Information processing device, alcohol intake estimation device, method for calculating indicator for estimating alcohol intake, alcohol intake estimation method, biological examination method, biological examination device, estimation device, learning model, estimation method, learning model generation method, and computer program

Abstract

The present disclosure provides an information processing device, alcohol intake estimation device, method for calculating an indicator for estimating alcohol intake, alcohol intake estimation method, biological examination method, biological examination device, estimation device, learning model, estimation method, learning model generation method, and computer program. The information processing device comprises a processing unit that acquires a total heart rate change representing the total sum of heart rate changes, which are amounts by which a subject's heart rate has changed. The processing unit calculates a first indicator for estimating alcohol intake, said indicator representing the total heart rate change in a state in which the subject is inactive and a state in which the heart rate of the subject exceeds a reference heart rate.

Description

Information processing apparatus, alcohol intake estimation apparatus, method for calculating alcohol intake estimation index, alcohol intake estimation method, biological examination method, biological examination apparatus, estimation apparatus, learning model, estimation method, learning model generation method, and computer program

[0001] The present disclosure relates to an information processing apparatus, an alcohol intake estimation apparatus, a method for calculating an alcohol intake estimation index, an alcohol intake estimation method, a biological examination method, a biological examination apparatus, an estimation apparatus, a learning model, an estimation method, a learning model generation method, and a computer program. [[ID=z4]]

[0002] The information processing apparatus and the like described in Patent Document 1 are intended to detect alcohol beverage intake with high accuracy. In relation to the above object, Patent Document 1 discloses, for example, that the change over time in the heart rate related to alcohol beverage intake has the characteristics of maintaining a relatively high value after the end of alcohol beverage intake, and decreasing after an activity that moves the body such as walking of the subject (paragraphs 0018 and 0019 of Patent Document 1, FIG. 4), and considering the decrease width of the heart rate after the walking of the subject is detected in the characteristic amount of the heart rate related to alcohol beverage intake (claims 1 and 7 of Patent Document 1).

[0003] Japanese Patent No. 6547837

[0004] However, as described above, even if the information processing apparatus and the like consider the decrease width of the heart rate after the walking of the subject is detected, the detection of alcohol beverage intake is basically based on what the characteristic amount of the heart rate of the subject is like (for example, whether it is rising, falling, an instantaneous sharp rise, or a sharp rise at a high frequency (paragraphs 0018 to 0020 of Patent Document 1)). Therefore, the information processing apparatus and the like cannot consider the possible influence of the walking of the subject on the heart rate of the subject immediately, that is, the heart rate obtained from the subject (claim 1 of Patent Document 1) is affected by the walking of the subject, and there is a possibility that the accuracy of the detection may decrease. For this reason, there is also a problem that reliable information cannot always be obtained when using the detection result of alcohol beverage intake.

[0005] The purpose of this disclosure is to estimate the alcohol consumption of subjects, taking into account the potential immediate effects of their physical activity on their heart rate.

[0006] The purpose of this disclosure is to provide a simpler method for estimating a subject's alcohol consumption than conventional methods, either as an alternative to or in addition to the above-mentioned purpose.

[0007] Another purpose of this disclosure is to obtain reliable information by utilizing subject alcohol consumption estimation index information obtained by taking into account the potential immediate effects of a subject's physical activity on their heart rate.

[0008] According to this disclosure, an information processing device is provided which includes a calculation unit that calculates a total heart rate change amount, which is the sum of the heart rate change amounts of a subject, and the calculation unit calculates a first index for estimating alcohol consumption, which represents the total heart rate change amount when the subject is inactive and the subject's heart rate exceeds a reference heart rate.

[0009] According to this disclosure, an estimation device is provided for estimating biological state information of a subject, comprising: an estimation unit that inputs input information to a learning model and acquires output information from the learning model; and a storage unit that stores the output information, wherein the output information includes the biological state information, the biological state information includes information representing the physical and mental state of the subject, information representing the state of the environment that affects the physical and mental state of the subject, information representing the state of the body affected by the physical and mental state of the subject, or information regarding the subject's alcohol consumption, the input information includes at least an index information for estimating alcohol consumption, the index information for estimating alcohol consumption includes information regarding the total change in heart rate in a state inactive of the subject and in a state in which the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

[0010] The present disclosure provides a learning model for causing a computer to function to estimate bio-related state information of a subject, wherein the computer is configured to take input information as input and output information as output, the output information includes the bio-related state information, the bio-related state information includes information representing the physical and mental state of the subject, information representing the state of the environment affecting the physical and mental state of the subject, information representing the state of the body affected by the physical and mental state of the subject, or information regarding the subject's alcohol consumption, the input information includes at least an index information for estimating alcohol consumption, the index information for estimating alcohol consumption includes information regarding the total change in heart rate in a state where the subject is inactive and the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

[0011] According to this disclosure, an estimation method for estimating biological state information of a subject is provided, comprising the steps of: inputting input information into a learning model; and obtaining output information from the learning model into which the input information has been input, wherein the output information includes the biological state information, the biological state information includes information representing the physical and mental state of the subject, information representing the state of the environment affecting the physical and mental state of the subject, information representing the biological state affected by the physical and mental state of the subject, or information regarding the subject's alcohol consumption, the input information includes at least an index information for estimating alcohol consumption, the index information for estimating alcohol consumption includes information regarding the total change in heart rate in a state inactive of the subject and in a state in which the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

[0012] According to this disclosure, a computer program is provided that causes a computer to perform the above estimation method.

[0013] According to this disclosure, the process includes the steps of acquiring a training dataset and generating a learning model that outputs output information when input information is input by performing training using the training dataset, wherein the training dataset includes at least indicator information for estimating the drinking of a training subject and bio-related state information of the training subject, the indicator information for estimating the drinking of a training subject includes information on the total change in heart rate when the training subject is inactive and the training subject's heart rate exceeds a reference heart rate, the total change in heart rate of the training subject represents the sum of heart rate changes which are changes in the training subject's heart rate, and the bio-related state information of the training subject includes information representing the physical and mental state of the training subject, information representing the state of the environment that affects the physical and mental state of the training subject, and the training subject A learning model generation method is provided, wherein the output information includes information representing the state of a living organism affected by the physical and mental state of the body, or information regarding the alcohol consumption of the learning subject, the output information includes biological state information of the subject, the biological state information of the subject includes information representing the physical and mental state of the subject, information representing the state of the environment affecting the physical and mental state of the subject, information representing the state of a living organism affected by the physical and mental state of the subject, or information regarding the alcohol consumption of the subject, the input information includes at least index information for estimating the alcohol consumption of the subject, the index information for estimating the alcohol consumption of the subject includes information regarding the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate, and the total change in heart rate of the subject represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

[0014] According to this disclosure, a computer program is provided that causes a computer to execute the above-described learning model generation method.

[0015] According to this disclosure, it is possible to estimate a subject's alcohol consumption while taking into account the potential immediate effects that walking may have on the subject's heart rate.

[0016] According to this disclosure, reliable information can be obtained by utilizing an index for estimating a subject's alcohol consumption, which takes into account the potential immediate impact of a subject's physical activity on their heart rate.

[0017] This diagram shows the configuration of the SKS biomedical testing system of Embodiment 1. This diagram shows the configuration of the WT wearable terminal of Embodiment 1. This diagram shows the configuration of the IS alcohol consumption estimation device of Embodiment 1. This diagram shows the configuration of the SK biomedical testing device of Embodiment 1. This diagram is a flowchart showing the operation of the SKS biomedical testing system of Embodiment 1. This is a time chart (part 1) showing the operation of the SKS biomedical testing system of Embodiment 1. This is a time chart (part 2) showing the operation of the SKS biomedical testing system of Embodiment 1. This is a time chart (part 3) showing the operation of the SKS biomedical testing system of Embodiment 1. This is a time chart (part 4) showing the operation of the SKS biomedical testing system of Embodiment 1. This is a time chart (part 5) showing the operation of the SKS biomedical testing system of Embodiment 1. This is a time chart (part 6) showing the operation of the SKS biomedical testing system of Embodiment 1. This is a time chart (part 7) showing the operation of the SKS biomedical testing system of Embodiment 1. This is a time chart (part 8) showing the operation of the SKS biomedical testing system of Embodiment 1. This is a time chart (part 9) showing the operation of the SKS biomedical testing system of Embodiment 1. This is a time chart showing the operation of the SKS biomedical examination system of Embodiment 3. This shows the hardware configuration of the SKS biomedical examination system of Embodiments 1 to 3. This shows the hardware configuration based on the software implementation of the SKS biomedical examination system of Embodiments 1 to 3. This is a flowchart showing an example of alcohol consumption estimation processing according to one embodiment. This is a flowchart showing an example of biomedical examination processing according to one embodiment. This is a block diagram showing an example of the configuration of an information processing system according to Embodiment 4. This is a diagram showing an example of a learning dataset according to Embodiment 4. This is a diagram showing an example of input information and output information according to Embodiment 4. This is a block diagram showing an example of the configuration of an estimation system according to Embodiment 4. This is a flowchart showing an example of an estimation method executed by the estimation system according to Embodiment 4. This is a block diagram showing an example of the configuration of an estimation device according to Embodiment 4. This is a diagram schematically showing an example of a deep neural network according to Embodiment 4. This is a diagram schematically showing another example of a deep neural network according to Embodiment 4. This is a diagram schematically showing yet another example of a deep neural network according to Embodiment 4. This is a flowchart showing an example of estimation processing according to Embodiment 4.This is a block diagram showing an example configuration of a learning device according to Embodiment 4. This is a flowchart showing an example of a learning method using the learning device according to Embodiment 4. This is a block diagram showing an example configuration of a learning data creation device according to Embodiment 4. This is a flowchart showing an example of a learning data creation method using the learning data creation device according to Embodiment 4.

[0018] Preferred embodiments of this disclosure will be described in detail below with reference to the attached drawings. In this specification and the drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant descriptions will be omitted.

[0019] Embodiment 1 of the biological testing system according to this disclosure will be described. <Embodiment 1> The biological testing system SKS of Embodiment 1 will be described. <Configuration of Embodiment 1> <Configuration of the biological testing system SKS> Figure 1 shows the configuration of the biological testing system SKS of Embodiment 1.

[0020] The first embodiment of the biomedical examination system SKS includes, as shown in Figure 1, wearable terminals WT1 to WTm (where m is an integer of 2 or more), an alcohol consumption estimation device IS, and a biomedical examination device SK. The wearable terminals WT1 to WTm, the alcohol consumption estimation device IS, and the biomedical examination device SK are interconnected via a network NW (e.g., the Internet), as shown in Figure 1. The alcohol consumption estimation device IS corresponds to an example of the "information processing device" of this disclosure.

[0021] In the following, for the sake of ease of explanation and understanding, for example, the same name and multiple codes may be collectively referred to as one name and one code. For example, wearable terminals WT1 to WTm may be collectively referred to as wearable terminal WT.

[0022] Wearable devices WT1 to WTm are equipped with functions to measure, for example, heart rate HR(t) and step count ST(t) (both illustrated, for example, in Figure 6), and are detachable from the arm, head, etc. Wearable devices WT1 to WTm are used by subjects HI1 to HIm, for example, subjects whose presence or absence of alcohol consumption and amount of alcohol consumption are to be estimated. For example, wearable device WT1 is used by subject HI1, wearable device WT2 is used by subject HI2, ... wearable device WTm is used by subject HIm.

[0023] Here, "heart rate" is a broad concept that includes, for example, "pulse." A subject is an example of a "subject" in this disclosure.

[0024] The alcohol consumption estimation device IS is used by administrator KA, who manages the alcohol consumption estimation device IS.

[0025] The biomedical examination device SK is used, for example, by an examiner KE to examine the biological state of subjects HI1 to HIm.

[0026] <Configuration of Wearable Terminal WT> Figure 2 shows the configuration of the wearable terminal WT of Embodiment 1.

[0027] The wearable terminal WT of Embodiment 1, as shown in Figure 2, includes an input / output unit NS(WT), a processing unit SY(WT), a storage unit KI(WT), and a communication unit TU(WT).

[0028] The input / output unit NS (WT) includes conventionally known sensors and acquires the subject HI's biological information SJ (for example, heart rate HR, step count ST (shown in Figure 6), blood pressure KE, respiratory rate KO, and electroencephalogram NO (shown in Figure 15)).

[0029] The processing unit SY(WT) performs, for example, processing related to the subject HI's biological information SJ.

[0030] The memory unit KI(WT) stores, for example, the data necessary for processing by the processing unit SY(WT).

[0031] The communications unit TU (WT) communicates via the network NW (shown in Figure 1) and, for example, transmits the subject HI's biological information SJ, as described above, to the alcohol consumption estimation device IS and the examiner KE.

[0032] <Configuration of the alcohol consumption estimation device IS> Figure 3 shows the configuration of the alcohol consumption estimation device IS of Embodiment 1.

[0033] The alcohol consumption estimation device IS of Embodiment 1 includes an input / output unit NS (IS), a processing unit SY (IS), a storage unit KI (IS), and a communication unit TU (IS), as shown in Figure 3.

[0034] The input / output unit NS (IS) is used by administrator KA to, for example, provide input to control the operation of the alcohol consumption estimation device IS, and to provide output to monitor its operation.

[0035] The processing unit SY(IS) performs processing related to the alcohol consumption of subjects HI1 to HIm (whether or not they drank alcohol, the amount of alcohol consumed, etc.).

[0036] The memory unit KI(IS) stores, for example, the data necessary for processing by the processing unit SY(IS).

[0037] The communications unit TU (IS) communicates via the network NW and, for example, receives biometric information SJ from wearable terminals WT1 to WTm of subjects HI1 to HIm, particularly heart rate HR and step count ST. It also transmits, for example, a first index f(t) and a second index F(d) (shown in Figures 12 to 14) for estimating the alcohol consumption of subjects HI1 to HIm to the biomedical examination device SK.

[0038] <Configuration of the biomedical testing device SK> Figure 4 shows the configuration of the biomedical testing device SK of Embodiment 1.

[0039] The biological inspection device SK of Embodiment 1, as shown in Figure 4, includes an input / output unit NS(SK), a processing unit SY(SK), a storage unit KI(SK), and a communication unit TU(SK).

[0040] The input / output unit NS (SK) is used by the examiner KE to perform input and output for examining the biological state of subjects HI1 to HIm.

[0041] The processing unit SY (SK) performs, for example, processing related to the biological information SJ (shown in Figure 15) of subjects HI1 to HIm.

[0042] The memory unit KI(SK) stores, for example, data necessary for the processing by the processing unit SY(SK).

[0043] The communication unit TU(SK) performs communication via the network NW. For example, it receives biometric information SJ of the subjects HI1 to HIm, particularly blood pressure KE, respiratory rate KO, and electroencephalogram NO (shown in FIG. 15), from the wearable terminals WT1 to WTm. Also, for example, it receives a first index f(t) and a second index F(d) for estimating the drinking of the subjects HI1 to HIm from the drinking estimation device IS.

[0044] <Operation of Embodiment 1>FIG. 5 is a flowchart showing the operation of the biological examination system SKS of Embodiment 1.

[0045] FIGS. 6 to 14 are time charts showing the operation of the biological examination system SKS of Embodiment 1.

[0046] The operation of the biological examination system SKS of Embodiment 1 will be described with reference to the flowchart of FIG. 5 and the time charts of FIGS. 6 to 14. FIG. 5 shows a method for calculating an index for drinking estimation.

[0047] Hereinafter, for the sake of easy explanation and understanding, it is assumed that the subject to be examined for drinking (presence or absence of drinking, amount of drinking, etc.) is the subject HI1 among the subjects HI1 to HIm.

[0048] Step S1: The wearable terminal WT1 (shown in FIG. 1) acquires biometric information SJ (heart rate HR, number of steps ST, blood pressure KE, respiratory rate KO, electroencephalogram NO, etc.) of the subject HI1 for a predetermined period (for example, 1 minute, 3 minutes, 5 minutes). The wearable terminal WT1 transmits the acquired heart rate HR and number of steps ST to the drinking estimation device IS, and also transmits the acquired blood pressure KE, respiratory rate KO, electroencephalogram NO, etc. to the biological examination device SK.

[0049] The alcohol consumption estimation device IS (shown in Figure 1) receives the heart rate HR and step count ST of subject HI1 from the wearable terminal WT1, and calculates the heart rate HR(t) and step count ST(t) for each unit of time (t) (for example, 1 minute). In other words, the alcohol consumption estimation device IS acquires the heart rate HR(t) and step count ST(t). Alternatively, the alcohol consumption estimation device IS may acquire the heart rate HR(t) and step count ST(t) by receiving them from the wearable terminal WT1.

[0050] As a result, the alcohol consumption estimation device IS calculates, for example, the heart rate HR and step count ST at 20:59 (1 minute), which is heart rate HR (20:59), step count ST (20:59); the heart rate HR and step count ST at 21:00 (1 minute), which is heart rate HR (21:00), step count ST (21:00); ...; the heart rate HR and step count ST at 21:54 (1 minute), which is heart rate HR (21:54), step count ST (21:54); and the heart rate HR and step count ST at 21:55 (1 minute), which is heart rate HR (21:55), step count ST (21:55).

[0051] Step S2: The alcohol consumption estimation device IS calculates the heart rate change ΔHR(t), which is the difference in the heart rate HR(t) of subject HI1 between adjacent and consecutive unit time periods. More specifically, the alcohol consumption estimation device IS calculates the heart rate change ΔHR(t) for subject HI1, which is the difference between the heart rate HR(t-1) in the first unit time period (t-1) and the heart rate HR(t) in the second unit time period (t) following the first unit time period (t-1).

