Cognitive function evaluation system

Abstract

To alternatively and precisely perform an evaluation of a cognitive function.SOLUTION: A cognitive function evaluation system S includes an input unit 31 and an evaluation estimation unit 36. The evaluation estimation unit 36 performs machine learning of the correlation of input data and teacher data by using a serum vitamin D level, body fat information, a sex, and an age as the input data and by using a value associated with a cognitive function as the teacher data. In that way, the evaluation estimation unit 36 estimates a value associated with the cognitive function when the serum vitamin D level, the body fat information, the sex, and the age are input.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a cognitive function evaluation system.

Background Art

[0002] In Japan, the aging process is advancing, and it is estimated that the number of people with dementia will increase to about 7 million by 2030. Although there is no cure for dementia, it is possible to delay the onset and progression through preventive management. Therefore, in order to take preventive measures to reduce the risk of developing dementia, it is important to know as early as possible about the decline in cognitive function.

[0003] As a method for measuring the decline in cognitive function, for example, there is MMSE (Mini Mental State Examination). MMSE is a neuropsychological test conducted when there is a suspicion of dementia, and it can objectively measure the decline in cognitive function in points. However, since MMSE requires an interview by an expert, it incurs a lot of costs and also takes about 10 minutes for the test. Therefore, MMSE has a problem in terms of easily measuring the decline in cognitive function.

[0004] Also, as a screening test for detecting mild cognitive impairment (MCI (Mild Cognitive Impairment)), there is MoCA-J (Montreal Cognitive Assessment-Japanese). However, MoCA-J also requires an interview by an expert, similar to MMSE.

[0005] Therefore, for example, a dementia / memory loss testing device such as that in Patent Document 1 has been developed. This device is said to be able to easily perform a dementia test by acting on the right brain using music. However, in this device, the correlation with the conventional evaluation of cognitive function is unclear, and it cannot be used as an alternative means for the conventional tests related to cognitive function.

Prior Art Documents

Patent Documents

[0006] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2010-178920 [Summary of the Invention] [Problems to be Solved by the Invention]

[0007] In view of the above circumstances, the problem to be solved by the present invention is to provide a cognitive function evaluation system that can accurately substitute for the evaluation related to cognitive function. [Means for Solving the Problems]

[0008] To solve the above problems, the cognitive function evaluation system according to the present invention includes an evaluation estimation unit, The evaluation estimation unit is pre-trained by teacher data that uses the serum vitamin D value, information related to body fat (hereinafter referred to as body fat information), gender, and age as input data and a value related to cognitive function as output data. When the serum vitamin D value, body fat information, gender, and age are input, it estimates a value related to cognitive function.

[0009] The above cognitive function evaluation system preferably The value related to cognitive function includes at least one value among the values corresponding to the test results of MMSE (Mini Mental State Examination), MoCA-J (Montreal Cognitive Assessment-Japanese), grip strength, SMI (Skeletal Muscle Index), and WHOQOL (World Health Organization Quality of Life).

[0010] The above cognitive function evaluation system preferably further includes a vitamin D value estimation unit, The vitamin D value estimation unit is pre-trained by machine learning using teacher data with body fat information, gender, information on the frequency of vitamin D intake during a predetermined period (hereinafter referred to as vitamin D intake information), and information on the frequency of ultraviolet ray exposure during a predetermined period (hereinafter referred to as ultraviolet ray exposure information) as input data and the serum vitamin D value as output data. When body fat information, gender, vitamin D intake information, and ultraviolet ray exposure information are input, it estimates the serum vitamin D value. The evaluation and estimation unit estimates a value related to the cognitive function with the serum vitamin D value estimated by the vitamin D value estimation unit input.

[0011] The above cognitive function evaluation system preferably further includes a vitamin D value estimation unit, The vitamin D value estimation unit calculates the serum vitamin D value according to a predetermined formula using body fat information, gender, information on the frequency of vitamin D intake during a predetermined period (hereinafter referred to as vitamin D intake information), and information on the frequency of ultraviolet ray exposure during a predetermined period (hereinafter referred to as ultraviolet ray exposure information). The evaluation and estimation unit estimates a value related to the cognitive function with the serum vitamin D value calculated by the vitamin D value estimation unit input.

