user's brain or image data relating to the function of the user's brain (step YES in S11, hereinafter "step" is omitted). In this embodiment, the brain state image data acquisition unit 41 acquires MRI image data of the brain as image data relating to the morphology of the user's brain. In this case, for example, brain MRI image data transmitted from the terminal device 14 is acquired. Next, the brain volume-related data extracting unit 42 extracts brain volume data from the acquired brain state image data, in this case, MRI image data of the brain (S12). In this case, in the MRI image data of the brain, the intracranial volume data is extracted, and the whole brain gray matter data is also extracted. Then, the ratio of the total brain gray matter volume to the intracranial volume data is calculated and extracted as brain volume-related data.

After that, the lifestyle index value data derivation unit 43 calculates the correlation between the brain volume-related data and the lifestyle index value data, which is data related to the lifestyle, from the extracted brain volume-related data, which is the data related to the brain volume of the user. Future user's lifestyle index value data is derived based on the relationship data 51 (S13). In this case, the lifestyle index value data deriving unit 43 derives the future user's lifestyle index value data from the brain volume data based on the second relational expression data 55 . Specifically, for example, the ratio of the total brain gray matter of the user at present is obtained from the MRI image data, and from the ratio of the total brain gray matter, the prediction data of the SBP value data, which is the lifestyle index value data, is obtained by formula (1). to derive The first difference data 53 and the equation (3) obtained by the formula (2) and the second difference data are the differences between the actually measured SBP value data at the first point in time and the SBP value data at the second point in time. be. Therefore, if the first point in time is the present and the second point in time is the future, the SBP value at the second point in time=the SBP value at the first point in time+the second differential data. From this relational expression, predictive data of the user's future SBP value data is derived. That is, the second difference data 54 is derived based on the actually measured SBP value data at the first point in time and the actually measured SBP value data at the second point in time, and the second difference data 54 and the first difference data 53 are used. accurately derives the future user's SBP value data. The user's future lifestyle habits are predicted using the derived future user's SBP value data.

According to the lifestyle prediction system 11 of the present disclosure, it is possible to derive future lifestyle index value data from the user's brain state image data by reflecting the correction that takes into account the first difference data and the second difference data. . Therefore, according to the lifestyle prediction system 11, it is possible to accurately predict a future lifestyle.

Further, according to the server 13 as the lifestyle prediction device described above, the user's Future lifestyle index value data can be derived from the brain state image data. Therefore, according to the said lifestyle prediction apparatus, a future lifestyle can be correctly predicted.

According to the above embodiment, the server 13 includes a server hard disk 23 that stores the first difference data, the second difference data, the first relational expression data and the second relational expression data. Therefore, the data stored in the server hard disk 23 can be used to efficiently derive the future lifestyle index value data.

According to the above embodiments, the data on brain volume includes data on the ratio of total brain gray matter volume to intracranial volume. By using such data related to the volume ratio of total brain gray matter, future lifestyle index value data of the user can be more accurately derived, and future lifestyle can be predicted.

(Other embodiments)
In the above embodiment, the brain state image data acquisition unit acquires MRI image data of the brain as image data related to brain morphology. The brain morphology-related image data acquired by may include at least one of brain MRI image data and brain X-ray CT image data. Further, the image data relating to brain function acquired by the brain state image data acquisition unit may include at least one of PET image data of the brain and SPECT image data of the brain. By using such image data on brain morphology or image data on brain function as brain state image data, it is possible to accurately predict future lifestyle habits more reliably.

In the lifestyle prediction device, the lifestyle index value data includes SBP value data, but is not limited to this, and the lifestyle index value data includes diastolic blood pressure (DBP) value data, At least one of smoking amount value data, drinking amount value data, and BMI (Body Mass Index) value data may be included. Such lifestyle index values are closely related to lifestyle-related diseases that are likely to occur in the future. can.

In the above embodiment, both the first relational expression data and the second relational expression data have linear relational expressions. At least one of the data may include a linear relational expression. By doing so, at least one of the first relational expression data and the second relational expression can be simplified to easily derive the lifestyle index value data. At least one of the first relational expression data and the second relational expression data may be non-linear relational expression data. Specifically, for example, one may be linear relational expression data and the other may be nonlinear relational expression data. By using the non-linear relational expression data, the relational expression can be refined and more accurate lifestyle index value data can be derived. Note that such a relational expression can be derived by deeply analyzing each parameter using deep learning.