[0052] As a result, the alcohol consumption estimation device IS calculates, for example, the difference between the heart rate HR (20:59), which is the heart rate at 20:59 (corresponding to (t-1)), and the heart rate HR (21:00), which is the heart rate at 21:00 (corresponding to t), as shown in Figure 7. It also calculates, for example, the difference between the heart rate HR (21:54), which is the heart rate at 21:54 (corresponding to (t-1)), and the heart rate HR (21:55), which is the heart rate at 21:55 (corresponding to t), as the heart rate change ΔHR (21:55).

[0053] Step S3: The alcohol consumption estimation device IS obtains the total heart rate change SumΔHR(t). Typically, the alcohol consumption estimation device IS calculates the total heart rate change SumΔHR(t) by summing the heart rate change ΔHR(t) over a third time period Δt3 (e.g., 80 minutes) that is longer than a unit time (t).

[0054] As a result, the alcohol consumption estimation device IS calculates the total heart rate change SumΔHR(21:00) by summing up the heart rate changes ΔHR(19:40), ΔHR(19:41), ..., ΔHR(20:59), and ΔHR(21:00), for the third time period Δt3, for example, from 19:40 (corresponding to (t-Δt3), but not shown) to 21:00 (corresponding to t), as shown in Figure 8.

[0055] Similarly, the alcohol consumption estimation device IS calculates the total heart rate change SumΔHR(21:55) by summing up the heart rate changes ΔHR(20:35), ΔHR(20:36), ..., ΔHR(21:54), and ΔHR(21:55) for a third time period Δt3, for example, from 20:35 (corresponding to the time (t-Δt3)) to 21:55 (corresponding to t).

[0056] In other words, the alcohol consumption estimation device IS calculates the total heart rate change SumΔHR(t) for each first time window, while shifting the first time window, which has a third time interval Δt3, by a unit time (t).

[0057] Step S4: The alcohol consumption estimation device IS defines a function L_ST(t) that indicates whether or not subject HI1 is walking at each unit time (t), based on the number of steps ST of subject HI1 (obtained in step S1).

[0058] As a result, the alcohol consumption estimation device IS defines the function L_ST(20:00) = 1 (no walking) as indicating whether or not walking occurred at 20:00 (1 minute), as shown in Figure 9, since the step count ST(20:00), which indicates the step count ST at 20:00 (1 minute), is 0.

[0059] In contrast, the alcohol consumption estimation device IS, as shown in Figure 6, for example, shows the number of steps ST at 20:08 (1 minute), and since the number of steps ST(20:08) ≈ 20, it is defined as the function L_ST(20:08) = 0 (walking present) which indicates whether walking occurred at 20:08, as shown in Figure 9.

[0060] Step S5: The alcohol consumption estimation device IS selects a representative heart rate repHR(t), which is the heart rate HR that represents a fourth time interval Δt4 (e.g., 40 minutes) longer than the unit time (t), based on the heart rate HR(t) (obtained in Step S1).

[0061] As a result, the alcohol consumption estimation device IS, as shown in Figure 10, selects at least one of the heart rates HR(19:20), HR(19:21), ..., HR(19:59), HR(20:00) as the representative heart rate repHR(20:00) for the fourth time period Δt4, for example, from 19:20 (corresponding to (t - Δt4), but not shown) to 20:00 (corresponding to t).

[0062] Similarly, the alcohol consumption estimation device IS selects, for example, within the fourth time period Δt4 from 21:15 (corresponding to (t - Δt4)) to 21:55 (corresponding to t), at least one of the heart rates HR(21:15), HR(21:16), ..., HR(21:54), HR(21:55) as the representative heart rate repHR(21:55).

[0063] In other words, the alcohol consumption estimation device IS selects a representative heart rate repHR(t) for each second time window, while shifting the second time window, which has a fourth time interval Δt4, by a unit time (t).

[0064] The alcohol consumption estimation device IS may use, for example, the minimum, mode, median, and mean values ​​for heart rates HR(19:20), HR(19:21), ..., HR(19:59), and HR(20:00) as the representative heart rate repHR(20:00) mentioned above. Similarly, the representative heart rate repHR(21:55) may use, for example, the minimum, mode, median, and mean values ​​for heart rates HR(21:15), HR(21:16), ..., HR(21:54), and HR(21:55) mentioned above.

[0065] Step S6: The alcohol consumption estimation device IS defines a function L_DR(t) that indicates whether the representative heart rate repHR(t) (calculated in Step S5) exceeds the reference heart rate criHR (e.g., 80 beats / min) in which subject HI1 is presumed to have been drinking alcohol.

[0066] As a result, the alcohol consumption estimation device IS defines a function L_DR(20:00) = 0 (no possibility of alcohol consumption) that indicates the possibility of alcohol consumption at 20:00, as shown in Figure 11, since, for example, the representative heart rate repHR(20:00) at 20:00 is < 80 beats / min, as shown in Figure 10.

[0067] In contrast, the alcohol consumption estimation device IS, as shown in Figure 10, defines a function L_DR(21:00) = 1 (possible alcohol consumption) indicating the likelihood of alcohol consumption at 21:00, for example, if the representative heart rate at 21:00, repHR(21:00), is ≥ 80 beats / min, as shown in Figure 11.

[0068] Step S7: The alcohol consumption estimation device IS defines a first index f(t) using the summation of heart rate changes SumΔHR(t) (calculated in step S3), the function L_ST(t) (defined in step S4), and the function L_DR(t) (defined in step S6).

[0069] The alcohol consumption estimation device IS defines the first index f(t) by multiplying the summation heart rate change SumΔHR(t), the function L_ST(t), and the function L_DR(t), as shown in the following formula (F1). In other words, the first index f(t) represents the summation heart rate change SumΔHR(t) when the subject is inactive and the subject's heart rate exceeds the reference heart rate criHR.

[0070] The first index f(t) = Total change in heart rate SumΔHR(t) * function L_ST(t) * function L_DR(t) ... (F1)

[0071] The alcohol consumption estimation device IS uses the above formula to define, for example, for 20:00, the first index f(20:00) = total heart rate change SumΔHR(20:00) * function L_ST(20:00) * function L_DR(20:00), and similarly, for example for 21:55, the first index f(21:55) = total heart rate change SumΔHR(21:55) * function L_ST(21:55) * function L_DR(21:55). As a result, the first index f(t) is given, as shown in Figure 12.

[0072] Step S8: The alcohol consumption estimation device IS defines a second index F(d) using the first index f(t) (defined in Step S7).

[0073] More specifically, the alcohol consumption estimation device IS defines the second indicator F(d) for February 2, 2024, as shown in Figure 13, by summing the first indicator f(t) for each day, for example, the first indicator f(t) at 00:00 on February 2, 2024, the first indicator f(t) at 00:01 on February 2, 2024, ..., the first indicator f(t) at 23:58 on February 2, 2024, and the first indicator f(t) at 23:59 on February 2, 2024. "One day" corresponds to the "fifth time Tf" described later.

[0074] The alcohol consumption estimation device IS also defines the second index F(d) for February 2, 2024, as shown in Figure 14, by averaging the first index f(t) for each day, for example, the first index f(t) at 00:00 on February 2, 2024, the first index f(t) at 00:01 on February 2, 2024, ..., the first index f(t) at 23:58 on February 2, 2024, and the first index f(t) at 23:59 on February 2, 2024, instead of the above sum. "One day" corresponds to the "fifth time Tf" described later.

[0075] Specifically, the processes in steps S1 to S8 of Figure 5 (method for calculating the alcohol consumption estimation index) are performed by the processing unit SY(IS).

[0076] <Effects of Embodiment 1> As described above, in the biological examination system SKS of Embodiment 1, the alcohol consumption estimation device IS calculates the total heart rate change amount SumΔHR(t) for subject HI1 based on the heart rate HR and step count ST per unit time (t), and defines the functions L_ST(t) and L_DR(t). Then, using the total heart rate change amount SumΔHR(t), the function L_ST(t), and the function L_DR(t), it defines a first index f(t), and further defines a second index F(d) from the first index f(t). This makes it possible to consider the potential immediate influence that subject HI1's walking (step count ST) may have on subject HI1's heart rate (heart rate HR).

[0077] In the first embodiment of the SKS biomedical examination system, as described above, the steps ST(t) of heart rate HR(t), the total change in heart rate SumΔHR(t) are calculated, the functions L_ST(t) and L_DR(t) are defined, and the first index f(t) and the second index F(d) are defined. Therefore, unlike conventional systems (Patent No. 6547837) in which the data to be used for calculation is stored in a database in advance, the estimation of alcohol consumption in subject HI1 can be performed more easily than in the conventional system described above.

[0078] <Embodiment 2> The SKS biomedical testing system of Embodiment 2 will be described. <Configuration of Embodiment 2> The SKS biomedical testing system of Embodiment 2 has the same configuration as the SKS biomedical testing system of Embodiment 1 (shown in Figures 1 to 4).

[0079] <Operation of Embodiment 2> The operation of the SKS biological testing system in Embodiment 2 is basically the same as the operation of the SKS biological testing system in Embodiment 1.

[0080] <Estimation using the first index f(t)> In the biological examination system SKS of Embodiment 2, on the other hand, unlike the biological examination system SKS of Embodiment 1, the alcohol consumption estimation device IS uses a first index f(t) (shown in Figure 12) relating to subject HI1 to estimate, for example, (1) whether or not subject HI1 has consumed alcohol, and (2) if subject HI1 has been estimated to have consumed alcohol in (1) above, to estimate the time of day when subject HI1 consumed alcohol.

[0081] The alcohol consumption estimation device IS, for example as shown in Figure 12, estimates that (1) subject HI1 may have been drinking alcohol during a time when the first index f(t) is not zero, and (2) estimates that the time when subject HI1 drank alcohol was particularly between 20:35 and 22:00, which is a time when the first index f(t) tends to rise. From Figure 12, the alcohol consumption estimation device IS also predicts that subject HI1 may have been drinking alcohol even after 22:00, as it suggests that the first index f(t) is not zero even after 22:00.

[0082] <Estimation using the second index F(d)> In the biological examination system SKS of Embodiment 2, unlike the biological examination system SKS of Embodiment 1, the alcohol consumption estimation device IS uses a second index F(d) (shown in Figures 13 and 14) relating to subject HI1 to estimate, for example, (1) whether or not subject HI1 has consumed alcohol, and (2) if subject HI1 has consumed alcohol as estimated in (1) above, which day subject HI1 consumed alcohol.

[0083] The alcohol consumption estimation device IS also estimates, for example, as shown in Figures 13 and 14, that (1) subject HI1 may have consumed alcohol on days when the second index F(d) is not 0, and (2) the days on which subject HI consumed alcohol are, for example, February 2nd, February 5th, ... February 24th, February 27th.

[0084] <Correspondence> The processing unit SY (IS) of the alcohol consumption estimation device IS (shown in Figure 3) corresponds to the first estimation unit and the second estimation unit.

[0085] <Effects of Embodiment 2> As described above, in the SKS biomedical examination system of Embodiment 2, by using the first index f(t) and the second index F(d) of Embodiment 1, it is possible to estimate whether or not subject HI1 has been drinking alcohol, and what time of day and day subject HI1 was drinking alcohol.

[0086] <Embodiment 3> The SKS biological testing system of Embodiment 3 will be described. <Configuration of Embodiment 3> The SKS biological testing system of Embodiment 3 has the same configuration as the SKS biological testing system of Embodiment 1 (shown in Figures 1 to 4).

[0087] <Operation of Embodiment 3> The operation of the SKS biological testing system in Embodiment 3 is basically the same as the operation of the SKS biological testing system in Embodiment 1.

[0088] In the third embodiment of the SKS biological testing system, unlike the first embodiment of the SKS biological testing system, the biological testing device SK (shown in Figures 1 and 4) examines the biological state of subjects HI1 to HIm.

[0089] Figure 15 is a time chart showing the operation of the SKS biomedical examination system of Embodiment 3.

[0090] The biomedical device SK extracts or excludes alcohol-related biomedical information IKSJ related to subject HI1's alcohol consumption from subject HI1's biomedical information SJ (for example, heart rate HR, step count ST (shown in Figure 6), blood pressure KE, respiratory rate KO, electroencephalogram NO (shown in Figure 15)) using at least one of the first index f(t) and the second index F(d) defined in Embodiment 1, and examines subject HI1's biological state using the alcohol-related biomedical information IKSJ or the biomedical information SJ other than the alcohol-related biomedical information IKSJ.

[0091] As shown in Figure 15, the alcohol-related biometric information IKSJ is, for example, the blood pressure KE(2) during the second period KI(2), which is the estimated period during which subject HI1 was drinking alcohol, based on the first index f(t) and the second index F(d) of the blood pressure KE, which is one of the biometric information SJs of subject HI1. Similarly, as shown in Figure 15, the alcohol-related biometric information IKSJ is the respiratory rate KO(2) during the second period KI(2), which is one of the biometric information SJs of subject HI1, and the electroencephalogram NO(2) during the second period KI(2), which is one of the electroencephalogram SJs of subject HI1.

[0092] <When using alcohol-related biometric information IKSJ> When the biomedical device SK attempts to examine the biological state of subject HI1 based on the biometric information SJ affected by alcohol consumption by subject HI1, it uses the alcohol-related biometric information IKSJ, namely blood pressure KE(2), respiratory rate KO(2), and electroencephalogram NO(2), as shown in Figure 15.

[0093] <When using biometric information other than IKSJ related to alcohol consumption> When the biomedical device SK attempts to examine the biological state of subject HI1 based on biometric information SJ that is not affected by alcohol consumption by subject HI1, it uses blood pressure KE(1), KE(3), respiratory rate KO(1), KO(3), electroencephalogram NO(1), NO(3), which are other than the biometric information IKSJ related to alcohol consumption, as shown in Figure 15.

[0094] <Correspondence> The processing unit SY (SK) of the biomedical testing device SK (shown in Figure 4) corresponds to the extraction / removal unit and the testing unit.

[0095] <Effects of Embodiment 3> As described above, in the biological testing system SKS of Embodiment 3, the biological testing device SK can perform a biological examination of subject HI1 using alcohol-related biological information IKSJ, or using biological information SJ from which alcohol-related biological information IKSJ has been removed, depending on whether or not the effects of alcohol consumption by subject HI1 are to be considered.

[0096] <Hardware configuration of Embodiment 1> Figure 16 shows the hardware configuration of the SKS biological examination system according to Embodiments 1 to 3.

[0097] The biological inspection system SKS of Embodiments 1 to 3 includes a processing circuit SYO as shown in Figure 16 in order to perform the functions described above, and further includes an input circuit NYU and an output circuit SYU as necessary.

[0098] The processing circuit SYO is dedicated hardware. The processing circuit SYO primarily implements the functions of the processing units SY(WT), SY(IS), and SY(SK) of the wearable terminal WT, the alcohol consumption estimation device IS, and the biomedical examination device SK (shown in Figures 2 to 4).

[0099] The processing circuit SYO may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination thereof.

[0100] The input circuit NYU and output circuit SYU exchange inputs and outputs related to the operation of the processing circuit SYO with, for example, the wearable terminal WT, the alcohol consumption estimation device IS, and the biomedical examination device SK.

[0101] <Hardware configuration based on software implementation of Embodiment 1> Figure 17 shows the hardware configuration based on software implementation of the SKS biological examination system of Embodiments 1 to 3.

[0102] The biological inspection system SKS of Embodiments 1 to 3 includes a processor PRO and a memory circuit KIO, as shown in Figure 17, and optionally further includes an input circuit NYU and an output circuit SYU.

[0103] The processor PRO is a CPU (Central Processing Unit, also known as a microprocessor, microcomputer, or DSP (Digital Signal Processing)) that executes programs. The processor PRO primarily implements the functions of the processing units SY(WT), SY(IS), and SY(SK) of the wearable terminal WT, the alcohol consumption estimation device IS, and the biomedical examination device SK.

[0104] The PRO processor implements the above-described functions through software, firmware, or a combination of software and firmware. The software and firmware are written as programs and stored in the memory circuit KIO.

[0105] The processor PRO realizes the above-described functions by reading and executing the above-described program from the memory circuit KIO. The above-described program can also be said to cause the computer to execute the procedures and methods of the processing units SY(WT), SY(IS), and SY(SK) of the wearable terminal WT, the alcohol estimation device IS, and the biomedical examination device SK.

[0106] Here, the memory circuit KIO includes, for example, non-volatile or volatile semiconductor memories such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), and EEPROM (Electrically Erasable Programmable Read-Only Memory), as well as magnetic disks, flexible disks, optical disks, compact disks, minidiscs, DVDs (Digital Versatile Discs), etc.

[0107] Of the functions of the processing units SY(WT), SY(IS), and SY(SK) of the wearable terminal WT, alcohol estimation device IS, and biomedical examination device SK, some functions may be implemented by the processing circuit SYO (shown in Figure 16), while other functions may be implemented by the processor PRO (shown in Figure 17).

[0108] As described above, the functions of the processing units SY(WT), SY(IS), and SY(SK) of the wearable terminal WT, alcohol estimation device IS, and biomedical examination device SK can be realized through hardware, software, firmware, or a combination thereof.

[0109] The input circuit NYU and output circuit SYU exchange inputs and outputs related to the operation of the processor PRO with external devices such as the wearable terminal WT, the alcohol consumption estimation device IS, and the biomedical examination device SK.