[0012] The serum vitamin D value estimation system according to the present invention includes a vitamin D value estimation unit, The vitamin D value estimation unit is pre-trained by machine learning using teacher data with information on body fat (hereinafter referred to as body fat information), gender, information on the frequency of vitamin D intake during a predetermined period (hereinafter referred to as vitamin D intake information), and information on the frequency of ultraviolet ray exposure during a predetermined period (hereinafter referred to as ultraviolet ray exposure information) as input data and the serum vitamin D value as output data. When body fat information, gender, vitamin D intake information, and ultraviolet ray exposure information are input, it estimates the serum vitamin D value.

[0013] The serum vitamin D value estimation system according to the present invention includes a vitamin D value estimation unit, The vitamin D value estimation unit calculates the serum vitamin D value using a predetermined formula that uses information related to body fat (hereinafter referred to as body fat information), gender, information related to the frequency of vitamin D intake during a predetermined period (hereinafter referred to as vitamin D intake information), and information related to the frequency of ultraviolet ray exposure during a predetermined period (hereinafter referred to as ultraviolet ray exposure information).

[0014] The cognitive function evaluation system or the serum vitamin D value estimation system preferably The vitamin D intake information includes information on the intake frequency of supplements containing vitamin D and fish during a predetermined period.

[0015] The cognitive function evaluation system or the serum vitamin D value estimation system preferably The vitamin D intake information includes the intake frequency of vitamin D supplements, multivitamin supplements containing vitamin D, fish eaten with bones, dried fish, fatty fish, and lean fish during a predetermined period.

[0016] The cognitive function evaluation system or the serum vitamin D value estimation system preferably The vitamin D intake information includes the intake frequency of vitamin D supplements, multivitamin supplements containing vitamin D, fish eaten with bones, dried fish, fatty fish, and lean fish during a predetermined period, In terms of the vitamin D content, the vitamin D supplement is set in the range of 2.5 to 10.0 μg, the multivitamin supplement is 1.0 to 5.0 μg, the fish eaten with bones is 1.5 to 14.0 μg, the dried fish is 0.5 to 6.0 μg, the fatty fish is 3.0 to 16.0 μg, and the lean fish is 4.0 to 32.0 μg.

[0017] The cognitive function evaluation system or the serum vitamin D value estimation system preferably The vitamin D intake information includes the intake frequency of vitamin D supplements, multivitamin supplements containing vitamin D, fish eaten with bones, dried fish, fatty fish, lean fish, and eggs during a predetermined period, In each vitamin D content related to the calculation of serum vitamin D value, the vitamin D supplement is set in the range of 2.5 to 10.0 μg, the multivitamin supplement is 1.0 to 5.0 μg, the fish eaten with bones is 1.5 to 14.0 μg, the dried fish is 0.5 to 6.0 μg, the fatty fish is 3.0 to 16.0 μg, the lean fish is 4.0 to 32.0 μg, and the egg is 0.5 to 4.0 μg.

[0018] The cognitive function evaluation system or the serum vitamin D value estimation system according to the present invention preferably includes body fat information including BMI (Body Mass Index) and body fat percentage, or either of them.

[0019] The cognitive function evaluation system or the serum vitamin D value estimation system according to the present invention preferably includes ultraviolet exposure information including information on ultraviolet exposure countermeasures.

[0020] In order to solve the above problems, the cognitive function evaluation program according to the present invention operates a computer as the above evaluation and estimation unit.

[0021] In order to solve the above problems, the serum vitamin D value estimation program according to the present invention operates a computer as the above vitamin D value estimation unit.

[0022] In order to solve the above problems, the cognitive function evaluation method according to the present invention pre-learns the correlation between the serum vitamin D value, information on body fat, gender, age, and the value related to cognitive function by machine learning in advance, and includes estimating the value related to cognitive function using an evaluation and estimation unit which is the learned model obtained by the above machine learning.

[0023] The serum vitamin D value estimation method according to the present invention Information on the frequency of vitamin D intake during a predetermined period (hereinafter referred to as vitamin D intake information), information on ultraviolet ray exposure during the predetermined period, information on gender and body fat are used as input data, and machine learning is performed in advance with teacher data having a serum vitamin D value as output data, and estimating the serum vitamin D value using a vitamin D value estimation unit which is a learned model obtained by the above machine learning.