FIG. 8 shows Log 10 data of the ratio of the total brain gray matter volume to the intracranial volume, which is data on the past brain volume, and systolic blood pressure (SBP), which is one of the lifestyle index values actually measured in the past. Value) is a graph showing the relationship with the data. In FIG. 8, the horizontal axis represents Log ₁₀ of the value obtained by multiplying the ratio (%) of the total brain gray matter volume by 100, that is, the common logarithm value of 100 times the ratio (%) of the total brain gray matter volume. , the vertical axis indicates the measured SBP value (mmHg). The common logarithmic value of the percentage of total brain gray matter volume has been derived based on brain MRI imaging data. Note that the p-value in the graph shown in FIG. 8 is, for example, 8.88×10 ⁻²⁴ . This value is sufficiently small and highly reliable.

FIG. 9 is a graph showing the relationship between the actually measured SBP value data at the first time point in this case and the SBP value data predicted from the Log 10 data of the ratio of the total brain gray matter volume at the first time point. In FIG. 9, the horizontal axis indicates the predicted SBP value (mmHg), and the vertical axis indicates the actually measured SBP value (mmHg). The predicted SBP value is data calculated based on the correlation shown in FIG. FIG. 9 is a graph corresponding to FIG. 5 described above. Note that the p-value in the graph shown in FIG. 9 is, for example, 5.21×10 ⁻⁸ . This value is sufficiently small and highly reliable.

FIG. 10 is a graph showing the relationship between the first difference data and the second difference data in this case. In FIG. 10, the horizontal axis indicates the first difference data 53 (mmHg), and the vertical axis indicates the second difference data 54 (mmHg). FIG. 10 is a graph corresponding to FIG. 6 described above. Note that the p-value in the graph shown in FIG. 10 is, for example, 5.39×10 ⁻⁹ . This value is sufficiently small and highly reliable.

Referring to FIGS. 8 to 10, Log ₁₀ of 100 times the past total brain gray matter volume ratio and actually measured SBP value data have a correlation represented by a linear relational expression. Then, it can be said that the ratio of the past total brain gray matter volume and the actually measured SBP value data have a correlation represented by a nonlinear relational expression. It can be said that the first difference data and the second difference data also have a correlation represented by a nonlinear relational expression.

A lifestyle prediction method according to the present disclosure is a lifestyle prediction method for predicting a future lifestyle of a user, and includes image data relating to the form of the user's brain or image data relating to the function of the user's brain. a step of obtaining image data; a step of extracting data about the volume of the user's brain from the obtained brain state image data of the user; and deriving lifestyle index value data of the future user based on correlation data with lifestyle index value data, which is data related to lifestyle habits. Correlation data includes first relational expression data representing the relationship between brain volume data and lifestyle index data created based on past brain volume data and lifestyle index value data actually measured in the past. , the difference between the lifestyle index value data actually measured at the first time point in the past and the prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first time point based on the data of the first relational expression. first difference data, second difference data that is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at the second time point after the first time point; and second relational expression data representing a relationship between the difference data and the second difference data. The step of deriving the lifestyle index value data of the future user derives the lifestyle index value data of the future user from the brain volume data based on the second relational expression data.

According to such a lifestyle prediction method, a future lifestyle can be accurately predicted.

A lifestyle prediction program according to the present disclosure is a program for a lifestyle prediction system that includes a server and is used in a lifestyle prediction system for predicting future lifestyle habits of a user, wherein the server is a form of the user's brain. a brain state image data acquisition unit that acquires brain state image data that is image data related to or image data related to the function of the user's brain; A brain volume-related data extraction unit for extracting data related to volume, and from the data related to brain volume of the user extracted by the brain volume-related data extraction unit, data related to brain volume and lifestyle index value data that is data related to lifestyle habits It functions as a lifestyle index value data deriving unit that derives the future lifestyle index value data of the user based on the correlation data with the . Correlation data includes first relational expression data representing the relationship between brain volume data and lifestyle index data created based on past brain volume data and lifestyle index value data actually measured in the past. , the difference between the lifestyle index value data actually measured at the first time point in the past and the prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first time point based on the data of the first relational expression. first difference data, second difference data that is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at the second time point after the first time point; and second relational expression data representing a relationship between the difference data and the second difference data. The lifestyle index value data deriving unit derives the future user's lifestyle index value data from the brain volume data based on the second relational expression data.