[0110] <Summary of Embodiments 1-3> As described above with reference to Figures 1-17, the processing unit SY(IS) calculates the total heart rate change SumΔHR(t), which is the sum of the heart rate change amounts ΔHR(t) of the subject. The processing unit SY(IS) then calculates a first index f(t) for estimating alcohol consumption, which represents the total heart rate change SumΔHR(t) when the subject is inactive (e.g., not walking) and the subject's heart rate exceeds the reference heart rate criHR. Therefore, the first index f(t) for estimating alcohol consumption is information that quantifies the sustained upward trend of the heart rate in the subject's inactive state. On the other hand, a sustained upward trend of the heart rate can be observed when drinking alcohol. Thus, the first index f(t) for estimating alcohol consumption is information that can estimate alcohol consumption. In particular, the first index f(t) for estimating alcohol consumption can reduce the influence of increased heart rate due to the subject's physical activity (e.g., walking) by quantifying the sustained upward trend in the subject's heart rate during inactivity. Therefore, the first index f(t) for estimating alcohol consumption is information that can estimate alcohol consumption with high accuracy.

[0111] In detail, the processing unit SY(IS) calculates the total heart rate change SumΔHR(t) by summing the heart rate change ΔHR(t), which is the difference between the subject's heart rate in a first unit time (t-1) and the subject's heart rate in a second unit time (t) that follows the first unit time (t-1), over a third time period Δt3 that is longer than the unit time (t). Specifically, the processing unit SY(IS) calculates the total heart rate change SumΔHR(t) by summing the heart rate change ΔHR(t), which is the difference between the subject's heart rate in a first unit time (t-1) and the subject's heart rate in a second unit time (t) following the first unit time (t-1), over a third time period Δt3 that is longer than the unit time (t). The "subject's inactive state" is indicated by the function L_ST(t), which is defined based on the subject's physical activity information measured for each unit time (t), and indicates whether the subject is active or not. The "state in which the subject's heart rate exceeds the reference heart rate criHR" is indicated by the function L_DR(t), which indicates whether the representative heart rate repHR(t) exceeds the reference heart rate criHR, which is presumed to indicate the possibility of alcohol consumption. The representative heart rate repHR(t) is a representative heart rate calculated (selected) within a fourth time interval Δt4 that is longer than a unit time (t), based on the subject's heart rate. Function L_ST(t) corresponds to an example of the “first function” of this disclosure. Function L_DR(t) corresponds to an example of the “second function” of this disclosure.

[0112] With this configuration, the function L_ST(t) binarizes the subject's active and inactive states, thus simplifying the calculation process of the first index f(t) for alcohol consumption estimation using the function L_ST(t). Furthermore, the function L_DR(t) binarizes the subject's heart rate into states where it exceeds the reference heart rate criHR and states where it does not, thus simplifying the calculation process of the first index f(t) for alcohol consumption estimation using the function L_DR(t). In addition, the influence of short-term heart rate fluctuations can be suppressed by using the representative heart rate repHR(t). It is preferable that the first unit time (t-1) and the second unit time (t) are consecutively adjacent, but they do not have to be consecutively adjacent.

[0113] As a preferred example, the subject's physical activity information for defining the function L_ST(t) is the subject's step count. The function L_ST(t) then indicates whether the subject is walking or not. For example, a step count of 0 indicates an inactive state (no walking). In this preferred example, the function L_ST(t) can be easily defined by the step count. However, the physical activity information may also be exercise intensity. Exercise intensity is indicated, for example, by METs (Metabolic equivalents). In this case, the function L_ST(t) indicates whether the subject is active or not. For example, a state of 1 MET indicates an inactive state.

[0114] As a preferred example, the processing unit SY(IS) calculates a second indicator F(d) for estimating alcohol consumption, which is the result of statistical processing, by performing statistical processing on a first indicator f(t) for estimating alcohol consumption within a fifth time period Tf. According to this preferred example, the first indicator f(t) for estimating alcohol consumption within a fifth time period Tf can be easily summarized without losing its characteristics. In the above example, the fifth time period Tf is one day, but it is not limited to this. The fifth time period Tf may be, for example, one week, one month, or one year. The fifth time period Tf corresponds to an example of a “predetermined time” in this disclosure.

[0115] In this case, for example, statistical processing refers to the process of calculating a sum or the process of calculating an average. In this case, the results of the statistical processing of the first indicator f(t) for estimating alcohol consumption within the fifth time period Tf are easily grasped intuitively. Note that statistical processing is not limited to these, and may also include, for example, variance or standard deviation.

[0116] As a preferred example, the first index f(t) for estimating alcohol consumption is represented by the summation change in heart rate, SumΔHR(t), which indicates the increase in the subject's heart rate (see formula (F3) below). Therefore, the first index f(t) for estimating alcohol consumption can more accurately quantify the sustained increase in the subject's heart rate while they are inactive.

[0117] In this case, for example, a first index f(t) for estimating alcohol consumption can be defined by formulas (F2) and (F3). In formula (F3), "SumΔHR(t) > 0" indicates an increase in the sum change in heart rate, SumΔHR(t).

[0118] First index f(t) = 0 (SumΔHR(t) ≤ 0) ... (F2) First index f(t) = Total change in heart rate SumΔHR(t) * function L_ST(t) * function L_DR(t) (SumΔHR(t) > 0) ... (F3)

[0119] Furthermore, the processing unit SY(IS) executes an alcohol consumption estimation process (alcohol consumption estimation method). Figure 18 is a flowchart of the alcohol consumption estimation process. As shown in Figure 18, the alcohol consumption estimation process includes steps S301 and S302. First, in step S301, the processing unit SY(IS), which functions as a first estimation unit, estimates whether or not the subject has consumed alcohol using a first index f(t) for alcohol consumption estimation. Next, in step S302, the processing unit SY(IS), which functions as a second estimation unit, estimates when the subject consumed alcohol using the first index f(t) for alcohol consumption estimation (estimation of the timing of alcohol consumption). For example, the processing unit SY(IS) uses the first index f(t) for alcohol consumption estimation to estimate what time of day the subject consumed alcohol.

[0120] Furthermore, the processing unit SY(IS) may perform another alcohol estimation process (alcohol estimation method). Specifically, as shown in Figure 18, first, in step S301, the processing unit SY(IS), which functions as the first estimation unit, estimates whether or not the subject drank alcohol using a second alcohol estimation index F(d). Next, in step S302, the processing unit SY(IS), which functions as the second estimation unit, estimates when the subject drank alcohol using the second alcohol estimation index F(d). In this case, the second alcohol estimation index F(d) is an index obtained by performing statistical processing on the first alcohol estimation index f(t) within a fifth time period Tf. For example, the processing unit SY(IS) uses the second alcohol estimation index F(d) to estimate which day the subject drank alcohol. In this case, the second alcohol estimation index F(d) is an index obtained by performing statistical processing on the first alcohol estimation index f(t) over a one-day period.

[0121] As explained above with reference to Figure 18, by using the first indicator f(t) or the second indicator F(d) for estimating alcohol consumption, it is possible to estimate the subject's alcohol consumption while taking into account the potential immediate effects of the subject's physical activity on the subject's heart rate.

[0122] Furthermore, the processing unit SY(SK) of the biomedical testing device SK executes a biomedical testing process (biological testing method). Figure 19 is a flowchart of the biomedical testing process. As shown in Figure 19, the biomedical testing process includes steps S401 and S402. First, in step S401, the processing unit SY(SK), which functions as an extraction / exclusion unit, extracts or excludes alcohol-related biological information IKSJ related to the subject's alcohol consumption from the subject's biological information using at least one of the first index f(t) for alcohol consumption estimation and the second index F(d) for alcohol consumption estimation. Next, in step S402, the processing unit SY(SK), which functions as an inspection unit, inspects the subject's biological state using the extracted alcohol-related biological information IKSJ, or the biological information other than the alcohol-related biological information IKSJ.

[0123] As explained above with reference to Figure 19, it is possible to perform tests that focus on IKSJ, which is alcohol-related biometric information, or to perform tests that focus on biometric information that is not related to alcohol consumption.

[0124] Here, the first indicator f(t) for estimating alcohol consumption and the second indicator F(d) for estimating alcohol consumption correspond to an example of the indicator information for estimating alcohol consumption in Embodiment 4, which will be described later.

[0125] Next, an information processing system according to Embodiment 4 of the present disclosure will be described. <Embodiment 4> The information processing system according to Embodiment 4 of the present disclosure uses a learning model to estimate the subject's bio-related state information from at least the subject's alcohol consumption estimation index information. The subject's bio-related state information includes information representing the subject's mental and physical state, information representing the environmental state that affects the subject's mental and physical state, information representing the biological state affected by the subject's mental and physical state, or information regarding the subject's alcohol consumption. The learning model is constructed by performing learning using a learning dataset. The learning dataset includes at least the learning subject's alcohol consumption estimation index information and the learning subject's bio-related state information. The learning subject's bio-related state information includes information representing the learning subject's mental and physical state, information representing the environmental state that affects the subject's mental and physical state, information representing the biological state affected by the learning subject's mental and physical state, or information regarding the learning subject's alcohol consumption. In this specification, mental and physical refers to the mind and / or body.

[0126] A subject corresponds to an example of a “subject” in this disclosure. A learning subject corresponds to an example of a “learning subject” in this disclosure.

[0127] The alcohol consumption estimation index information for the learning subject and the alcohol consumption estimation index information for the subject in Embodiment 4 include the first index f(t) or the second index F(d) described in Embodiments 1 to 3. In Embodiments 1 to 3, the SKS biomedical examination system calculated alcohol consumption estimation index information considering the potential immediate effects of the subject's physical activity (e.g., walking) on ​​the subject's heart rate. In Embodiment 4, the information processing system obtains highly reliable bio-related state information by utilizing the subject's alcohol consumption estimation index information obtained considering the potential immediate effects of the subject's physical activity (e.g., walking) on ​​the subject's heart rate.

[0128] [Information Processing System 1] An information processing system 1 according to Embodiment 4 of the present disclosure will be described with reference to Figures 20 to 33. Figure 20 is a block diagram showing an example configuration of the information processing system 1. As shown in Figure 20, the information processing system 1 comprises a learning data creation device 2, a learning device 3, and an estimation device 4. Each of the learning data creation device 2, the learning device 3, and the estimation device 4 is a computer. The estimation device 4 comprises a pre-processing unit 41, an estimation unit 42, and a learning model TM1. The estimation device 4 may further comprise a post-processing unit 43.

[0129] The learning data creation device 2 acquires the learning subject's biological state raw data A1 and the correct answer information Z1. The biological state raw data A1 is raw data indicating the learning subject's biological state. A learning subject is a subject from whom learning raw data is acquired. Typically, a learning subject is a human being. The biological state raw data A1 includes the learning subject's vital raw data and / or behavioral raw data. The vital raw data is raw data indicating the vital information of the organism. The behavioral raw data is raw data indicating the behavioral information of the organism. The correct answer information Z1 indicates biological-related state information.

[0130] The training data creation device 2 creates a training dataset F1 based on the raw biological state data A1 and the ground truth information Z1. The training dataset F1 includes feature information G1 and ground truth labels B1 of the training subject. Feature information G1 includes at least the training subject's alcohol consumption estimation index information D1. Feature information G1 is an explanatory variable. Feature information G1 is created based on the raw biological state data A1. Ground truth labels B1 is the target variable. Ground truth labels B1 are created based on the ground truth information Z1. Ground truth labels B1 include biological related state information M1.

[0131] The learning device 3 performs learning using the learning dataset F1 by executing a machine learning algorithm. The learning device 3 generates a learning model TM1 by repeatedly performing learning using multiple learning datasets F1. The learning device 3 performs supervised learning. The learning model TM1 outputs output information L2 when input information K2 is input. Typically, the learning model TM1 is a pre-trained model. Also, the learning model TM1 is a computer program.

[0132] Meanwhile, the estimation device 4 uses the learning model TM1 to estimate the subject's bio-related state information M2. The subject is typically a human.

[0133] Specifically, the preprocessing unit 41 performs preprocessing on the subject's biological state raw data A2 and generates input information K2, which is the result of the preprocessing. The biological state raw data A2 is raw data indicating the subject's biological state. The biological state raw data A2 includes the subject's vital raw data and / or behavioral raw data. The vital raw data is raw data indicating the vital information of the organism. The behavioral raw data is raw data indicating the behavioral information of the organism.

[0134] The preprocessing unit 41 includes at least an alcohol consumption index calculation unit 411. The alcohol consumption index calculation unit 411 calculates an estimated alcohol consumption index information D2 for the subject based on the subject's raw vital data and behavioral data. Therefore, the input information K2 includes at least the subject's estimated alcohol consumption index information D2.

[0135] Specifically, the alcohol consumption index calculation unit 411 calculates the alcohol consumption estimation index information D2 in the same manner as the alcohol consumption estimation device IS in any of Embodiments 1 to 3. In other words, the alcohol consumption index calculation unit 411 has the same functions as the alcohol consumption estimation device IS (processing unit SY (IS)). Note that the preprocessing unit 41 may obtain the subject's alcohol consumption estimation index information D2 from the alcohol consumption estimation device IS without providing the alcohol consumption index calculation unit 411. In this case, for example, the information processing system 1 may include the alcohol consumption estimation device IS.

[0136] The estimation unit 42 inputs the input information K2 to the learning model TM1. As a result, the learning model TM1 outputs output information L2. The input information K2 is an explanatory variable. The output information L2 is an objective variable. The output information L2 includes the subject's biological state information M2. In other words, the learning model TM1 takes the input information K2 as input and estimates the subject's biological state information M2. The estimation unit 42 obtains the output information L2 from the learning model TM1. The post-processing unit 43 may perform post-processing on the output information L2 and generate output information N2, which is the result of the post-processing. The output information N2 includes the biological state information M3 after post-processing.

[0137] In the following, when it is not necessary to distinguish between the biological status information M2 and M3, the biological status information M2 and M3 of the subject may be referred to as "biological status information MX".

[0138] In Embodiment 4, the estimation device 4 can output (estimate) bio-related state information MX with high estimation accuracy by inputting at least the subject's alcohol consumption estimation index information D2 to the learning model TM1. In this case, for example, the estimation device 4 does not require diagnosis and evaluation by medical professionals, examinations using medical devices such as medical imaging diagnostic equipment, or collection and examination of bodily fluids such as blood. Medical professionals are, for example, doctors or nurses. In addition, the estimation device 4 automatically and continuously acquires raw bio-state data A2 from a wearable device (e.g., a wearable terminal WT) and / or a mobile terminal. Therefore, bio-related state information MX can be continuously obtained while reducing the burden on the subject.

[0139] [Training Dataset F1] Next, we will explain the correlation between the alcohol consumption estimation index information D1 and the biological state information M1 of the training subjects in the training dataset F1. Figure 21 is a diagram showing an example of the training dataset F1. As shown in Figure 21, the training dataset F1 includes feature information G1 and ground truth labels B1 of the training subjects. Feature information G1 includes at least the alcohol consumption estimation index information D1. On the other hand, ground truth labels B1 include biological state information M1. In ground truth labels B1, the biological state information M1 is, for example, measured values ​​(e.g., diagnosis results by a doctor, test results by a medical device, biological test results using biomarkers, or evaluation results using various indicators).

[0140] Specifically, the biological state information M1 includes information M100 representing the mental and physical state of the learning subject (hereinafter referred to as "mental and physical state information M100"), information M110 representing the environmental state that affects the mental and physical state of the learning subject (hereinafter referred to as "mental and physical environment information M110"), information representing the biological state affected by the mental and physical state of the learning subject (hereinafter referred to as "passive biological information M120"), or information regarding the learning subject's alcohol consumption (hereinafter referred to as "alcohol consumption information M130"). On the other hand, the alcohol consumption estimation index information D1 includes information regarding the summation of heart rate change SumΔHR(t) in a state where the learning subject is inactive and the learning subject's heart rate exceeds the reference heart rate. The summation of heart rate change SumΔHR(t) represents the sum of the heart rate change ΔHR(t), which is the change in the learning subject's heart rate. In other words, the alcohol consumption estimation index information D1 is information that quantifies the sustained upward trend in the heart rate of a learning subject in an inactive state. On the other hand, a sustained upward trend in heart rate can be observed when drinking alcohol. Therefore, the alcohol consumption estimation index information D1 is information that can estimate alcohol consumption. In particular, by quantifying the sustained upward trend in the heart rate of a learning subject in an inactive state, the alcohol consumption estimation index information D1 can reduce the influence of the increase in heart rate due to the learning subject's physical activity (e.g., walking). Therefore, the alcohol consumption estimation index information D1 is information that can estimate alcohol consumption with high accuracy.

[0141] Furthermore, alcohol consumption can affect the state of various physical and mental functions. Therefore, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and the physical and mental state information M100. In addition, the presence, amount, and frequency of alcohol consumption can be influenced by the state of the surrounding environment of the organism. For example, the presence, amount, and frequency of alcohol consumption may be affected by stress from air pollution or a low-oxygen environment. Therefore, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and the physical and mental environment information M110. Moreover, direct drinkers, such as organisms that have consumed alcohol or organisms with alcohol dependence, can affect other organisms. For example, in the case of alcohol dependence, it may cause mental burden on family members, etc., or drinking during pregnancy may affect the fetus. Therefore, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and passive organism information M120. Furthermore, as can be understood from Embodiments 1 to 3, there is a correlation between the alcohol consumption estimation index information D1 and the alcohol consumption information M130.