[0024] The method for estimating a serum vitamin D value according to the present invention includes calculating a serum vitamin D value by a predetermined formula using information on the frequency of vitamin D intake during a predetermined period (hereinafter referred to as vitamin D intake information), information on ultraviolet ray exposure during the predetermined period, gender, and information on body fat.

[0025] The method for estimating a serum vitamin D value preferably wherein the vitamin D intake information includes the intake frequencies of vitamin D supplements, multivitamin supplements containing vitamin D, fish eaten with bones, dried fish, fatty fish, and lean fish during a predetermined period.

Advantages of the Invention

[0026] The cognitive function evaluation system according to the present invention can accurately substitute for the evaluation regarding cognitive function.

Brief Description of the Drawings

[0027]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0028] Hereinafter, an embodiment of the cognitive function evaluation system according to the present invention will be described with reference to the accompanying drawings.

[0029] As shown in FIG. 1, the cognitive function evaluation system S according to the present embodiment includes an electronic device D. The cognitive function evaluation system S is configured to output an evaluation (value) regarding the cognitive function based on the operation of the user U.

[0030] <Hardware Configuration of the Electronic Device> First, the hardware configuration of the electronic device D will be described. As shown in FIG. 2, the electronic device D includes an input means 1, a display means 2, and a computer (control unit) 3. In the present embodiment, the electronic device D is a smartphone, but this is merely an example and is not limited thereto. For example, the electronic device D may be configured by a tablet or a personal computer.

[0031] The input means 1 receives the input operation of the user U. As shown in FIG. 1, the input means 1 in the present embodiment is a touch panel and buttons, but this is merely an example and is not limited thereto. For example, when the electronic device D is configured by a personal computer, the input means 1 may be configured by a keyboard, a mouse, etc.

[0032] As shown in FIG. 1, the display means 2 is configured by a liquid crystal display. However, this is also merely an example. For example, when the electronic device D is configured by a personal computer, the display means 2 may be configured by a monitor connected to the personal computer.

[0033] The computer 3 has a storage means, an arithmetic means, and a memory.

[0034] The storage means is a flash memory built in the electronic device D, but this is merely an example and is not limited thereto. For example, the storage means may be an external memory such as a USB memory or an SD card, or may be configured by so-called online storage. The storage means stores an OS (Operation System), applications, and other programs. The applications include a serum vitamin D value estimation program and a cognitive function evaluation program that operate the computer 3 as a vitamin D value estimation unit 35 and an evaluation estimation unit 36 to be described later. The arithmetic means consists of a CPU (Central Processing Unit) or the like, and executes each program in the application based on the input operation of the user U.

[0035] <Functional configuration of the cognitive function evaluation system> Next, the functional configuration of the cognitive function evaluation system S will be described. The cognitive function evaluation system S includes a control unit 3. As shown in FIG. 2, the control unit 3 has a display unit 30, an input unit 31, an age calculation unit 32, a BMI calculation unit 33, a storage unit 34, a vitamin D value estimation unit 35, an evaluation estimation unit 36, and an advice unit 37.

[0036] As shown in FIG. 3, the display unit 30 causes the display means 2 to display each question. In the present embodiment, the display unit 30 causes the display means 2 to display questions regarding gender, date of birth, height [cm], weight [kg], muscle mass [%], body fat percentage [%], frequency information on vitamin D intake in the past week (hereinafter referred to as vitamin D intake information), and ultraviolet ray exposure information in the past week (hereinafter referred to as ultraviolet ray exposure information).

[0037] Questions regarding vitamin D intake information include questions about the frequency of taking supplements containing vitamin D and the frequency of eating fish. Questions (1) and (2) in Figure 3 are about the frequency of taking supplements in the past week, and questions (3) to (6) are about the frequency of eating fish in the past week.

[0038] Also, question (7) in Figure 3 is about the frequency of exposure to ultraviolet rays in the past week, and question (8) is about the frequency of taking measures against ultraviolet ray exposure in the past week. In question (8), sunscreen is indicated as "apply", and this sunscreen includes spray type. Also, the measures against ultraviolet ray exposure in the present invention are not limited to "sunscreen", and for example, measures against ultraviolet ray exposure such as "gloves" and "hats" with large brims may be included, and these measures against ultraviolet ray exposure may be included in the questions.