According to such a lifestyle prediction program, a future lifestyle can be accurately predicted.

A storage medium according to the present disclosure includes a server, is used in a lifestyle prediction system for predicting future lifestyle habits of a user, and is a computer-readable recording medium, wherein the server relates to the morphology of the user's brain. a brain state image data acquisition unit for acquiring image data or brain state image data that is image data relating to the functions of the user's brain; a brain volume-related data extraction unit that extracts data related to the brain volume, and from the data related to the user's brain volume extracted by the brain volume-related data extraction unit, data related to brain volume and lifestyle index value data that is data related to lifestyle habits Based on the correlation data, it functions as a lifestyle index value data deriving unit that derives the future lifestyle index value data of the user. Correlation data includes first relational expression data representing the relationship between brain volume data and lifestyle index data created based on past brain volume data and lifestyle index value data actually measured in the past. , the difference between the lifestyle index value data actually measured at the first time point in the past and the prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first time point based on the data of the first relational expression. first difference data, second difference data that is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at the second time point after the first time point; and second relational expression data representing a relationship between the difference data and the second difference data. The lifestyle index value data deriving unit derives the future user's lifestyle index value data from the brain volume data based on the second relational expression data.

According to such a storage medium, it is possible to accurately predict future lifestyle habits.

It should be understood that the embodiments disclosed this time are illustrative in all respects and are not restrictive in any aspect. The scope of the present invention is not defined in the above-described sense, but is indicated by the scope of the claims, and is intended to include all modifications within the meaning and scope equivalent to the scope of the claims.

11 lifestyle prediction system 12 network 13 server 14 terminal device 15 touch panel 21 server network interface unit 22 server control unit 23 server hard disk 31 terminal device network interface unit 32 terminal device control unit 33 terminal device memory 41 brain state image data acquisition unit 42 brain volume related data extraction unit 43 lifestyle index value data derivation unit 51 correlation data 52 first relational expression data 53 first difference data 54 second difference data , 55 second relational expression data.

Claims (10)