[0142] Therefore, according to Embodiment 4, the learning model TM1 constructed by learning using the learning dataset F1 (drinking estimation index information D1 and biologically related state information M1) can output (generate) the subject's biologically related state information M2 (Figure 22) when the subject's drinking estimation index information D2 (Figure 22) is input. In other words, the estimation device 4 can estimate the subject's biologically related state information M2 by using the learning model TM1. In particular, the learning model TM1 learns the correlation between the drinking estimation index information D1 and the biologically related state information M1, which are information that can estimate drinking with high accuracy. Therefore, by inputting the drinking estimation index information D2 into the learning model TM1, the subject's biologically related state information M2 can be estimated with high accuracy.

[0143] (Physical state information in mental-physical state information M100) For example, the mental-physical state information M100 of the biological-related state information M1 may include information representing the physical state (hereinafter referred to as "physical state information"). Since the alcohol consumption estimation index information D1 is information that can estimate alcohol consumption with high accuracy, the learning model TM1 can accurately learn the relationship between alcohol consumption and the physical state (for example, worsening or improvement of physical illness and symptoms). In other words, it can learn changes in the physical state caused by alcohol consumption with high accuracy.

[0144] Physical condition information may include, for example, information on diseases or symptoms of the nervous system and the cranial nervous system (hereinafter referred to as "nervous system information"), information on diseases or symptoms of the circulatory system (hereinafter referred to as "circulatory system information"), information on diseases or symptoms of the respiratory system (hereinafter referred to as "respiratory system information"), information on diseases or symptoms of the digestive system and the hepatobiliary and pancreatic regions (hereinafter referred to as "digestive system information"), information on diseases or symptoms of the endocrine and metabolic systems (hereinafter referred to as "endocrine and metabolic system information"), information on diseases or symptoms of the renal and urinary tract (hereinafter referred to as "renal and urinary tract information"), information on diseases or symptoms of sex hormones and the reproductive system (hereinafter referred to as "reproductive system information"), information on diseases or symptoms of the skin and sensory organs (hereinafter referred to as "skin and sensory organ information"), information on diseases or symptoms related to immunity and infectious diseases (hereinafter referred to as "immune and infectious disease-related information"), information on tumor symptoms (hereinafter referred to as "tumor-related information"), information on pre-disease conditions of the body (hereinafter referred to as "pre-disease information"), or information representing the physical condition that can be indicated by the results of a biopsy (hereinafter referred to as "biopsy information").

[0145] The neurological information may include, for example, information describing conditions related to alcoholic dementia, Wernicke encephalopathy, Korsakoff syndrome, peripheral neuropathy, tremor or delirium, or Parkinson's disease.

[0146] For example, chronic heavy drinking can lead to hippocampal atrophy and memory impairment, potentially resulting in alcohol-induced dementia. For example, Wernicke's encephalopathy is an acute brain injury caused by vitamin B1 deficiency resulting from alcohol consumption and malnutrition. For example, Korsakoff syndrome is a syndrome caused by alcohol consumption. For example, peripheral neuropathy is a disorder in which alcohol consumption leads to neurodegeneration, causing numbness or pain in the limbs. For example, tremors or delirium can occur in severe cases of alcohol dependence or during alcohol withdrawal. For example, it is sometimes said that people with a history of moderate alcohol consumption have a lower incidence of Parkinson's disease. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and neurological information.

[0147] Cardiovascular information may include information describing conditions such as hypertension, atrial fibrillation or arrhythmia, cardiomyopathy (alcoholic cardiomyopathy), ischemic heart disease (angina pectoris, myocardial infarction), or cerebral hemorrhage or cerebral infarction. For example, long-term alcohol consumption may increase both systolic and diastolic blood pressure. For example, attacks of atrial fibrillation or arrhythmia may be acutely triggered by alcohol consumption. For example, with regard to cardiomyopathy (alcoholic cardiomyopathy), years of heavy drinking may lead to myocardial dilation and fibrosis, resulting in heart failure. For example, with regard to ischemic heart disease (angina pectoris, myocardial infarction), a link between alcohol consumption and the promotion of arteriosclerosis and alcohol consumption and vascular endothelial damage has been observed. In addition, for example, moderate consumption of certain types of alcohol (e.g., red wine) is thought to increase HDF cholesterol and suppress platelet aggregation, potentially leading to a reduction in the risk of ischemic heart disease. For example, with regard to cerebral hemorrhage or cerebral infarction, it is conceivable that alcohol consumption may cause blood pressure fluctuations or coagulation abnormalities. These results suggest a correlation between the alcohol consumption estimation index D1 and cardiovascular system information.

[0148] Respiratory system information may include information representing conditions such as sleep apnea syndrome, aspiration pneumonia, or bronchial asthma. For example, regarding sleep apnea syndrome, alcohol consumption is thought to cause upper airway muscle relaxation, increasing the risk of temporary respiratory interruption. For example, a decrease in the swallowing reflex due to alcohol consumption is thought to increase the risk of aspiration, thereby increasing the risk of aspiration pneumonia. For example, some alcoholic beverages can trigger bronchial asthma attacks. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and the respiratory system information.

[0149] Digestive system information may include, for example, information describing conditions related to alcoholic fatty liver, alcoholic hepatitis or cirrhosis, gastritis or gastric ulcers, esophageal varices, or chronic pancreatitis.

[0150] For example, alcoholic fatty liver disease is said to occur frequently in the early stages of drinking. For example, alcoholic hepatitis or cirrhosis are diseases that can develop due to excessive alcohol consumption or drinking habits. For example, gastritis or gastric ulcers may be worsened by the combination of certain drugs (such as NSAIDs) in addition to mucosal damage caused by alcohol. For example, esophageal varices are a symptom that can arise from cirrhosis caused by alcohol consumption. For example, alcohol consumption is considered one of the biggest risk factors for pancreatitis. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and digestive system information.

[0151] Endocrine and metabolic information may include, for example, information describing conditions related to hypoglycemia, type 2 diabetes, or gout (hyperuricemia).

[0152] For example, hypoglycemia can occur due to the suppression of hepatic gluconeogenesis caused by alcohol consumption (especially when consumed on an empty stomach). For example, type 2 diabetes is at risk due to increased insulin resistance from alcohol consumption and the promotion of obesity by drinking habits. For example, gout (hyperuricemia) is thought to develop due to impaired uric acid excretion caused by the accumulation of purines and lactic acid from certain alcoholic beverages. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and endocrine metabolic system information.

[0153] Renal and urinary tract information may include, for example, information describing conditions related to dehydration or hyponatremia due to diuresis, urinary tract stones, or renal dysfunction.

[0154] For example, diuretic dehydration or hyponatremia can occur because alcohol consumption suppresses ADH (antidiuretic hormone), leading to polyuria and dehydration. For example, urinary tract stones can occur because alcohol consumption induces dehydration, and continued dehydration increases the risk of stone formation. For example, renal dysfunction may occur due to the development of hypertension and diabetes as a result of long-term alcohol consumption. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and renal and urinary tract information.

[0155] Reproductive system information may include, for example, information describing conditions related to irregular or amenorrhea, ovulation disorders or infertility, polycystic ovary syndrome (PCOS), osteoporosis, erectile dysfunction (ED), or male infertility.

[0156] For example, menstrual irregularities or amenorrhea may result from alcohol consumption disrupting the feedback loop between the hypothalamus, pituitary gland, and ovarian axis. For example, ovulation disorders or infertility may occur due to hormonal abnormalities caused by alcohol consumption. Hormonal abnormalities are indicated, for example, by changes in the ratio of FSH (follicle-stimulating hormone) to LH (luteinizing hormone). For example, the worsening of polycystic ovary syndrome may be caused by increased insulin resistance due to alcohol consumption. For example, if malnutrition occurs in addition to decreased estrogen secretion due to alcohol consumption, the risk of osteoporosis may increase as a result. For example, erectile dysfunction may result from alcohol consumption inhibiting vascular responses and neurotransmission. For example, a decrease in spermatogenesis has been confirmed with chronic alcohol consumption in relation to male infertility. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and reproductive system information.

[0157] Information from the skin sensory system may include, for example, information describing conditions such as facial flushing, atopic dermatitis, nystagmus or diplopia, eye strain or photophobia, or auditory hypersensitivity or tinnitus.

[0158] For example, regarding facial flushing, the flushing reaction during alcohol consumption is particularly pronounced in individuals with an inactive form of ALDH2 (aldehyde dehydrogenase 2). For example, in the case of atopic dermatitis, there have been cases where the histamine response is enhanced by alcohol consumption. For example, symptoms such as nystagmus or diplopia may be caused by cerebellar damage due to chronic heavy drinking or by the central nervous system depressant effect of alcohol. For example, eye strain or photophobia is seen during hangovers or chronic drinking. For example, abnormal sensory processing has been reported in long-term drinkers regarding auditory hypersensitivity or tinnitus. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and skin sensory organ system information.

[0159] Immunoinfectious disease-related information may include, for example, information describing susceptibility to infection (e.g., common cold, pneumonia, skin infections), tuberculosis, or vaccine efficacy.

[0160] For example, regarding increased susceptibility to infection, it is thought that alcohol consumption reduces neutrophil function and mucosal defense, increasing the risk of infection with various infectious diseases. For instance, tuberculosis has a high prevalence among individuals with chronic alcohol dependence. For example, alcohol consumption can suppress the immune response, potentially leading to reduced vaccine effectiveness. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index D1 and information related to immune infections.

[0161] Tumor-related information may include, for example, information describing a condition related to oral or pharyngeal cancer, esophageal cancer (squamous cell carcinoma), hepatocellular carcinoma (HCC), colorectal cancer, breast cancer (especially in premenopausal women), or pancreatic cancer.

[0162] For example, oral and pharyngeal cancers have been strongly associated with alcohol consumption. For example, esophageal cancer (squamous cell carcinoma) is reported to be particularly high-risk in people with aldehyde metabolism disorders. For example, hepatocellular carcinoma can develop against a background of alcoholic cirrhosis. For example, colorectal cancer has been reported to have an increased risk in long-term heavy drinkers. For example, breast cancer (especially in premenopausal women) can be caused by impaired estrogen metabolism affected by alcohol consumption. For example, pancreatic cancer may have an increased risk through its association with alcoholic pancreatitis. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and tumor-related information.

[0163] Pre-disease information may include, for example, information describing chronic fatigue or chronic fatigue syndrome, cold sensitivity or poor peripheral circulation, edema (swelling), headache or hangover, indigestion, constipation, diarrhea, or conditions related to libido. In this case, pre-disease information is information related to the physical body.

[0164] For example, chronic fatigue or chronic fatigue syndrome may occur due to alcohol consumption causing decreased sleep quality, blood glucose fluctuations, or liver dysfunction. For example, cold extremities or poor peripheral circulation may occur due to temporary vasodilation and subsequent vasoconstriction during alcohol consumption. For example, edema (swelling) may occur due to the vasoactive effects of alcohol and hormonal imbalances. For example, headaches or hangovers may occur due to dehydration, vasoconstriction and dilation, and the effects of acetaldehyde caused by alcohol consumption. For example, indigestion, constipation, and diarrhea may occur due to disruption of the intestinal environment caused by alcohol consumption. Disruption of the intestinal environment can also occur with chronic light drinking. For example, decreased libido may occur due to disruptions in sex hormones, blood flow, and neurotransmission caused by alcohol consumption. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and pre-disease information.

[0165] The biomedical information may include, for example, information representing the physical condition that can be indicated by biomarker test results, medical device test results, or evaluation results using an assessment scale (e.g., a subjective scale). Medical devices include, for example, blood pressure monitors or medical imaging devices (e.g., X-ray machines, CT scanners, MRI scanners, ultrasound diagnostic devices).

[0166] Biomarker test results include, for example, salivary cortisol, salivary amylase, or urinary 6-melatonin sulfate (aMT6s) levels. Salivary cortisol tends to rise in the short term and decrease in the long term with alcohol consumption. Salivary amylase functions as an indicator of autonomic nervous system stress and tends to rise with alcohol withdrawal or long-term stress. Urinary 6-melatonin sulfate functions as an indicator of circadian rhythm (sleep rhythm), and chronic alcohol consumption can cause disruption of this rhythm.

[0167] Biomarker test results include, for example, test results related to oxidative stress, immune function, or cytokines. An indicator of oxidative stress may be, for example, 8-OHdG or 8-isoprostane. An indicator related to immunity may be, for example, HHV-6 or HHV-7. An indicator related to cytokines may be, for example, TGF-β. These are biomarkers indicating stress. Furthermore, moderate alcohol consumption may reduce mental stress through effects such as mood enhancement. On the other hand, excessive alcohol consumption may cause mental stress.

[0168] From the above, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and the information representing the physical state that can be indicated by the biomarker test results.

[0169] Medical device test results include, for example, blood pressure readings from a blood pressure monitor. For instance, acute heavy drinking can cause a transient increase in blood pressure. Also, moderate drinking can sometimes improve stress-induced hypertension. Furthermore, physical conditions indicated by medical imaging diagnostic device test results may be attributable to alcohol consumption. From these points, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and the information representing physical conditions that can be indicated by medical device test results.

[0170] The evaluation results from the assessment scale are, for example, the results from a fatigue scale or a lethargy scale. Light drinking may reduce fatigue and lethargy. On the other hand, excessive drinking may cause fatigue and lethargy. From these, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and the information representing the physical state that can be indicated by the evaluation results from the assessment scale.

[0171] Furthermore, information regarding the physical condition may include, for example, information regarding accompanying symptoms of a physical disease or symptom, or information regarding side effects or complications of treatment for a physical disease or symptom.

[0172] (Mental state information in mental state information M100) For example, the mental state information M100 of the biological state information M1 may include information representing the mental state (hereinafter referred to as "mental state information"). Since the alcohol consumption estimation index information D1 is information that can estimate alcohol consumption with high accuracy, the learning model TM1 can accurately learn the relationship between alcohol consumption and the mental state (for example, worsening or improvement of mental illness and symptoms). In other words, it can learn changes in the mental state caused by alcohol consumption with high accuracy.

[0173] Mental state information may include, for example, information describing a mental illness (hereinafter referred to as "mental illness information"), information describing a mental symptom (hereinafter referred to as "mental symptom information"), or information describing a mental state that can be indicated by the results of a neuropsychological test (hereinafter referred to as "neuropsychological test information").

[0174] Mental illness information may include, for example, information describing conditions relating to alcohol use disorder, depression or anxiety disorder, personality disorder (e.g., borderline personality disorder), or sleep disorder.

[0175] For example, alcohol use disorder is a disorder in which the amount and frequency of alcohol consumption become uncontrollable. For example, alcohol may temporarily alleviate depressive and anxiety symptoms, but it can also be a factor in the exacerbation of these symptoms. For example, it is thought that the decline in cognitive inhibitory function due to alcohol makes impulsivity and interpersonal problems more likely to surface when drinking. This may cause exacerbation of personality disorders. For example, sleep disorders may occur due to the disruption of sleep and REM inhibition caused by alcohol. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and mental illness information.

[0176] Mental symptom information may include, for example, information describing conditions related to cognitive function, impulsivity, irritability, or concentration.

[0177] For example, cognitive decline may occur due to the suppression of frontal lobe function caused by alcohol consumption. For example, the decrease in cognitive inhibitory function due to alcohol consumption may increase impulsivity. Increased impulsivity may induce problematic behaviors such as verbal abuse, violence, or gambling. For example, rebound arousal, hypoglycemia, or neurotransmitter disruption caused by alcohol consumption may result in irritability (e.g., irritability) or decreased concentration. From these results, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and the mental symptom information.

[0178] Neuropsychological test information may include, for example, information representing mental states that can be indicated by results of mood tests, anxiety tests, attention or sustained concentration tests, executive function tests, memory function tests, or stress tests.

[0179] The results of mood state tests include, for example, the results of tests such as POMS (Profile of Mood States). In POMS, the "tension-anxiety" score and the "anger-hostility" score temporarily decrease immediately after alcohol consumption. Also, in POMS, the "fatigue" score and the "depression-downness" score tend to increase the morning after drinking or in cases of chronic drinking.

[0180] Anxiety-related test results include, for example, the results of tests such as the STAI (State-Trait Anxiety Inventory). In the STAI, acute alcohol consumption temporarily lowers the "state anxiety" score. Also, in the STAI, chronic alcohol consumption or alcohol dependence is associated with a high "trait anxiety" score.

[0181] Attention or sustained concentration test results are obtained from tests such as the Psychomotor Vigilance Task (PVT). In the PVT, reaction delays and errors increase after alcohol consumption. Furthermore, in the PVT, performance declines are significantly more pronounced in individuals who are both intoxicated and sleep-deprived.

[0182] Executive function tests include results from tests such as the WCST (Wisconsin Card Sorting Test), Stroop test, and TMT (Trail Making Test). Executive function and frontal lobe function are affected by alcohol consumption. Chronic drinkers may also experience decreased flexibility and impulse control disorders. These are reflected in the results of tests such as the WCST.

[0183] Memory function tests include results from tests such as the Digit Span and the Wechsler Memory Scale (WMS). Long-term alcohol consumption impairs working memory. This decline in working memory is reflected in the results of tests such as the Digit Span.

[0184] Stress test results include, for example, the results of tests using psychological stress scales. Moderate alcohol consumption may reduce mental burden through effects such as mood enhancement. On the other hand, excessive alcohol consumption may cause mental burden. Such mental burden is reflected in psychological stress scales, etc.