[0039] As shown in Figure 4, the input unit 31 receives the input of gender, date of birth, height [cm], weight [kg], muscle mass [%], body fat percentage [%], vitamin D intake information, and ultraviolet ray exposure information via the input means 1. Each input information is stored in the storage unit 34. In the present embodiment, each vitamin D intake information and each ultraviolet ray exposure information are configured to be input by the user U selecting the number of stars in the range of 1 to 5. However, this is also merely an example, and the input method of vitamin D intake information and ultraviolet ray exposure information is not limited thereto.

[0040] Note that the predetermined period in questions (1) to (8) of the present embodiment is the past week, but this is merely an example and is not limited thereto. For example, the predetermined period may be set to 3 days or more, or may be configured such that the user U can select an arbitrary period.

[0041] The age calculation unit 32 calculates the age of the user U based on the input birthday. The calculated age is transmitted to the evaluation estimation unit 36. Note that the input unit 31 may be configured to directly input the age by the user U. In this case, the cognitive function evaluation system S may not include the age calculation unit 32.

[0042] The BMI calculation unit 33 calculates the BMI (Body Mass Index) according to the following formula (1) based on the input height and weight. The calculated BMI is transmitted to the vitamin D value estimation unit 35 and the evaluation estimation unit 36. Note that the input unit 31 may be configured to directly input the BMI by the user U. In this case, the cognitive function evaluation system S may not include the BMI calculation unit 33. Formula (1) ··· {weight (kg)} ÷ {square of height (m)}

[0043] The storage unit 34 stores coefficients (vitamin D content) corresponding to each of the questions (1) to (6). The vitamin D supplement for question (1) is set in the range of 2.5 to 10.0 μg, the multivitamin supplement for question (2) is set in the range of 1.0 to 5.0 μg, the fish eaten with bones for question (3) is set in the range of 1.5 to 14.0 μg, the dried fish for question (4) is set in the range of 0.5 to 6.0 μg, the fatty fish for question (5) is set in the range of 3.0 to 16.0 μg, and the lean fish for question (6) is set in the range of 4.0 to 32.0 μg.

[0044] In the present embodiment, among the coefficients corresponding to each of the questions (1) to (6), the vitamin D supplement for question (1) is 5 μg, the multivitamin supplement for question (2) is 2.5 μg, the fish eaten with bones for question (3) is 5 μg, the dried fish for question (4) is 3 μg, the fatty fish for question (5) is 15 μg, and the lean fish for question (6) is 16 μg. As described above, the vitamin D value of the vitamin D supplement is set to be larger than that of the multivitamin supplement, and the vitamin D values (coefficients) related to the intake of fish are set to be larger in the order of lean fish, fatty fish, fish eaten with bones, and dried fish.

[0045] When the vitamin D value estimation unit 35 is input with body fat information, gender, vitamin D intake information, and ultraviolet exposure information, it calculates the serum vitamin D value according to the following formulas (2) to (5). The correspondence between the number of stars related to the input of questions, the display of the display means 2 corresponding to the number of stars, each frequency in formula (2), and the average daily intake vitamin D value [μg] of questions (1) to (6) is shown in Table 1 below. Formula (2) ··· The sum of {each frequency × each coefficient (vitamin D content)} (the sum of each average daily intake vitamin D value) = a Formula (3) ··· (The number of stars in question (7) - the number of stars in question (8)) × 0.25 × BMI × 0.05 = b Formula (4) ··· In the case of men (body fat percentage + BMI) × 0.1 = c ··· In the case of women (body fat percentage + BMI) × 0.1 ÷ 2 = c Formula (5) ··· a + b + c = serum vitamin D value [ng / ml]

[0046]

Table 1

[0047] In the case of the input example shown in FIG. 4, the vitamin D value estimation unit 35 calculates the serum vitamin D value to be 29.4 [ng / ml] according to the above formulas (2) to (5). Note that the value (estimated value) calculated by the vitamin D value estimation unit 35 had a positive correlation when compared with the actually measured serum vitamin D value, as shown in FIG. 5. The calculated serum vitamin D value is transmitted to the evaluation estimation unit 36.

[0048] Note that an upper limit may be provided for the serum vitamin D value estimated by the vitamin D value estimation unit 35. In this case, the upper limit of the serum vitamin D value may be set, for example, in the range of 35 to 45 [ng / ml]. Thereby, in the evaluation estimation unit 36 described later, an evaluation (value) regarding the recognition function can be output more appropriately. Also, the upper limit value may be set to be changed according to the residence and gender of the user U.