A lifestyle prediction device for predicting a future user's lifestyle,
a brain state image data acquisition unit that acquires brain state image data that is image data related to the user's brain morphology or image data related to the user's brain function;
a brain volume-related data extraction unit that extracts data related to the brain volume of the user from the brain state image data of the user acquired by the brain state image data acquisition unit;
Based on correlation data between the brain volume-related data and lifestyle index value data, which is data related to lifestyle habits, from the user's brain volume-related data extracted by the brain volume-related data extraction unit, the future a lifestyle index value data deriving unit that derives the lifestyle index value data of the user;
The correlation data are
first relational expression data representing the relationship between the brain volume data and the lifestyle index value data created based on the past brain volume data and the lifestyle index value data actually measured in the past;
the lifestyle index value data actually measured at the first point in time in the past; prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first point in time based on the first relational expression data; the first difference data, which is the difference between
Second difference data, which is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at a second time point after the first time point;
and second relational expression data representing the relationship between the first difference data and the second difference data,
The lifestyle prediction device, wherein the lifestyle index value data deriving unit derives the future lifestyle index value data of the user from the brain volume data based on the second relational expression data. the image data on the morphology of the brain includes at least one of the MRI image data of the brain and the X-ray CT image data of the brain;
2. The lifestyle prediction device according to claim 1, wherein said image data relating to brain function includes at least one of PET image data of said brain and SPECT image data of said brain. The lifestyle prediction device according to claim 1 or 2, wherein the data on brain volume includes data on a ratio of total brain gray matter volume to intracranial volume. The lifestyle index value data includes at least one of systolic blood pressure value data, diastolic blood pressure value data, smoking amount value data, drinking amount value data, and BMI value data. 3. The lifestyle prediction device according to 2. 3. The lifestyle prediction device according to claim 1, wherein at least one of said first relational expression data and said second relational expression data includes a linear relational expression. Claim 1 or Claim 2, further comprising a storage unit that stores at least one of the first difference data, the second difference data, the first relational expression data, and the second relational expression data. The lifestyle prediction device described. A lifestyle prediction system comprising a server for predicting a future user's lifestyle,
Said server
a brain state image data acquisition unit that externally acquires brain state image data, which is image data relating to the form of the user's brain or image data relating to the function of the user's brain;
a brain volume-related data extraction unit that extracts data related to the brain volume of the user from the brain state image data of the user acquired by the brain state image data acquisition unit;
Based on correlation data between the brain volume-related data and lifestyle index value data, which is data related to lifestyle habits, from the user's brain volume-related data extracted by the brain volume-related data extraction unit, the future a lifestyle index value data deriving unit that derives the lifestyle index value data of the user;
The correlation data are
first relational expression data representing the relationship between the brain volume data and the lifestyle index value data created based on the past brain volume data and the lifestyle index value data actually measured in the past;
the lifestyle index value data actually measured at the first point in time in the past; prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first point in time based on the first relational expression data; the first difference data, which is the difference between
Second difference data, which is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at a second time point after the first time point;
and second relational expression data representing the relationship between the first difference data and the second difference data,
The lifestyle prediction system, wherein the lifestyle index value data deriving unit derives the future lifestyle index value data of the user from the brain volume data based on the second relational expression data. A lifestyle prediction method for predicting a future user's lifestyle, comprising:
a step of acquiring brain state image data, which is image data relating to the morphology of the user's brain or image data relating to the function of the user's brain;
a step of extracting data related to the volume of the user's brain from the acquired brain state image data of the user;
Future lifestyle index value data of the user based on correlation data between the brain volume data and the lifestyle index value data, which is data about lifestyle habits, from the extracted brain volume data of the user. and deriving
The correlation data are
first relational expression data representing the relationship between the brain volume data and the lifestyle index value data created based on the past brain volume data and the lifestyle index value data actually measured in the past;
the lifestyle index value data actually measured at the first point in time in the past; prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first point in time based on the first relational expression data; the first difference data, which is the difference between
Second difference data, which is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at a second time point after the first time point;
and second relational expression data representing the relationship between the first difference data and the second difference data,
The step of deriving the future lifestyle index value data of the user derives the future lifestyle index value data of the user from the brain volume data based on the second relational expression data . A lifestyle prediction method executed by a habit prediction device . A lifestyle prediction system program comprising a server and used in a lifestyle prediction system for predicting the lifestyle of a future user,
the server,
a brain state image data acquisition unit that acquires brain state image data that is image data related to the user's brain morphology or image data related to the user's brain function;
a brain volume-related data extraction unit for extracting data related to the volume of the user's brain from the brain state image data of the user acquired by the brain state image data acquisition unit; From the data on the brain volume of the user, the future lifestyle index value data of the user is derived based on the correlation data between the data on the brain volume and the lifestyle index value data that is the data on the lifestyle. Function as a lifestyle index value data derivation unit,
The correlation data are
first relational expression data representing the relationship between the brain volume data and the lifestyle index value data created based on the past brain volume data and the lifestyle index value data actually measured in the past;
the lifestyle index value data actually measured at the first point in time in the past; prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first point in time based on the first relational expression data; the first difference data, which is the difference between
Second difference data, which is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at a second time point after the first time point;
and second relational expression data representing the relationship between the first difference data and the second difference data,
A program for a lifestyle prediction system, wherein the lifestyle index value data deriving unit derives the future lifestyle index value data of the user from the brain volume data based on the second relational expression data. A computer-readable recording medium comprising a server and used in a lifestyle prediction system for predicting the lifestyle of a future user,
the server,
a brain state image data acquisition unit that acquires brain state image data that is image data related to the user's brain morphology or image data related to the user's brain function;
a brain volume-related data extraction unit for extracting data related to the volume of the user's brain from the brain state image data of the user acquired by the brain state image data acquisition unit; From the data on the brain volume of the user, the future lifestyle index value data of the user is derived based on the correlation data between the data on the brain volume and the lifestyle index value data that is the data on the lifestyle. Function as a lifestyle index value data derivation unit,
The correlation data are
first relational expression data representing the relationship between the brain volume data and the lifestyle index value data created based on the past brain volume data and the lifestyle index value data actually measured in the past;
the lifestyle index value data actually measured at the first point in time in the past; prediction data obtained by predicting the lifestyle index value data from the data related to the brain volume at the first point in time based on the first relational expression data; the first difference data, which is the difference between
Second difference data, which is the difference between the lifestyle index value data actually measured at the first time point and the lifestyle index value data actually measured at a second time point after the first time point;
and second relational expression data representing the relationship between the first difference data and the second difference data,
The lifestyle index value data derivation unit records a lifestyle prediction system program for deriving the future lifestyle index value data of the user from the brain volume data based on the second relational expression data. computer-readable recording medium.

Images