[0185] Mental load information may include, for example, information representing a mental state that can be indicated by the results of a mental state test. The results of a mental state test (e.g., stress) are, for example, the results of a psychological stress scale test.

[0186] From the above, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and the neuropsychological test information.

[0187] Furthermore, mental state information may include, for example, information regarding accompanying symptoms of mental illnesses or symptoms, or information regarding side effects or complications of treatment for mental illnesses or symptoms.

[0188] (Information on intake of specific components in mental and physical state information M100) For example, the mental and physical state information M100 of the biological state information M1 may include information on the intake of specific components (hereinafter referred to as "information on intake of specific components"). Intake of specific components may include, for example, taking a specific drug, or ingesting food or poisons containing a specific component.

[0189] In the information on specific component intake, "specific component" refers to a component other than alcohol that causes a sustained increase in heart rate in the learning subject during inactivity. As mentioned above, the alcohol consumption estimation index information D1 is information that quantifies the sustained increase in heart rate in the learning subject during inactivity. Therefore, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and the information on specific component intake. For this reason, by training the learning model TM1 on the relationship between the alcohol consumption estimation index information D1 and the information on specific component intake, it is possible to estimate the information on specific component intake related to the intake of specific components other than alcohol.

[0190] Specific ingredients include, for example, caffeine (coffee, green tea, energy drinks), nicotine (smoking, heated tobacco products), theobromine (chocolate, cocoa), guarana, mate, yohimbine, ephedrines (cold medicines, herbal medicines), thyroid hormones (such as levothyroxine), antidepressants (tricyclic antidepressants, SSRIs, SNRIs), β-agonists (bronchodilators), antihistamines (first generation), herbal medicines (containing ephedra: such as Ma Huang Tang and Bo Feng Tong Sheng San), amino acids (L-tyrosine, L-phenylalanine), GABA supplements, BCAA amino acids (energy supplements), capsaicin (chili peppers), carbon monoxide (CO), organic solvents (toluene, benzene), or carbohydrates (high GI foods).

[0191] Furthermore, information regarding the intake of specific ingredients may include, for example, information regarding side effects associated with the intake of those specific ingredients.

[0192] (QOL information in mental and physical state information M100) For example, the mental and physical state information M100 of the biological state information M1 may include information on the quality of life (QOL) of the learning subject (hereinafter referred to as "QOL information"). QOL information may indicate, for example, the physical state, mental state, or quality of life related to the intake of specific components.

[0193] For example, the presence or absence of alcohol consumption, the amount consumed, habits, and total amount consumed can directly lead to a decline in quality of life. Furthermore, it is believed that alcohol consumption can cause or exacerbate physical and mental changes, discomfort, or abnormalities, thus reducing quality of life. From these points, it can be inferred that there is a correlation between the alcohol consumption estimation index D1 and QOL information.

[0194] For example, QOL information may include information about the decline in social functioning of learning subjects due to alcohol consumption. Information about the decline in social functioning may include, for example, information indicating workplace absenteeism, social isolation, or divorce. Such declines in social functioning may result from decreased motivation or breakdown of relationships associated with long-term alcohol consumption. Therefore, it can be inferred that there is a correlation between QOL information indicating a decline in social functioning and the alcohol consumption estimation index information D1.

[0195] This section explains how QOL information can indicate the occurrence of divorce. Frequent and heavy drinking can create financial burdens. Furthermore, frequent and heavy drinking can repeatedly cause temporary declines in cognitive and social functioning. In addition, frequent and heavy drinking can increase the risk of various physical and mental ailments and diseases, and may increase the risk of domestic violence. Since these factors can be contributing to divorce, it is thought that there is a certain correlation, albeit indirect, between drinking and divorce. The drinking estimation index information D1 is information that can estimate drinking with high accuracy. Therefore, by training the learning model TM1 with the relationship between the drinking estimation index information D1 and QOL information indicating the occurrence of divorce, the learning model TM1 can predict the occurrence of future divorces. In this case, for example, the drinking estimation index information D1 quantified over a long period of time, such as one month to several years, is used.

[0196] (Atmospheric environmental information in mental-body environmental information M110) For example, the mental-body environmental information M110 of the biological-related state information M1 may include information on atmospheric environmental factors (hereinafter referred to as "atmospheric environmental information"). The mental-body environmental information M110 may be represented, for example, by information indicating whether there is an abnormality in the environmental state, by information indicating the degree of the environmental state in stages, or by information indicating the environmental state as points.

[0197] Atmospheric environmental factors include the concentration of air pollutants or oxygen concentration. Furthermore, factors such as whether or not someone drinks alcohol, the amount consumed, and the frequency of drinking may be affected by stress from air pollution or low-oxygen environments. Therefore, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and atmospheric environmental information.

[0198] (Fetal effect information in passive biological information M120) For example, passive biological information M120 may include information regarding effects on the fetus (hereinafter referred to as "fetal effect information").

[0199] Fetal impact information may include, for example, information describing the effects of fetal alcohol syndrome on the fetus (e.g., whether or not there is a disability), or information regarding developmental disorders in children (e.g., whether or not there is a developmental disorder). Developmental disorders include, for example, attention deficit or behavioral disorders.

[0200] For example, Fetal Alcohol Syndrome is a syndrome in which alcohol consumption during pregnancy causes serious effects on the fetus's brain, facial features, and development. For example, attention deficits or behavioral disorders in children may be caused by alcohol consumption during pregnancy. From these points, it can be inferred that there is a correlation between the alcohol consumption estimation index information D1 and the fetal effect information.

[0201] (Information on interpersonal effects in passive biological information M120) For example, passive biological information M120 may include information on interpersonal effects (hereinafter referred to as "interpersonal effects information").

[0202] Interpersonal impact information may include, for example, information on interpersonal violence (e.g., whether or not interpersonal violence occurred, whether or not there was a risk of interpersonal violence occurring, whether or not a public institution was notified or consulted regarding interpersonal violence), or information on the mental burden on family members or cohabitants (e.g., whether or not there was a mental burden). Interpersonal violence may include, for example, child abuse or domestic violence.

[0203] For example, child abuse and domestic violence may occur due to decreased inhibition or increased impulsivity caused by alcohol consumption. Similarly, family members or cohabitants of alcoholics may experience mental strain due to the need for assistance and the impact of their symptoms. Therefore, it can be inferred that there is a correlation between the alcohol consumption estimation index D1 and interpersonal impact information.

[0204] (Alcohol Information M130) For example, alcohol information M130 includes information indicating whether or not alcohol is consumed, whether or not there is a drinking habit, information indicating the frequency of drinking (e.g., daily drinking, drinking two to three times a week, etc.), or information indicating the amount of alcohol consumed (e.g., total amount of alcohol consumed over a certain period).

[0205] (How to represent biological status information M1) Biological status information M1 is represented, for example, by information expressed as a score, binary classification, or multi-class classification, based on the results of a diagnosis or evaluation by a medical professional, the individual, or a third party. Diagnosis and evaluation include diagnosis and evaluation based on the results of examinations using medical devices such as medical imaging devices, and diagnosis and evaluation based on the results of biological examinations such as biomarkers. In binary classification, biological status information M1 is, for example, information indicating the presence or absence of disease, symptoms, abnormalities, effects, or intake. In multi-class classification, biological status information M1 is, for example, information indicating the degree of disease, symptoms, abnormalities, effects, quality of life, or environmental conditions in stages. For example, in multi-class classification, the severity of disease or symptoms may be indicated.

[0206] For example, in the case of tuberculosis, in binary classification by healthcare professionals, the biological status information M1 is either information indicating a "diagnosis" of tuberculosis (e.g., 1) or information indicating "no diagnosis" of tuberculosis (e.g., 0), based on the diagnostic results from chest X-ray or CT.

[0207] Furthermore, the biological status information M1 may include, for example, values ​​of evaluation or test indicators for evaluating or testing diseases, symptoms, abnormalities, effects, quality of life, or environmental conditions, information obtained by classifying such values ​​into binary categories, or information obtained by classifying such values ​​into multiple classes. Evaluation or test indicators include, for example, disease activity indicators, biopsy indicators, severity scores, functional assessment scales, prognosis scores, symptom assessment scales, or tests. Symptom assessment scales include, for example, assessment scales such as pain scales (e.g., VAS: Visual Analogue Scale) or subjective scales. Tests include, for example, PVT.

[0208] For example, in the binary classification of hyperuricemia in men, the biological status information M1 indicates "diagnosis of hyperuricemia" (e.g., 1) when the serum uric acid level is greater than 7.0 [mg / dL], and indicates "no diagnosis of hyperuricemia" (e.g., 0) when the serum uric acid level is 7.0 [mg / dL] or less.

[0209] Furthermore, the biological status information M1 may be, for example, a numerical value indicating the result of a biological test such as a biomarker, information obtained by classifying the numerical value into binary categories, or information obtained by classifying the numerical value into multiple classes.

[0210] Furthermore, the biological state information M1 may be represented, for example, by information showing a score, binary classification, or multi-class classification of the overall result combining the results from the above example.

[0211] As described above, the biological state information M1 may directly represent the mental and physical state of the learning subject or the environment, or the effect on the biological body, using numerical values ​​(direct numerical output), binary classification (binary classification output), or multi-class classification (multi-class classification output).

[0212] (Feature Information G1) Next, with reference to Figure 21, feature information G1 will be explained. Feature information G1 may further include one or more of the following: vital information H1 of the learning subject, behavioral information J1 of the learning subject, information R1 of the learning subject's environment (hereinafter referred to as "environmental information R1"), information V1 of the learning subject's physical and mental state reported by the subject or a third party, and attribute information Q1 of the learning subject. This is because these pieces of information are correlated with bio-related state information M1, and therefore the estimation accuracy of bio-related state information M2 by the learning model TM1 can be further improved. For example, feature information G1 may include vital information H1 and / or behavioral information J1 in addition to the alcohol consumption estimation index information D1.

[0213] First, we will explain the correlation between vital information H1 and biological state information M1.

[0214] For example, vital information H1 is an objective indicator that shows the physiological functional state of a learning subject and directly responds to changes in the body's state. Since information representing the body's state is reflected in these measurements, the two are closely related. Therefore, it can be inferred that there is a correlation between vital information H1 and the body state information in biologically related state information M1.

[0215] For example, vital information H1 is an objective indicator that reflects the autonomic nervous system activity of a learning subject and fluctuates in response to changes in mental state. Since mental state influences vital information through stress responses and emotional changes, the two are closely related. Therefore, it can be inferred that there is a correlation between vital information H1 and the mental state information of biologically related state information M1.

[0216] For example, vital information H1 objectively shows the biological response to the intake of specific components such as pharmaceuticals. Since these components directly affect vital indicators such as heart rate, changes in intake status are reflected in vital information. Therefore, it can be inferred that there is a correlation between vital information H1 and the specific component intake information in biological state information M1.

[0217] For example, vital information H1 is an objective indicator that reflects the state of the learning subject's biological functions and is related to changes in quality of life. A decline in quality of life leads to physical and mental stress, which manifests as abnormalities in vital information, so the two influence each other. Therefore, it can be inferred that there is a correlation between vital information H1 and QOL information from biological state information M1.

[0218] For example, vital information H1 is an objective indicator showing the environmental response of a learning subject's body, and it reacts sensitively to changes in environmental conditions. Environmental abnormalities such as hypoxia and air pollution activate the body's compensatory functions, which are observed as fluctuations in vital information. Therefore, it can be inferred that there is a correlation between vital information H1 and the atmospheric environment information of the body-related state information M1.

[0219] For example, vital information H1 is an objective indicator of the mother's physiological state and fluctuates due to factors such as alcohol consumption. Changes in the mother's state can also affect the fetus. Therefore, it can be inferred that there is a correlation between vital information H1 and fetal impact information.

[0220] For example, vital information H1 is an objective indicator that shows physiological changes caused by alcohol consumption. Excessive alcohol consumption can lead to interpersonal consequences such as interpersonal violence and mental burden on family members. Therefore, it can be inferred that there is a correlation between vital information H1 and information on interpersonal effects.

[0221] For example, behavioral information J1 is an objective indicator showing the exercise intensity and exercise patterns of learning subjects, and is closely related to their physical condition. Physical illnesses and functional impairments manifest as a decrease in exercise intensity and step count, while appropriate physical activity improves physical function. Therefore, it can be inferred that there is a correlation between behavioral information J1 and the physical condition information in biologically related state information M1.

[0222] For example, behavioral information J1 is an objective indicator showing the exercise intensity and exercise patterns of learning subjects, and reflects their mental state. Changes in mental state, such as anxiety or depression, manifest as changes in exercise intensity and activity rhythm, and are reflected in the fluctuation patterns of behavioral information. Therefore, it can be inferred that there is a correlation between behavioral information J1 and the mental state information in biologically related state information M1.

[0223] For example, behavioral information J1 is an objective indicator showing the exercise intensity and exercise patterns of learning subjects, and is related to the intake status of specific components. Ingestion of specific components such as drugs affects physical activity patterns and exercise intensity, which is observed as a change in behavioral information. Therefore, it can be inferred that there is a correlation between behavioral information J1 and the specific component intake information in biological state information M1.

[0224] For example, behavioral information J1 is an objective indicator showing the exercise intensity and exercise patterns of learning subjects, and is directly related to quality of life. A decline in quality of life manifests as a decrease in exercise intensity and a decrease in the number of steps taken, and is clearly reflected in the behavioral information. Therefore, it can be inferred that there is a correlation between behavioral information J1 and the QOL information of biologically related state information M1.

[0225] For example, behavioral information J1 is an objective indicator showing the exercise intensity and exercise patterns of learning subjects, and it changes in accordance with environmental conditions. Atmospheric abnormalities such as hypoxia and air pollution limit exercise intensity and physical activity, and are detected as changes in behavioral information. Therefore, it can be inferred that there is a correlation between behavioral information J1 and the atmospheric environment information of biologically related state information M1.

[0226] For example, behavioral information J1 is an objective indicator of the mother's lifestyle and physical activity, and it changes in response to factors such as alcohol consumption. Changes in the mother's behavioral patterns and lifestyle can affect the health and development of the fetus. Therefore, it can be inferred that there is a correlation between behavioral information J1 and fetal impact information.

[0227] For example, behavioral information J1 is an objective indicator showing changes in behavior due to alcohol consumption, etc. Changes in behavior can lead to interpersonal effects such as interpersonal violence or mental burden on family members. Therefore, it can be inferred that there is a correlation between behavioral information J1 and information on interpersonal effects.

[0228] For example, environmental information R1 is an objective indicator of the external conditions in which the learning subject is placed, and can cause changes related to physical state, mental state, intake of specific components, quality of life (QOL), air environment, fetal effects, or interpersonal effects. Therefore, it can be inferred that there is a correlation between environmental information R1 and each of these pieces of information in biological state information M1.

[0229] The information V1 reported by the learning subject themselves, that is, the self-reported information, captures each aspect of the biological state information M1 from a subjective perspective, and it can be inferred that there is a correlation between the two.

[0230] Third-party report information W1 captures each aspect of the biological condition information M1 from an objective perspective, and it can be inferred that there is a correlation between the two.

[0231] Attribute information Q1 represents the basic characteristics of the learning subject and indicates the fundamental background related to the content of the biological state information M1. Therefore, it can be inferred that there is a correlation between attribute information Q1 and biological state information M1.

[0232] In particular, if feature information G1 further includes one or more of the following information from the learning subject's vital information H1, behavioral information J1, environmental information R1, reporting information V1, and attribute information Q1, then at least the content or timing of the feature information G1 (explanatory variable) and the biological state information M1 (dependent variable) will be different. This is because the dependent variable, which is the target of estimation, and the explanatory variables used to explain the dependent variable for estimation will be different, at least in terms of content or timing. For example, even if the content of feature information G1 and biological state information M1 is the same, if feature information G1 is past information and biological state information M1 is current or latest information, then the timing of the feature information G1 and biological state information M1 will be different.

[0233] Next, we will explain the details of the contents of vital information H1, etc.

[0234] Vital information H1 includes information about heart rate. In this specification, heart rate is not limited to heart rate based on electrocardiogram waveforms obtained by electrocardiography (ECG), but also includes pulse rate based on pulse wave waveforms. In other words, heart rate based on electrocardiogram waveforms and pulse rate based on pulse wave waveforms are treated as "heart rate" in this specification. The method for obtaining pulse wave waveforms is not particularly limited and may be obtained, for example, by photoplethysmography (PPG).

[0235] Information related to heart rate includes, for example, heart rate (HR), R-R interval, time-domain indicators related to heart rate, frequency-domain indicators related to heart rate, or nonlinear indicators related to heart rate. Heart rate is the number of times the heart beats in a given period of time, and is expressed, for example, as beats per minute (bpm). A given period of time in heart rate may be referred to as the first given period. The R-R interval is the time interval from one QRS wave to the next in the electrocardiogram waveform. Time-domain indicators related to heart rate, frequency-domain indicators related to heart rate, and nonlinear indicators related to heart rate are collectively called heart rate variability (HRV) indicators. In this specification, for example, pulse rate (PR) is treated as heart rate, pulse interval (PI) is treated as R-R interval, and pulse variability is treated as heart rate variability.