[0049] The evaluation and estimation unit 36 has previously performed machine learning using a deep neural network with teacher data that takes the serum vitamin D value, body fat information, gender, and age as input data and the test result (value) of the MMSE as output data. Thereby, when the serum vitamin D value, body fat information, gender, and age are input, the evaluation and estimation unit 36 estimates the test result (value) of the MMSE. The body fat information preferably includes both the BMI and the body fat percentage, but may include only either the BMI or the body fat percentage.

[0050] Also, the evaluation and estimation unit 36 has previously performed machine learning using a deep neural network with teacher data that takes the serum vitamin D value, body fat information, gender, and age as input data and the test results (values) of the MoCA-J, grip strength, SMI (Skeletal Muscle Index), and WHOQOL (World Health Organization Quality of Life) as output data. Thereby, when the serum vitamin D value, body fat information, gender, and age are input, the evaluation and estimation unit 36 estimates not only the test result (value) of the MMSE but also the test results (values) by the MoCA-J, grip strength, SMI, and WHOQOL.

[0051] Note that grip strength is known to be an indicator of cognitive function, and the "value related to cognitive function" in the present invention includes "grip strength". Also, the SMI (skeletal muscle index) is said to be correlated with cognitive function, and the "value related to cognitive function" in the present invention includes the test result (value) by the "SMI". Furthermore, it is known that the QOL of the elderly is affected by the cognitive function of the patient himself / herself, and the "value related to cognitive function" in the present invention includes the test result (value) by the "WHOQOL".

[0052] The accuracy of each result estimated by this evaluation and estimation unit 36 is shown in Table 2 below.

[0053]

Table 2

[0054] As shown in Table 2, the evaluation and estimation unit 36 estimated each value with high accuracy in the validation set and the test set. Only the result of MoCA-J was not as expected, which may be affected by the ability of the MoCA-J measurer. Note that the evaluation and estimation unit 36 used TensorFlow as a deep learning tool, but it is not limited to this.

[0055] The storage unit 34 stores a plurality of advices selected by the advice unit 37. The plurality of advices may include advice on whether to visit a hospital, in addition to advice on vitamin D intake and ultraviolet exposure.

[0056] The advice unit 37 selects advice on vitamin D intake and ultraviolet exposure based on the serum vitamin D value calculated by the vitamin D value estimation unit 35. In addition, the advice unit 37 selects advice on whether to visit a hospital based on each value output by the evaluation and estimation unit 36. The advice unit 37 may be set to select advice on exercise and diet for burning body fat, for example, when the vitamin D value is low and the input body fat percentage is high.

[0057] As shown in FIG. 6, the display unit 30 causes the display means 2 to display the data input by the user U, each value estimated by the evaluation and estimation unit 36, and the advice. Thereby, the user U can alternatively recognize the evaluation regarding the conventional cognitive function. Moreover, the user can know how to improve the future life from the displayed advice.

[0058] Although not shown in FIG. 6, the display unit 30 causes the display means 2 to display the serum vitamin D value estimated by the vitamin D value estimation unit 35 and the advice on vitamin D intake and ultraviolet exposure selected by the advice unit 37. As a result, the user U can recognize the adequacy of the serum vitamin D value and, if the serum vitamin D is insufficient, the method for improving it.

[0059] Next, with reference to FIG. 7, the operation of the cognitive function evaluation system S will be described again.

[0060] (1) First, the cognitive function evaluation system S receives the input of vitamin D intake information, ultraviolet exposure information, age, gender, BMI, and body fat percentage by the input unit 31. The input information is stored in the storage unit 34 and transmitted to the vitamin D value estimation unit 35 and the evaluation estimation unit 36, respectively.

[0061] (2) Next, the cognitive function evaluation system S estimates the serum vitamin D value by the vitamin D value estimation unit 35 based on the vitamin D intake information, ultraviolet exposure information, gender, BMI, and body fat percentage. The estimated serum vitamin D value is transmitted to the evaluation estimation unit 36.

[0062] (3) Next, the cognitive function evaluation system S estimates the test results (values) by MMSE, MoCA-J, grip strength, SMI, and WHOQOL based on the estimated serum vitamin D value, age, gender, BMI, and body fat percentage by the evaluation estimation unit 36.