[0236] Time-domain metrics include, for example, SDNN (Standard deviation of NN intervals), RMSSD (Root Mean Square of Successive Differences), CVRR (Coefficent of Variation of RR intervals), SDRR (Standard deviation of RR intervals), SDANN (Standard Deviation of the Average NN intervals for each 5-minute segment of a 24-hour HRV recording), SDNN index, NN50 (the number of pairs of successive NN intervals that differ by more than 50 ms), pNN50 (the proportion of NN50 divided by the total number of NN intervals), HR Max, HR Min, (HR Max - HR Min), HTI (HRV Triangular Index), or TINN (Triangular Interpolation of the NN Interval Histogram). HRV stands for heart rate variability. RMSSD, in terms of heart rate, is the square root of the average of the squared differences between consecutive adjacent R-R intervals.

[0237] Frequency domain indicators include, for example, LF (low-frequency power or peak), HF (high-frequency power or peak), LF / HF, Total Power, LF Norm, HF Norm, ULF (ultra-low frequency power), or VLF (very low frequency power) of heart rate variability.

[0238] Nonlinear indices include, for example, entropy, SD1 (standard deviation in the direction orthogonal to y=x in a Poincaré plot), SD2 (standard deviation in the direction along y=x in a Poincaré plot), SD1 / SD2, ApEn (Approximate entropy), SampEn (Sample entropy), DFA α1 (Detrended Fluctuation Analysis, which describes short-term fluctuations), DFA α2 (Detrended Fluctuation Analysis, which describes long-term fluctuations), (DFA α1) / (DFA α2), CVI (Cardiac Vagal Index), or CSI (Cardiac Sympathetic Index).

[0239] Vital information H1 may include, for example, one or more of the following: blood pressure information, respiratory information, body temperature information, blood information, and electroencephalogram (EEG) information. Respiratory information may include, for example, respiratory rate, respiratory rate, or respiratory volume. Body temperature information may include, for example, at least one of the following: skin temperature, body temperature, and core body temperature. Blood information may include, for example, arterial blood oxygen saturation (e.g., SpO2). ２ ) or blood sugar level.

[0240] Activity information J1 includes at least one of the following: information on the number of steps and information on exercise intensity. Information on the number of steps and information on exercise intensity are types of exercise volume information. The number of steps and exercise intensity may be collectively referred to as exercise volume. Information on the number of steps may, for example, indicate the number of steps (spm: steps per minute) within a certain period of time (e.g., 1 minute). A certain period of time in terms of steps may be referred to as a second certain period of time. Information on exercise intensity may, for example, be indicated by METs. METs may, for example, indicate METs over a certain period of time (e.g., 1 minute). A certain period of time in terms of METs may be referred to as a third certain period of time. METs is a unit that expresses the intensity of physical activity as a multiple of the resting state. Note that information on exercise intensity is not limited to METs, and may also be indicated by, for example, a relative value to maximum oxygen uptake (%VO2max), or by a method using maximum heart rate (%HRmax, %MHR).

[0241] Furthermore, the behavioral information J1 may include, for example, one or more pieces of information from energy consumption and body movement information. Body movement information may be, for example, the amount or frequency of a specific body movement (e.g., turning over or struggling). In addition, the behavioral information J1 may include the time the learning subject wore a biological state detection device (e.g., a wearable device).

[0242] Furthermore, behavioral information J1 may include sleep information of the learning subject. Sleep information is information about the body's sleep. Sleep information includes at least one of the following: information about sleep duration and information about sleep rhythm. Sleep information may also include information about sleep state and / or information about sleep-related medications.

[0243] Furthermore, behavioral information J1 may include, for example, smoking information. Smoking information may include, for example, information indicating whether or not a person smokes or their smoking history.

[0244] Environmental information R1 is information that indicates the environment in which living organisms inhabit. Environmental information R1 includes, for example, one or more pieces of information from seasonal information and meteorological information. Seasonal information may be indicated by a seasonal name such as winter, or by a month such as January. Meteorological information includes, for example, weather or atmospheric pressure information.

[0245] Self-reported information V1 includes subjective reports of the physical and mental state of the learning subject themselves. Self-reported information V1 includes, for example, self-reports by the learning subject themselves regarding their physical condition, mental state, or quality of life. For example, self-reported information V1 may be in text format, may be the result of a binary or multi-class classification of subjective evaluations, or may be expressed numerically as the result of an evaluation scale, questionnaire, or survey. Self-reports regarding physical condition include, but are not limited to, subjective reports on fatigue, sleep (quantity, quality), exercise level, or the presence or severity of various physical symptoms. Self-reports regarding mental state include, but are not limited to, subjective reports on anxiety or psychological stress.

[0246] Third-party report information W1 includes objective reports by a third party indicating the physical and mental state of the learning subject. The third party may be, for example, a medical professional, a close relative, or a cohabitant. Third-party report information W1 includes, for example, objective reports by a third party regarding the physical state, mental state, or quality of life of the learning subject. For example, third-party report information W1 may be in text format, may be the result of a binary or multi-class classification of an objective evaluation, or may be expressed numerically as the result of an evaluation scale, questionnaire, or survey. Objective reports regarding the physical state include, but are not limited to, objective reports on fatigue, sleep (quantity, quality), exercise level, or the presence and severity of various physical symptoms. Objective reports regarding the mental state include, but are not limited to, objective reports on anxiety or psychological stress.

[0247] The information W1 reported by oneself or a third party may, for example, be current or up-to-date information, or it may be past information (information reported in the past).

[0248] The learning subject attribute information Q1 is information that indicates the attributes of the learning subject. Attribute information Q1 includes, for example, one or more of the following: basic information, anthropometric information, medical history information, diagnostic history information, drug use history information, lifestyle information, exercise-related information, genetic information, and disability-related information. Basic information includes, for example, information on age, sex, race, and / or place of residence. Anthropometric information includes, for example, information on height, weight, and / or BMI. Medical history information includes, for example, information on past medical history, surgical history, and / or current medical history. Diagnostic history information includes, for example, the results of health examinations or information on diagnostic images obtained by medical imaging equipment. Drug use history information includes, for example, current medication status and / or past medication information. Exercise-related information includes, for example, information on whether or not there is an exercise habit, the type of exercise, and / or the frequency of exercise. Disability-related information includes, for example, information on whether or not there is a disability, the type of disability, and / or the degree of disability.

[0249] In Embodiment 4, the alcohol consumption estimation indicator information D1, vital information H1, and behavioral information J1 may each include the alcohol consumption estimation indicator information, vital information, and behavioral information for a period indicated by specific conditions. The specific conditions are conditions related to sleep, conditions related to wakefulness, conditions related to the time of day caused by the sun, or a combination of two or more of these conditions.

[0250] The periods indicated by conditions related to sleep are, for example, the period during sleep, a predetermined period before falling asleep (a predetermined period immediately before falling asleep), or a predetermined period after falling asleep (a predetermined period immediately after falling asleep). The periods indicated by conditions related to wakefulness are, for example, the period while awake, a predetermined period before getting up (a predetermined period immediately before getting up), or a predetermined period after getting up (a predetermined period immediately after getting up). The periods indicated by conditions related to the time of day caused by the sun are, for example, a predetermined period during the day, a predetermined period at night, a predetermined period in the morning, a predetermined period in the afternoon, or a predetermined period at night.

[0251] However, for the alcohol consumption estimation index information D1, the period during sleep, the predetermined period after falling asleep, and the predetermined period before waking up are not included in the "period indicated by specific conditions." This is because drinking behavior does not occur during these periods.

[0252] Furthermore, if the alcohol consumption estimation index information D1 is the second alcohol consumption estimation index F(d), the period indicated by the specific conditions corresponds to the fifth time Tf used when calculating the second alcohol consumption estimation index F(d).

[0253] The reason why the alcohol consumption estimation indicator information D1, vital information H1, and behavioral information J1 for a period indicated by specific conditions are effective in estimating the biological state information M2 is that biological functions or states exhibit diurnal fluctuations due to abnormalities in the mind, body, or environment. For example, biological functions or states may fluctuate depending on the time of day due to abnormalities in the mind, body, or environment. Therefore, by learning the alcohol consumption estimation indicator information D1, vital information H1, and behavioral information J1 for different times of day, it may be possible to estimate abnormalities in the mind, body, or environment with high accuracy. In particular, since the specific conditions are fundamental elements of biological rhythms, setting periods based on these conditions is especially effective in estimating the biological state information M2.

[0254] For example, in the time immediately before falling asleep, while the body is at rest, alcohol metabolism due to drinking continues, so heart rate variability due to drinking may be clearly apparent and less affected by other factors. Also, for example, immediately after waking up, the effects of drinking the previous night may alter the normal physiological transition from sleep to wakefulness and affect the heart rate pattern. Therefore, by learning indicator information D1 for estimating alcohol consumption during a predetermined period before falling asleep and / or a predetermined period after waking up, it may be possible to estimate biological state information M2 with high accuracy.

[0255] For example, the alcohol consumption estimation index information D1 may include the difference between a second alcohol consumption estimation index in a first period indicated by specific conditions and a second alcohol consumption estimation index in a second period indicated by specific conditions. The first and second periods are, for example, a predetermined period after waking up and a predetermined period before falling asleep, a morning period and an afternoon period, or a daytime period and a nighttime period.

[0256] [Input Information K2, Output Information L2] Next, with reference to Figure 22, the input information K2 input to the learning model TM1 and the output information L2 output from the learning model TM1 will be explained. Figure 22 is a diagram showing an example of input information K2 and output information L2. As shown in Figure 22, the input information K2 includes at least the subject's alcohol consumption estimation index information D2. The alcohol consumption estimation index information D2 includes information on the sum heart rate change SumΔHR(t) when the subject is inactive and the subject's heart rate exceeds the reference heart rate. Otherwise, the alcohol consumption estimation index information D2 is the same as the alcohol consumption estimation index information D1 for the learning subject.

[0257] The input information K2 may further include one or more of the following: the subject's vital information H2, the subject's behavioral information J2, the subject's environment information R2 (hereinafter referred to as "environmental information R2"), the subject's self-reported information V2 regarding the subject's physical and mental state, and the subject's attribute information Q2. For example, the input information K2 may include vital information H2 and / or behavioral information J2 in addition to the alcohol consumption estimation index information D2. If the input information K2 includes vital information H2, the feature information G1 (Figure 21) includes vital information H1; if the input information K2 includes behavioral information J2, the feature information G1 includes behavioral information J1; if the input information K2 includes environmental information R2, the feature information G1 includes environmental information R1; if the input information K2 includes self-reported information V2, the feature information G1 includes self-reported information V1; and if the input information K2 includes attribute information Q2, the feature information G1 includes attribute information Q1.

[0258] Vital information H2 includes information about heart rate. Furthermore, the heart rate information in vital information H2 is the same as the heart rate information in vital information H1 shown in Figure 21.

[0259] Behavioral information J2 includes at least one of the following: information on the number of steps and information on exercise intensity. Behavioral information J2 may further include the subject's sleep information. Otherwise, the information on the number of steps, exercise intensity, and sleep information in behavioral information J2 are the same as the information on the number of steps, exercise intensity, and sleep information in behavioral information J1 in Figure 21, respectively.

[0260] Environmental information R2 is information indicating the environment in which the organism lives or works. Self-reported information V2 includes subjective reports of the subject's physical and mental state. Self-reported information V2 may be, for example, current or latest information, or past information (information reported in the past). Attribute information Q2 is information indicating the subject's attributes. In addition, environmental information R2, self-reported information V2, and attribute information Q2 in Figure 22 are the same as environmental information R1, self-reported information V1, and attribute information Q1 in Figure 21, respectively.

[0261] At least one of the alcohol consumption estimation indicator information D2, vital information H2, and behavioral information J2 may include alcohol consumption estimation indicator information, vital information, or behavioral information during a period indicated by specific conditions. The specific conditions are conditions related to sleep, conditions related to wakefulness, conditions related to the time of day caused by the sun, or a combination of two or more of these conditions. The specific conditions for alcohol consumption estimation indicator information D2, vital information H2, and behavioral information J2 are the same as the specific conditions for alcohol consumption estimation indicator information D1, vital information H1, and behavioral information J1 in Figure 21.

[0262] On the other hand, the output information L2 includes the subject's biological state information M2. The biological state information M2 includes information M200 representing the subject's mental and physical state (hereinafter referred to as "mental and physical state information M200"), information M210 representing the state of the environment that affects the subject's mental and physical state (hereinafter referred to as "mental and physical environment information M210"), information representing the state of the body affected by the subject's mental and physical state (hereinafter referred to as "passive biological information M220"), or information regarding the subject's alcohol consumption (hereinafter referred to as "alcohol consumption information M230"). The mental and physical state information M200, mental and physical environment information M210, passive biological information M220, and alcohol consumption information M230 in Figure 22 are the same as the mental and physical state information M100, mental and physical environment information M110, passive biological information M120, and alcohol consumption information M130 in Figure 21, respectively.

[0263] For example, the mental and physical state information M200 includes information representing the physical state (physical state information), information representing the mental state (mental state information), information regarding the intake of specific components (specific component intake information), or information regarding quality of life (QOL information). The QOL information indicates the quality of life related to the physical state, mental state, or intake of specific components.

[0264] Physical condition information may include, for example, information on diseases or symptoms of the nervous and cranial systems, information on diseases or symptoms of the circulatory system, information on diseases or symptoms of the respiratory system, information on diseases or symptoms of the digestive and hepatobiliary-pancreatic regions, information on diseases or symptoms of the endocrine and metabolic systems, information on diseases or symptoms of the renal and urinary systems, information on diseases or symptoms of sex hormones and the reproductive system, information on diseases or symptoms of the skin and sensory organs, information on diseases or symptoms related to immunity and infectious diseases, information on tumor symptoms, information on pre-disease conditions of the body, or information representing a physical condition that can be indicated by the results of a biopsy. Mental condition information may include, for example, information representing a state related to mental illness, information representing a state related to mental symptoms, or information representing a mental state that can be indicated by the results of a neuropsychological test. For example, the mind-body environment information M210 may include information on atmospheric environmental factors. For example, the passive biological information M220 may include information on effects on the fetus, or information on interpersonal effects.

[0265] In the explanation of Figure 21, for example, the learning subject, feature information G1, alcohol estimation index information D1, vital information H1, behavioral information J1, environmental information R1, self-reported information V1, attribute information Q1, correct label B1, biologically related state information M1, mental / physical state information M100, mental / physical / environmental information M110, passive biological information M120, and alcohol information M130 can be replaced with the subject, input information K2, alcohol estimation index information D2, vital information H2, behavioral information J2, environmental information R2, self-reported information V2, attribute information Q2, output information L2, biologically related state information M2, mental / physical state information M200, mental / physical / environmental information M210, passive biological information M220, and alcohol information M230, respectively, thereby substituting for the explanation of input information K2 and output information L2.

[0266] Typically, the alcohol consumption estimation indicator information D2, vital information H2, and behavioral information J2 are information created based on raw biological state data A2 acquired from a wearable device (e.g., the biological state detection device 103 described later) and / or a mobile terminal (e.g., the first terminal 102 described later). However, this information may also be acquired from a wearable device and / or a mobile terminal.

[0267] [Estimation System 40] Next, the stages of using the learning model TM1 will be described with reference to Figures 23 to 29. Figure 23 is a block diagram showing an example configuration of the estimation system 40 according to Embodiment 4. As shown in Figure 23, the estimation system 40 is connected to a network NW. The network NW includes, for example, the Internet, a private network, a public telephone network, a LAN (Local Area Network), and a short-range wireless network. The estimation system 40 is part of the information processing system 1 in Figure 20.

[0268] At least one client system 100 is connected to the network NW. The client system 100 comprises a cloud server 101, a plurality of first terminals 102, and a plurality of biological state detection devices 103. The cloud server 101, the first terminals 102, and the biological state detection devices 103 are connected to the network NW. In addition, a plurality of second terminals 200 are connected to the network NW.

[0269] The biological state detection device 103 detects the biological state of a subject. The biological state detection device 103 has, for example, a sensor that detects the biological state by contact or non-contact. The biological state detection device 103 outputs biological state raw data A2, which is raw data indicating the biological state of the subject. The biological state raw data A2 includes the subject's vital raw data and / or behavioral raw data. The vital raw data is raw data indicating the vital information of the organism. The behavioral raw data is raw data indicating the behavioral information of the organism. The sensor that detects the vital raw data includes, for example, an optical sensor (light-emitting element and light-receiving element) that performs PPG. In this case, for example, the subject's pulse wave waveform is detected. Therefore, the biological state detection device 103 calculates the pulse rate (bpm) based on the pulse wave waveform. The arterial blood oxygen saturation may be measured by the optical sensor. The sensor that detects the vital raw data may include, for example, a sensor that detects the subject's body temperature or skin temperature. The sensor that detects the behavioral raw data includes, for example, an acceleration sensor and / or a gyroscope sensor. In this case, for example, the subject's step count and exercise intensity (METs) are detected. That is, the biological state detection device 103 calculates the step count and exercise intensity (METs) based on the output of the acceleration sensor and / or gyroscope sensor. The sensors for detecting behavioral data may include, for example, an acceleration sensor and / or gyroscope sensor, as well as a microphone. In this case, for example, the subject's sleep state is detected.

[0270] The biological state detection device 103 is, for example, a wearable device. The wearable device is worn by the subject. The wearable device is, for example, a wristwatch, a ring, or a sticker. The wearable device may also be, for example, the wearable terminal WT shown in Figure 1. The biological state detection device 103 is synchronized with the first terminal 102 and transmits the raw biological state data A2 to the first terminal 102. The first terminal 102 is, for example, a mobile terminal such as a smartphone. The mobile terminal is carried by the subject. The biological state detection device 103 may also transmit the raw biological state data A2 to the cloud server 101 or the estimation system 40 via the network NW.

[0271] The biological state detection device 103 may be wearable, portable, or stationary, or it may be a dedicated detector (measuring instrument) for detecting the biological state. Furthermore, the first terminal 102 may have some or all of the functions of the biological state detection device 103. Also, for example, the first terminal 102 may be a personal computer (PC).

[0272] The first terminal 102 transmits the subject's raw biological state data A2 to the cloud server 101 via the network NW. The cloud server 101 transmits the raw biological state data A2 to the estimation system 40 via the network NW. The estimation system 40 processes the raw biological state data A2. Alternatively, the first terminal 102 may transmit the raw biological state data A2 to the estimation system 40 via the network NW without providing a cloud server 101. The raw biological state data A2 may also be transmitted directly from the biological state detection device 103 to the estimation system 40. One or more of the environmental information R2, self-report information V2, and attribute information Q2 may be transmitted from the first terminal 102 to the estimation system 40 via the cloud server 101 or directly. Some of this information may also be transmitted from the second terminal 200 to the estimation system 40 via the cloud server 101 or directly.

[0273] More specifically, the estimation system 40 comprises a relay server 44, a first database 45, an estimation device 4, a second database 46, and an information providing server 47. The relay server 44, the first database 45, the estimation device 4, the second database 46, and the information providing server 47 are connected to a network NW. The estimation device 4 is, for example, a server. The server is a computer.

[0274] Each of the relay server 44 and the information providing server 47 may include a processing unit, a communication unit, and a storage unit, and may also include an input unit and an output unit. The hardware configuration of the processing unit, communication unit, storage unit, input unit, and output unit is the same as the hardware configuration of the processing unit 400, communication unit 401, storage unit 402, input unit 403, and output unit 404 of the estimation device 4 shown in Figure 25, which will be described later, or the hardware configuration of the drinking estimation device IS shown in Figure 16 or Figure 17. In addition, the first database 45 and the second database 46 include at least a storage device such as a hard disk drive. The first database 45 and the second database 46 may have the same hardware configuration as the relay server 44 or the information providing server 47. The estimation system 40 does not need to include all or part of the relay server 44, the first database 45, the second database 46, and the information providing server 47.

[0275] [Estimation Method] Figure 24 is a flowchart showing an example of an estimation method performed by the estimation system 40. The estimation method estimates the subject's biological state information MX. The estimation method includes steps S51 to S55.

[0276] As shown in Figures 23 and 24, first, in step S51, the relay server 44 receives the subject's biological state raw data A2 from the cloud server 101. The relay server 44 may also receive one or more of the subject's environmental information R2, self-reported information V2, and attribute information Q2. The relay server 44 implements, for example, an API (Application Programming Interface).

[0277] Next, in step S52, the first database 45 stores the raw biological state data A2 received by the relay server 44. Alternatively, for example, the first database 45 may also store one or more of the following: environmental information R2, self-reported information V2, and attribute information Q2. Specifically, in the first database 45, a record 451 is assigned to each subject. The subject's identification information (personal identification information) is associated with the record 451. The raw biological state data A2 is then recorded in the subject's record 451. The record 451 may also contain one or more of the following: environmental information R2, self-reported information V2, and attribute information Q2.

[0278] Next, in step S53, the estimation device 4 estimates the subject's biological state information MX using the alcohol consumption estimation index information D2 based on the biological state raw data A2 stored in the first database 45 and the learning model TM1. Details of the estimation process will be described later. In addition, one or more of the environmental information R2, self-reported information V2, and attribute information Q2 may be used in the estimation process.

[0279] Next, in step S54, the second database 46 stores the subject's bio-related state information MX (output information L2) estimated by the estimation device 4. Specifically, in the second database 46, a record 461 is assigned to each subject. The subject's identification information (personal identification information) is associated with the record 461. The bio-related state information MX is then recorded in the subject's record 461. The second database 46 also stores input information K2, which includes the alcohol consumption estimation index information D2. Specifically, the input information K2 is recorded in the subject's record 461.

[0280] Next, in step S55, the information provision server 47 transmits the subject's biological status information MX stored in the second database 46 to the second terminal 200 and / or the first terminal 102 via the network NW. Specifically, the information provision server 47 displays the subject's biological status information MX on the second terminal 200 and / or the first terminal 102. The second terminal 200 is, for example, a terminal of a medical institution, a health management institution, or a research institution. The first terminal 102 is, for example, the subject's terminal. When step S5 is completed, the estimation method is finished.

[0281] In the example shown in Figure 23, the biological status information MX is transmitted to the first terminal 102 via the cloud server 101, but it may also be transmitted directly to the first terminal 102.

[0282] [Estimation Device 4] Figure 25 is a block diagram showing an example configuration of the estimation device 4 in Figure 23. As shown in Figure 25, the estimation device 4 comprises a processing unit 400, a communication unit 401, and a storage unit 402. The estimation device 4 may further comprise an input unit 403 and an output unit 404.

[0283] The input unit 403 is an input device for inputting various types of information to the processing unit 400. For example, the processing unit 400 may be a keyboard and pointing device, or a touch panel.

[0284] The output unit 404 outputs various types of information. The output unit 404 includes, for example, a display unit that displays various types of information. The display unit is, for example, a liquid crystal display or an organic electroluminescent display.

[0285] The communication unit 401 is connected to a network NW. The communication unit 401 communicates with external devices connected to the network NW. The communication unit 401 is a communication device that communicates according to a predetermined communication protocol, and includes, for example, a network interface controller. The predetermined communication protocol is, for example, a protocol compliant with Ethernet®, an Internet Protocol Suite, and a protocol compliant with a short-range wireless communication standard. The external devices are, for example, a first terminal 102, a second terminal 200, and a biological state detection device 103.

[0286] The storage unit 402 includes one or more storage devices and stores data and computer programs. The storage unit 402 includes a main memory such as a semiconductor memory and an auxiliary storage device such as a semiconductor memory and a hard disk drive. The storage unit 402 may also include removable media such as an optical disc. The storage unit 402 may be, for example, a non-temporary computer-readable storage medium.

[0287] The memory unit 402 stores the learning model TM1. The learning model TM1 is a trained model. The learning model TM1 is a computer program. The learning model TM1 causes the computer to function in order to estimate the subject's biological state information M2. Specifically, the learning model TM1 takes input information K2 as input and causes the computer to output output information L2.

[0288] The processing unit 400 performs various calculations. The processing unit 400 includes one or more processors. The processors are CPUs (Central Processing Units), GPUs (Graphics Processing Units), FPGAs (Field Programmable Gate Arrays), DSPs (Digital Signal Processors), or ASICs (Application Specific Integrated Circuits). The processors may be operated by computer programs or by hardwired logic.

[0289] Specifically, the processing unit 400 includes a pre-processing unit 41 and an estimation unit 42. The processing unit 400 may further include a post-processing unit 43. For example, the processor of the processing unit 400 functions as the pre-processing unit 41, estimation unit 42, and post-processing unit 43 by executing a computer program stored in the storage device of the storage unit 402. The pre-processing unit 41 includes an alcohol consumption index calculation unit 411. The pre-processing unit 41 may further include at least one of a statistical processing unit 412 and a condition processing unit 413. The post-processing unit 43 preferably includes an interpretation unit 431.

[0290] (Pre-processing unit 41) The pre-processing unit 41 acquires raw biological state data A2 from the first database 45. The storage unit 402 stores the raw biological state data A2. The raw vital data of the raw biological state data A2 includes, for example, heart rate data. The raw behavioral data of the raw biological state data A2 includes, for example, at least one of the following data: step count data and exercise intensity data. The pre-processing unit 41 may acquire one or more of the following information from the first database 45: environmental information R2, self-report information V2, and attribute information Q2.

[0291] Heart rate data may include, for example, information on heart rate (bpm). Heart rate data may also include, for example, information on the R-R interval. Heart rate is measured or aggregated at a sampling rate of, for example, at least once per minute. Heart rate data is accompanied by the corresponding measurement date, time, minute, second and subject identification information. Heart rate data includes information on heart rate arranged in time series. Heart rate data may also include information on the R-R interval arranged in time series.

[0292] Step count data includes step count (spm) information. Step counts are measured or aggregated, for example, at a sampling rate of at least once per minute. Step count data is accompanied by the corresponding measurement date, time (date, minute, second) and subject identification information. Step count data includes step count information arranged in chronological order.

[0293] Exercise intensity data includes METs information per unit of time (1 minute). In other words, exercise intensity is typically expressed in METs. Exercise intensity data is measured or aggregated, for example, at a sampling rate of at least once per minute. Exercise intensity data is accompanied by the corresponding measurement date, time (date, minute, second) and subject identification information. Exercise intensity data includes METs information arranged in a time series.

[0294] The preprocessing unit 41 performs preprocessing on the biological state raw data A2 and generates input information K2, which is the result of the preprocessing. The storage unit 402 stores the input information K2.

[0295] As an example, the alcohol consumption index calculation unit 411 of the preprocessing unit 41 calculates alcohol consumption estimation index information D2 based on the vital raw data (heart rate data and behavioral raw data, e.g., exercise intensity data or step count data) for each unit period UT. For example, the unit period UT corresponds to the fifth time Tf when calculating the second alcohol consumption estimation index F(d), and the alcohol consumption index calculation unit 411 calculates the second alcohol consumption estimation index F(d). The unit period UT is, for example, one day. Specifically, the alcohol consumption index calculation unit 411 calculates the alcohol consumption estimation index information D2 in the same manner as the alcohol consumption estimation device IS (processing unit SY(IS)) of Embodiments 1 to 3. The alcohol consumption index calculation unit 411 may also calculate the second alcohol consumption estimation index F(d) by performing statistical processing on the first alcohol consumption estimation index f(t), similar to the statistical processing performed by the statistical processing unit 412.

[0296] As an example, the statistical processing unit 412 of the preprocessing unit 41 performs statistical processing on each of the heart rate data, step count data, and exercise intensity data for each unit period UT. As a result, the statistical processing unit 412 outputs multiple statistical indicators, which are the results of the statistical processing, for each of the heart rate data, step count data, and exercise intensity data. The statistical processing by the statistical processing unit 412 represents mathematical processing to quantitatively express the characteristics and trends of the data to be processed. Therefore, by performing statistical processing, the characteristics and trends of the data to be processed (heart rate data, step count data, and exercise intensity data) can be accurately extracted, and the estimation accuracy of the bio-related state information M2 by the learning model TM1 can be further improved. The heart rate data, step count data, and exercise intensity data are, in detail, the extracted heart rate data, extracted step count data, and extracted exercise intensity data, which will be described later. In addition, the statistical processing unit 412 may perform difference processing and ratio processing for each unit period UT. Details of this will be described later.

[0297] Statistical indicators include, for example, the mean, sum, standard deviation, variance, mean square continuity error, xth percentile value, median, minimum, maximum, RMSSD, coefficient of variation CV (= standard deviation / mean), reciprocal of the coefficient of variation CV, mean change, percentage of changes greater than or equal to a specified value, statistically processed value of the moving median in the time window, and heart rate variability index. Furthermore, RMSSD is calculated not only for heart rate (bpm) but also for steps per minute (SPM) and exercise intensity (METs / min). In this case, RMSSD is the square root of the mean of the squared differences of consecutively adjacent values. Consecutively adjacent values ​​are, for example, the reciprocal of consecutively adjacent "heart rate (bpm)", consecutively adjacent "steps per minute (SPM)", or consecutively adjacent "exercise intensity (METs / min)".

[0298] The statistical indicators calculated by the statistical processing unit 412 constitute a part of the input information K2 (vital information H2, behavioral information J2). The statistical processing unit 412 may generate all or some of the multiple types of statistical indicators.

[0299] As an example, the condition processing unit 413 of the preprocessing unit 41 may perform processing (condition processing) according to specific conditions for each of the vital raw data and behavioral raw data of the biological state raw data A2 for each unit period UT.

[0300] Specifically, the condition processing unit 413 performs the process of extracting vital data from the subject's raw vital data for a period indicated by a specific condition, and / or the process of extracting behavioral data from the subject's raw behavioral data for a period indicated by a specific condition. This will be explained in detail below.

[0301] First, let's explain the processing of heart rate data from the vital data. The condition processing unit 413 extracts heart rate data for a period indicated by specific conditions from the heart rate data for a unit period UT. The specific conditions include at least one of the following conditions: conditions related to sleep, conditions related to wakefulness, and conditions related to the time of day caused by the sun.

[0302] For example, the condition processing unit 413 extracts heart rate data during wakefulness, during sleep, during the day, at night, at a predetermined time before falling asleep, at a predetermined time after falling asleep, at a predetermined time before waking up, and at a predetermined time after waking up from the heart rate data for a unit period UT. Hereinafter, the heart rate data extracted according to specific conditions may be referred to as "extracted heart rate data". Note that not all of these are required to be extracted.

[0303] The condition processing unit 413 passes multiple types of extracted heart rate data for a unit period UT to the statistical processing unit 412. The statistical processing unit 412 performs multiple types of statistical processing on each extracted heart rate data and outputs multiple statistical indicators for each extracted heart rate data. For example, the statistical processing unit 412 calculates multiple different statistical indicators from the same extracted heart rate data by performing multiple types of statistical processing on the same extracted heart rate data. Therefore, it is possible to extract features of the extracted heart rate data from the same extracted heart rate data from different perspectives. This contributes to improving the estimation accuracy of the biological state information M2 by the learning model TM1.

[0304] Furthermore, the statistical processing unit 412 may calculate the difference (hereinafter referred to as the day-night difference) between the statistical indicator obtained by statistically processing the "daytime heart rate data" and the statistical indicator obtained by statistically processing the "nighttime heart rate data" for each different statistical indicator (difference processing).

[0305] As described above, multiple types of statistical indicators and multiple types of diurnal differences are calculated for each of the multiple types of extracted heart rate data for each unit period UT, and constitute a part of the input information K2. In this case, the statistical indicators and diurnal differences are sometimes collectively referred to as heart rate statistical indicators. The storage unit 402 stores the multiple heart rate statistical indicators as part of the input information K2.

[0306] The heart rate statistical index is an example of vital information H2 from the input information K2. The extracted heart rate data is an example of "extracted raw vital data." Therefore, the statistical processing unit 412 generates vital information H2 by performing statistical processing on the extracted raw vital data.

[0307] The statistical processing unit 412 and the condition processing unit 413 may generate all or some of the multiple types of heart rate statistical indices.

[0308] Next, we will explain the processing of step count data from the raw behavioral data. The condition processing unit 413 extracts step count data for a period indicated by a specific condition from the step count data for a unit period UT. The specific condition in this case is the same as the specific condition for heart rate data.

[0309] For example, the condition processing unit 413 extracts step count data from step count data for a unit period UT, including step count data during wakefulness, step count data during sleep, step count data during the day, step count data at night, step count data for a predetermined time before falling asleep, step count data for a predetermined time after falling asleep, step count data for a predetermined time before waking up, and step count data for a predetermined time after waking up. Hereinafter, step count data extracted according to specific conditions may be referred to as "extracted step count data." Note that not all of these are required to be extracted.

[0310] The condition processing unit 413 passes multiple types of extracted step count data for a unit period UT to the statistical processing unit 412. The statistical processing unit 412 performs multiple types of statistical processing on each of the extracted step count data and outputs multiple statistical indicators for each of the extracted step count data. For example, the statistical processing unit 412 calculates multiple different statistical indicators from the same extracted step count data by performing multiple types of statistical processing on the same extracted step count data. Therefore, it is possible to extract features of the extracted step count data from different perspectives on the same extracted step count data. This contributes to improving the estimation accuracy of the biological state information M2 by the learning model TM1.

[0311] Furthermore, the statistical processing unit 412 may calculate the difference (hereinafter referred to as the day-night difference) between the statistical indicator obtained by statistically processing the "daytime step count data" and the statistical indicator obtained by statistically processing the "nighttime step count data" for each different statistical indicator (difference processing).

[0312] Furthermore, the statistical processing unit 412 may calculate the ratio of the period in which the number of steps is zero to the period indicated by specific conditions (hereinafter referred to as the occurrence ratio) (ratio processing).

[0313] For example, the statistical processing unit 412 calculates the percentage of occurrences during wakefulness, during sleep, during the day, at night, a predetermined time before falling asleep, a predetermined time after falling asleep, a predetermined time before waking up, and a predetermined time after waking up.

[0314] As described above, multiple types of statistical indicators, multiple types of day-night differences, and multiple types of occurrence rates are calculated for each of the multiple types of extracted step count data for each unit period UT, and constitute a part of the input information K2. In this case, the statistical indicators, day-night differences, and occurrence rates are sometimes collectively referred to as step count statistical indicators. The storage unit 402 stores the multiple step count statistical indicators as part of the input information K2.

[0315] The step count statistics index is an example of behavioral information J2 from input information K2. The extracted step count data is also an example of "extracted raw behavioral data." Therefore, the statistical processing unit 412 generates behavioral information J2 by ​​performing statistical processing on the extracted raw behavioral data.

[0316] The statistical processing unit 412 and the condition processing unit 413 may generate all or some of the multiple types of step count statistical indicators.

[0317] Next, we will explain the processing of exercise intensity data from the raw behavioral data. The condition processing unit 413 extracts exercise intensity data for a period indicated by a specific condition from the exercise intensity data for a unit period UT. The specific condition in this case is the same as the specific condition for heart rate data.