[0063] (4) Next, the cognitive function evaluation system S selects appropriate advice by the advice unit 37 based on the estimated serum vitamin D value and the test results (values) related to cognitive function.

[0064] (5) Next, the cognitive function evaluation system S causes the display unit 30 to display the input data, each estimation result, and the advice.

[0065] As a result, the cognitive function evaluation system S can accurately substitute for the evaluation of cognitive function without relying on experts. Consequently, the cognitive function evaluation system S can enable the user U to recognize each test result regarding cognitive function without incurring costs or taking time. Moreover, since the cognitive function evaluation system S also estimates the serum vitamin D level, the user U can recognize the serum vitamin D level without undergoing blood sampling. Furthermore, by outputting advice based on each estimated result, the cognitive function evaluation system S can also enable the user U to recognize improvement methods for each result.

[0066] As described above, the cognitive function evaluation system S according to one embodiment of the present invention has been explained. However, the cognitive function evaluation system according to the present invention is not limited to the above embodiment. For example, the cognitive function evaluation system according to the present invention may be implemented by combining or replacing each component with the following modification examples.

[0067] <Modification Example> · The cognitive function evaluation system S may further include a server computer. In this case, the evaluation estimation unit 36 and the vitamin D level estimation unit 35 may be constituted by the server computer. Then, each value to be input to the vitamin D level estimation unit 35 and the evaluation estimation unit 36 is transmitted from the electronic device D to this server computer through the network, and the values estimated by the evaluation estimation unit 36 and the vitamin D level estimation unit 35 are transmitted from the server computer to the electronic device D.

[0068] · The cognitive function evaluation system S may include a serum vitamin D level estimation system for estimating the serum vitamin D level with an electronic device separate from the electronic device D. In this case, this separate electronic device may be a smartphone, a personal computer, or an individual device provided with the input unit 31 and the vitamin D level estimation unit 35. Also, this input unit 31 receives the input of body fat information, gender, vitamin D intake information, and ultraviolet exposure information.

[0069] ·The cognitive function evaluation system S may ask the user U questions and obtain answers to the questions by voice. In this case, the cognitive function evaluation system S further includes a microphone and a speaker. Also, in this case, the cognitive function evaluation system S may output the values and advice estimated by the evaluation estimation unit 36 and the vitamin D value estimation unit 35 by voice.

[0070] ·Based on the gender, birthday, height, weight, and body fat percentage of the user U stored in the storage unit 34, the cognitive function evaluation system S may omit all or any of the inputs of gender, birthday, height, weight, and body fat percentage. In this case, the gender, birthday (or age), height, etc. stored are displayed on the screen related to the question.

[0071] ·The vitamin D value estimation unit 35 may be pre-trained by machine learning using teacher data with body fat information, gender, vitamin D intake information, and ultraviolet exposure information as input data and the serum vitamin D value as output data. In this case, when the body fat information, gender, vitamin D intake information, and ultraviolet exposure information are input, the vitamin D value estimation unit 35 estimates the serum vitamin D value without using the above formulas (3), (4), and (5). Note that the machine learning method in this modification example may be a deep neural network, but this is merely an example and is not limited thereto.

[0072] And the teacher data for machine learning in this case is as follows. The body fat information may be BMI and / or body fat percentage. The vitamin D intake information may be the average daily vitamin D intake value [μg] per day in a predetermined period calculated using formula (2). The ultraviolet exposure information may be the number of days of ultraviolet exposure in a predetermined period calculated by the following formula (6). Also, the predetermined period in these may be 7 days as in the above embodiment, but is not limited thereto. Formula (6) ··· Answer to question (7) - Answer to question (8)

[0073] · Vitamin D intake information may include egg intake information. In this case, the coefficient (vitamin D content) regarding egg intake may be set in the range of 0.5 to 4.0 μg. Also, questions regarding vitamin D intake information may include questions regarding vitamin D intake information of other foods such as dried shiitake mushrooms.

[0074] · The evaluation estimation unit 36 may estimate an evaluation regarding the cognitive function of the user U by using the serum vitamin D value actually measured through blood sampling instead of the serum vitamin D value estimated by the vitamin D value estimation unit 35. In this case, the input unit 31 receives the input of the actually measured serum vitamin D value. Also, in this case, the cognitive function evaluation system S may not include the vitamin D value estimation unit 35.

[0075] · As described above, the information on body fat percentage input to the evaluation estimation unit 36 may be either BMI or body fat percentage. In this case, the evaluation estimation unit 36 uses, as input data, the serum vitamin D value, gender, age, and BMI or body fat percentage, and performs machine learning by a deep neural network in advance using teacher data with the test results (values) of MMSE, MoCA-J, grip strength, SMI, and WHOQOL as output data. Then, when the serum vitamin D value, gender, age, and BMI or body fat percentage are input, the evaluation estimation unit 36 estimates the test results (values) by MMSE, MoCA-J, grip strength, SMI, and WHOQOL.

[0076] · The display method by the display unit 30 for each estimation result is not limited to characters. For example, the display unit 30 may display each estimation result by an image corresponding to each estimated value. In this case, the image may include clear, cloudy, and rainy weather. And, for example, an image of clear sky may be used when the serum vitamin D value is 30 [ng / ml] or more, an image of cloudy when it is 20 [ng / ml] or more and less than 30 [ng / ml], and an image of rain when it is less than 20 [ng / ml]. Also, for example, an image of clear sky may be used when the test result of MMSE is 27 to 30 points, an image of cloudy when it is 22 to 26 points, and an image of rain when it is 21 points or less.

[0077] ·The display unit 30 may change the color of the characters, the background color, and the image color to be displayed on the display means 2 based on the values estimated by the vitamin D value estimation unit 35 and the evaluation estimation unit 36.

Explanation of Signs

[0078] S Cognitive Function Evaluation System D Electronic Device U User 1 Input Means 2 Display Means 3 Control Unit (Computer) 30 Display Unit 31 Input Unit 32 Age Calculation Unit 33 BMI Calculation Unit 34 Memory Unit 35 Vitamin D Value Estimation Unit 36 Evaluation Estimation Unit 37 Advice Unit

Claims

1. Comprising an evaluation and estimation unit, The evaluation and estimation unit has been pre-trained by machine learning using teacher data with the serum vitamin D value, information on body fat (hereinafter referred to as body fat information), gender, and age as input data and a value related to cognitive function as output data. When the serum vitamin D value, the body fat information, the gender, and the age are input, it estimates the value related to cognitive function. A cognitive function evaluation system.

2. The value related to cognitive function includes at least one of the values corresponding to the test results of MMSE (Mini Mental State Examination), MoCA-J (Montreal Cognitive Assessment-Japanese), grip strength, SMI (Skeletal Muscle Index), and WHOQOL (World Health Organization Quality of Life). The cognitive function evaluation system according to Claim 1.

3. Further comprising a vitamin D value estimation unit, The vitamin D value estimation unit has been pre-trained by machine learning using teacher data with the body fat information, the gender, information on the frequency of vitamin D intake during a predetermined period (hereinafter referred to as vitamin D intake information), and information on the frequency of ultraviolet ray exposure during a predetermined period (hereinafter referred to as ultraviolet ray exposure information) as input data and the serum vitamin D value as output data. When the body fat information, the gender, the vitamin D intake information, and the ultraviolet ray exposure information are input, it estimates the serum vitamin D value. The evaluation and estimation unit estimates the value related to cognitive function with the serum vitamin D value estimated by the vitamin D value estimation unit input thereto. The cognitive function evaluation system according to Claim 1.

4. Further comprising a vitamin D value estimation unit, The vitamin D value estimation unit calculates the serum vitamin D value by a predetermined formula using the body fat information, the gender, information on the frequency of vitamin D intake during a predetermined period (hereinafter referred to as vitamin D intake information), and information on the frequency of ultraviolet ray exposure during a predetermined period (hereinafter referred to as ultraviolet ray exposure information). The evaluation and estimation unit estimates the value related to cognitive function with the serum vitamin D value calculated by the vitamin D value estimation unit input thereto. The cognitive function evaluation system according to Claim 1.

5. Comprising a vitamin D value estimation unit, The vitamin D value estimation unit has been pre-trained by machine learning using teacher data with information on body fat (hereinafter referred to as body fat information), gender, information on the frequency of vitamin D intake during a predetermined period (hereinafter referred to as vitamin D intake information), and information on the frequency of ultraviolet exposure during a predetermined period (hereinafter referred to as ultraviolet exposure information) as input data, and outputting the serum vitamin D value as output data. When the body fat information, the gender, the vitamin D intake information, and the ultraviolet exposure information are input, it is a serum vitamin D value estimation system that estimates the serum vitamin D value.

6. Comprising a vitamin D value estimation unit, The vitamin D value estimation unit is a serum vitamin D value estimation system that calculates the serum vitamin D value by a predetermined formula using information on body fat (hereinafter referred to as body fat information), gender, information on the frequency of vitamin D intake during a predetermined period (hereinafter referred to as vitamin D intake information), and information on the frequency of ultraviolet exposure during a predetermined period (hereinafter referred to as ultraviolet exposure information).

7. The cognitive function evaluation system or serum vitamin D value estimation system according to any one of claims 3 to 6, wherein the vitamin D intake information includes information on the intake frequency of supplements containing vitamin D and fish during a predetermined period.

8. The cognitive function evaluation system or serum vitamin D value estimation system according to any one of claims 3 to 6, wherein the vitamin D intake information includes the intake frequency of vitamin D supplements, multivitamin supplements containing vitamin D, fish eaten with bones, dried fish, fatty fish, and lean fish during a predetermined period.

9. The vitamin D intake information includes the intake frequency of vitamin D supplements, multivitamin supplements containing vitamin D, fish eaten with bones, dried fish, fatty fish, and lean fish during a predetermined period. In terms of the vitamin D content, the vitamin D supplement is set in the range of 2.5 to 10.0 μg, the multivitamin supplement is 1.0 to 5.0 μg, the fish eaten with bones is 1.5 to 14.0 μg, the dried fish is 0.5 to 6.0 μg, the fatty fish is 3.0 to 16.0 μg, and the lean fish is 4.0 to 32.0 μg. The cognitive function evaluation system or serum vitamin D value estimation system according to claim 4 or 6.

10. The vitamin D intake information includes the intake frequency of vitamin D supplements, multivitamin supplements containing vitamin D, fish eaten with bones, dried fish, fatty fish, lean fish, and eggs over a predetermined period. In each vitamin D content related to the calculation of the serum vitamin D value, the vitamin D supplement is set in the range of 2.5 to 10.0 μg, the multivitamin supplement is 1.0 to 5.0 μg, the fish eaten with bones is 1.5 to 14.0 μg, the dried fish is 0.5 to 6.0 μg, the fatty fish is 3.0 to 16.0 μg, the lean fish is 4.0 to 32.0 μg, and the egg is 0.5 to 4.0 μg. The cognitive function evaluation system or serum vitamin D value estimation system according to claim 4 or 6.

11. The body fat information includes BMI (Body Mass Index) and body fat percentage, or either of them. The cognitive function evaluation system or serum vitamin D value estimation system according to any one of claims 1 to 3 and 5.

12. The ultraviolet ray exposure information includes information related to ultraviolet ray exposure countermeasures. The cognitive function evaluation system or serum vitamin D value estimation system according to any one of claims 3 to 6.

13. A cognitive function evaluation program for operating a computer as the evaluation and estimation unit according to any one of claims 1 to 4.

14. A serum vitamin D value estimation program for operating a computer as the vitamin D value estimation unit according to any one of claims 3 to 6.

15. Previously performing machine learning on the correlation between the serum vitamin D value, information related to body fat, gender, age, and a value related to cognitive function. Estimating a value related to cognitive function using an evaluation and estimation unit that is a learned model obtained by the machine learning. A cognitive function evaluation method including this.

16. Previously performing machine learning using, as input data, information on the frequency of vitamin D intake over a predetermined period (hereinafter referred to as vitamin D intake information), information on ultraviolet ray exposure over a predetermined period, gender, and information related to body fat, and using, as output data, the serum vitamin D value, with teacher data. Estimating the serum vitamin D value using a vitamin D value estimation unit that is a learned model obtained by the machine learning. A serum vitamin D value estimation method including this.

17. A method for estimating serum vitamin D levels, including calculating serum vitamin D levels using information on the frequency of vitamin D intake during a predetermined period (hereinafter referred to as vitamin D intake information), information on ultraviolet exposure during a predetermined period, gender, and information on body fat according to a predetermined formula.

18. The method for estimating serum vitamin D levels according to claim 16 or 17, wherein the vitamin D intake information includes the intake frequencies of vitamin D supplements, multivitamin supplements containing vitamin D, fish eaten with bones, dried fish, fatty fish, and lean fish during a predetermined period.

Images