[0318] For example, the condition processing unit 413 extracts exercise intensity data from the exercise intensity data for a unit period UT, including exercise intensity data during wakefulness, exercise intensity data during sleep, daytime exercise intensity data, nighttime exercise intensity data, exercise intensity data for a predetermined time before falling asleep, exercise intensity data for a predetermined time after falling asleep, exercise intensity data for a predetermined time before waking up, and exercise intensity data for a predetermined time after waking up. Hereinafter, the exercise intensity data extracted according to specific conditions may be referred to as "extracted exercise intensity data." Note that not all of these data are required to be extracted.

[0319] The condition processing unit 413 passes multiple types of extracted exercise intensity data for a unit period UT to the statistical processing unit 412. The statistical processing unit 412 performs multiple types of statistical processing for each extracted exercise intensity data and outputs multiple statistical indicators for each extracted exercise intensity data. For example, the statistical processing unit 412 calculates multiple different statistical indicators from the same extracted exercise intensity data by performing multiple types of statistical processing on the same extracted exercise intensity data. Therefore, it is possible to extract features of the extracted exercise intensity data from different perspectives for the same extracted exercise intensity data. This contributes to improving the estimation accuracy of the biological state information M2 by the learning model TM1.

[0320] Furthermore, the statistical processing unit 412 may calculate the difference (hereinafter referred to as the day-night difference) between the statistical indicator obtained by statistically processing the "daytime exercise intensity data" and the statistical indicator obtained by statistically processing the "nighttime exercise intensity data" for each different statistical indicator (difference processing).

[0321] Furthermore, the statistical processing unit 412 may calculate the ratio of the period in which METs is less than or equal to a first specified value (hereinafter referred to as the first occurrence ratio) to the period indicated by the specific conditions (ratio processing). The condition processing unit 413 may calculate the ratio of the period in which METs is greater than or equal to a second specified value (hereinafter referred to as the second occurrence ratio) to the period indicated by the specific conditions (ratio processing).

[0322] For example, the statistical processing unit 412 calculates the first incidence rate and the second incidence rate for each of the following periods: while awake, while asleep, during the day, at night, a predetermined time before falling asleep, a predetermined time after falling asleep, a predetermined time before waking up, and a predetermined time after waking up.

[0323] As described above, multiple types of statistical indicators, multiple types of diurnal differences, and multiple types of occurrence rates are calculated for each of the multiple types of extracted exercise intensity data for each unit period UT, and constitute a part of the input information K2. In this case, the statistical indicators, diurnal differences, and occurrence rates are sometimes collectively referred to as exercise intensity statistical indicators. The storage unit 402 stores the multiple exercise intensity statistical indicators as part of the input information K2.

[0324] The exercise intensity statistical index is an example of behavioral information J2 from the input information K2. Furthermore, the extracted exercise intensity data is an example of "extracted raw behavioral data." Therefore, the statistical processing unit 412 generates behavioral information J2 by ​​performing statistical processing on the extracted raw behavioral data.

[0325] The statistical processing unit 412 and the condition processing unit 413 may generate all or some of the multiple types of exercise intensity statistical indices.

[0326] As described above, when calculating heart rate statistics, step count statistics, and exercise intensity statistics, statistical processing by the statistical processing unit 412 may be performed after processing by the condition processing unit 413. Therefore, according to Embodiment 4, statistical processing can be performed during a period that accurately reflects the physical and mental state and the environment of the living organism. As a result, the estimation accuracy of the biological state information M2 by the learning model TM1 can be further improved.

[0327] As described above, the statistical processing unit 412 may generate vital information H2 by performing statistical processing on the raw vital data extracted by the condition processing unit 413, and / or generate behavioral information J2 by ​​performing statistical processing on the raw behavioral data extracted by the condition processing unit 413.

[0328] Furthermore, the alcohol consumption index calculation unit 411 may receive, for example, extracted heart rate data and extracted step count data, or extracted heart rate data and extracted exercise intensity data from the condition processing unit 413. The alcohol consumption index calculation unit 411 may then calculate alcohol consumption estimation index information D2 based on the extracted heart rate data and extracted step count data, or extracted heart rate data and extracted exercise intensity data. In this case, for example, the period indicated by the specific conditions corresponds to the fifth time Tf when calculating the second alcohol consumption estimation index F(d), and the alcohol consumption index calculation unit 411 calculates the second alcohol consumption estimation index F(d).

[0329] As explained above with reference to Figures 21 to 25, the alcohol consumption estimation index information D2, heart rate statistics, step count statistics, and exercise intensity statistics constitute the input information K2. Hereinafter, the alcohol consumption estimation index information D2, heart rate statistics, step count statistics, exercise intensity statistics, environmental information R2, self-reported information V2, and attribute information Q2 may be collectively referred to as "features." The number of features is not particularly limited. Note that "features" may also be referred to as "feature FT."

[0330] (Estimation Unit 42) Referring to Figure 25, the estimation unit 42 uses the learning model TM1 to estimate the subject's biological state information M2. Specifically, the estimation unit 42 inputs input information K2 to the learning model TM1 and obtains output information L2 from the learning model TM1. The storage unit 402 stores the output information L2. For example, the learning model TM1 takes past input information K2 as input to estimate past biological state information M2. Alternatively, for example, the learning model TM1 takes current (latest) input information K2 as input to estimate current (latest) biological state information M2. Alternatively, for example, the learning model TM1 may take current (latest) input information K2 as input to estimate (predict) future biological state information M2.

[0331] As an example, the estimation unit 42 inputs the alcohol consumption estimation index information D2, heart rate statistical index, step count statistical index, and exercise intensity statistical index obtained during the unit period UT and / or the period indicated by specific conditions, as input information K2 to the learning model TM1.

[0332] The learning model TM1 takes input information K2 as input and outputs output information L2 according to a machine learning algorithm. The machine learning algorithm is not particularly limited, but examples include linear regression, Naive Bayes, Support Vector Machine, neural network, deep neural network (hereinafter referred to as DNN), decision tree, random forest, gradient boosting, or regularized regression. Regularized regression is, for example, L1 regularized regression or L2 regularized regression. Below, as an example, we will explain the case where the machine learning algorithm of the learning model TM1 is DNN.

[0333] (Deep Neural Network 50) Figure 26 is a schematic diagram showing an example of a DNN 50. As shown in Figure 26, the DNN 50 includes an input layer 51, a plurality of hidden layers 52, and an output layer 53. Figure 26 shows a fully connected example. The input layer 51 includes at least one node 511. In this case, the node 511 is input with index information D2 for alcohol consumption estimation. Each of the hidden layers 52 includes a plurality of nodes 521. The output layer 53 includes at least one node 531. In the example in Figure 26, the input layer 51 includes a plurality of nodes 511, and the output layer 53 includes a plurality of nodes 531.

[0334] Multiple nodes 511 of the input layer 51 are each input to multiple feature quantities FT that constitute the input information K2. Each node 521 of the hidden layer 52 uses the learned weights and biases and the activation function to convert the output of the previous layer into input to the next layer. Each node 531 of the output layer 53 uses the learned weights and biases and the activation function corresponding to the final output format to output the final result (estimation result) based on the output of the previous layer. In other words, the output layer 53 outputs output information L2. The output information L2 includes multiple biologically related state information M2 of different types. If the output layer 53 includes one node 531, the output information L2 includes one biologically related state information M2.

[0335] In the example shown in Figure 26, the state of mind and body is directly represented numerically by the biological state information M2 (direct numerical output). The number of nodes 531 is equal to the number of biological state information M2 to be output, and may be one or two or more.

[0336] On the other hand, the post-processing unit 43 in Figure 25 may output biological-related state information M3 by performing post-processing on the biological-related state information M2. This point will be explained with reference to Figures 27 and 28.

[0337] Figure 27 schematically shows another example of DNN50. In the example in Figure 27, the state of mind and body is indicated by binary classification using biological state information M3 (binary classification output).

[0338] Let's explain the first example of binary classification. In this first example, each node 531 directly outputs a numerical value as biological state information M2. In this case, the post-processing unit 43 performs binary classification by executing threshold processing on the numerical value indicated by the biological state information M2. The result of the binary classification is then output as biological state information M3.

[0339] A second example of binary classification will be explained. In this second example, the biological state information M2 output by each node 531 is, for example, a real value between 0 and 1 (hereinafter referred to as the "score value"). For example, if "0" indicates the first state and "1" indicates the second state, the score value indicates the probability that the state of mind, body, etc., is classified as the second state. In this case, the post-processing unit 43 performs probability calibration on the score value to calibrate it to a probability value with even higher reliability. Then, the post-processing unit 43 performs binary classification by performing threshold processing on the score value after probability calibration (hereinafter referred to as the "calibrated score value"). The result of the binary classification is then output as biological state information M3.

[0340] Furthermore, the number of nodes 531 is equal to the number of biological state information M2 to be output, and may be one or two or more. Also, the number of nodes 511 may be one or two or more.

[0341] Figure 28 schematically illustrates yet another example of DNN50. In the example in Figure 28, the state of mind, body, etc., is represented by multi-class classification using biological state information M3 (multi-class classification output). The state of mind, body, etc., is classified into U classes. U represents an integer of 3 or more. The number of classifications (number of classes) U is not particularly limited and can be set arbitrarily.

[0342] As shown in Figure 28, the DNN 50 includes an input layer 51, a plurality of intermediate layers 52, and at least one output layer 53A. In the example in Figure 28, the DNN 50 includes a plurality of output layers 53A.

[0343] Multiple output layers 53A are provided, each corresponding to multiple different types of biological state information M2. Each output layer 53A outputs output information L2. Specifically, each output layer 53A includes the same number of nodes 531A as the number of classifications (number of classes) U.

[0344] Let's focus on one output layer 53A. Each of the U nodes 531A corresponds to one of the U classes. Each node 531A outputs an output value VL. The output value VL is, for example, a real number between 0 and 1. The sum of the output values ​​VL of the U nodes 531A is "1". Therefore, the output value VL represents the probability value of being classified into the corresponding class. In other words, the class corresponding to the node 531A that outputs the largest output value VL has the highest probability of being correct. However, the post-processing unit 43 performs probability calibration on each output value VL, calibrating each output value VL to a probability value with even higher reliability. Then, the post-processing unit 43 obtains the largest output value VL among the U output values ​​VL after calibration by probability calibration. The largest output value VL corresponds to the biological state information M2. In this way, the U output values ​​VL (output information L2) from the output layer 53A substantially include the biological state information M2. Then, the post-processing unit 43 sets the class corresponding to node 531A, which output the largest output value VL after calibration, in the biological state information M3. In this manner, the post-processing unit 43 performs multi-class classification.

[0345] The post-processing unit 43 performs multi-class classification for each of the multiple output layers 53A. The number of output layers 53A is equal to the number of biological state information M2 to be output, and may be one or two or more. If there is one output layer 53A, one biological state information M2 and one biological state information M3 are output. The number of nodes 511 may be one or two or more.

[0346] (Post-processing unit 43) Returning to Figure 25, the interpretation unit 431 of the post-processing unit 43 calculates contribution information indicating the extent to which multiple feature quantities FT constituting the input information K2 contributed when the learning model TM1 estimated the biological-related state information M2. The contribution information includes the contribution of each feature quantity when estimating the biological-related state information M2. Specifically, the interpretation unit 431 calculates the contribution information based on the multiple feature quantities FT, the estimated biological-related state information M2, and the information of the learning model TM1. In this case, the interpretation unit 431 calculates the contribution information according to the model interpretation method. Examples of model interpretation methods include SHAP (Shapley Additive Explanations), ICE (Individual Conditional Expectation), LIME (Local Interpretable Model-agnostic Explanations), or AIME (Approximate Inverse Model Explanations). The model interpretation method is not limited to these, as it is a method that outputs information about the indicators that contributed to the estimation result and the breakdown of the contributions. The memory unit 402 stores contribution information. The second database 46 also stores contribution information. The information provision server 47 transmits the contribution information stored in the second database 46, along with the biological condition information MX, to the second terminal 200 and / or the first terminal 102 via the network NW. By checking the contribution information, the user can understand what factors led to the biological condition information MX.

[0347] As described above with reference to Figures 20 to 28, according to Embodiment 4, the estimation device 4 can obtain highly reliable biological state information MX by inputting at least the alcohol consumption estimation index information D2 to the learning model TM1. This is because the learning model TM1 utilizes the alcohol consumption estimation index information D2, which can estimate alcohol consumption with high accuracy.

[0348] In particular, in Embodiment 4, the input information K2 does not need to include the results of diagnosis and evaluation by medical professionals, the results of examinations by medical devices such as medical imaging diagnostic equipment, or the results of collection and examination of bodily fluids such as blood. Therefore, in the stage of using the learning model TM1, the estimation device 4 can output the biological state information MX with high estimation accuracy without using the results of diagnosis and evaluation by medical professionals, the results of examinations by medical devices such as medical imaging diagnostic equipment, or the results of collection and examination of bodily fluids such as blood. Thus, the biological state information MX can be obtained while reducing the burden on the subject. In this way, the biological state information MX can be obtained without reducing the subject's quality of life (QOL).

[0349] In addition, the estimation device 4 can continuously acquire raw biological state data A2 from a wearable device (biological state detection device 103) and / or a mobile terminal (first terminal 102). Therefore, it can continuously calculate and monitor biological state information MX. In particular, the raw biological state data A2 is automatically transmitted from the wearable device (biological state detection device 103) and / or the mobile terminal (first terminal 102) to the estimation system 40 (estimation device 4). As a result, the burden on the subject is further reduced. In other words, biological state information MX can be obtained while reducing the burden on the subject.

[0350] [Estimation process for biological state information M2] Next, the estimation process for biological state information M2, which is performed in step S53 of Figure 24, will be explained with reference to Figures 25 and 29. Figure 29 is a flowchart of an example of the estimation process. The estimation process is performed by the estimation device 4 in Figure 25. As shown in Figure 29, the estimation process includes steps S71 to S78.

[0351] First, in step S71, the condition processing unit 413 obtains the subject's biological state raw data A2 from the first database 45 (Figure 23). The biological state raw data A2 is stored in the storage unit 402.

[0352] Next, in step S72, the condition processing unit 413 performs processing (condition processing) according to specific conditions on the heart rate data, step count data, and exercise intensity data of the subject's biological state raw data A2 stored in the memory unit 402. As a result, the condition processing unit 413 outputs extracted heart rate data, extracted step count data, and extracted exercise intensity data. The specific conditions are, for example, conditions related to sleep, conditions related to wakefulness, and conditions related to the time of day caused by the sun.

[0353] Next, in step S73, the statistical processing unit 412 performs statistical processing and differential processing on each of the extracted heart rate data, extracted step count data, and extracted exercise intensity data stored in the storage unit 402. In addition, the statistical processing unit 412 performs ratio processing on each of the extracted step count data and extracted exercise intensity data stored in the storage unit 402. As a result, the statistical processing unit 412 outputs a heart rate statistical index based on the extracted heart rate data, a step count statistical index based on the extracted step count data, and an exercise intensity statistical index based on the extracted exercise intensity data. These indices are stored in the storage unit 402 as input information K2. Note that differential processing and ratio processing can be considered as statistical processing in a broad sense.

[0354] Furthermore, the statistical processing unit 412 calculates heart rate statistical indicators, step count statistical indicators, and exercise intensity statistical indicators by performing statistical processing on the heart rate data, step count data, and exercise intensity data for the unit period UT stored in the memory unit 402. These statistical indicators are stored in the memory unit 402 as input information K2.

[0355] The statistical processing unit 412 may perform either statistical processing on the data processed according to the specific conditions in step S72 (data for the period indicated by the specific conditions) or statistical processing on the data for the unit period UT, or it may perform both.

[0356] Next, in step S74, the alcohol consumption index calculation unit 411 calculates alcohol consumption estimation index information D2 based on the extracted heart rate data and extracted step count data, or extracted heart rate data and extracted exercise intensity data, stored in the memory unit 402 (alcohol consumption index calculation process). The alcohol consumption estimation index information D2 is stored in the memory unit 402 as input information K2.

[0357] Furthermore, the alcohol consumption index calculation unit 411 calculates alcohol consumption estimation index information D2 based on the heart rate data and step count data for a unit period UT, or the heart rate data and exercise intensity data for a unit period UT, which are stored in the memory unit 402 (alcohol consumption index calculation process). The alcohol consumption estimation index information D2 is stored in the memory unit 402 as input information K2.

[0358] The alcohol consumption index calculation unit 411 may perform either the alcohol consumption index calculation process based on data processed according to the specific conditions in step S72 (data for the period indicated by the specific conditions), or the alcohol consumption index calculation process based on data for the unit period UT, or it may perform both.

[0359] Next, in step S75, the estimation unit 42 inputs input information K2, which includes the features generated in steps S73 and S74, to the learning model TM1. As a result, the learning model TM1 outputs output information L2.

[0360] Next, in step S76, the estimation unit 42 acquires output information L2 from the learning model TM1. The output information L2 includes the subject's biological state information M2. The output information L2 is stored in the storage unit 402.

[0361] Next, in step S77, the estimation unit 42 determines whether or not post-processing is required for the biological state information M2.

[0362] If it is determined in step S77 that no post-processing is required (NO), the estimation process is completed and the process returns to the main routine shown in Figure 24. Cases where no post-processing is required include, for example, when the biological status information M2 is shown by direct numerical output (for example, Figure 26), or when contribution information is not calculated.

[0363] On the other hand, if it is determined in step S77 that post-processing is necessary (YES), the process proceeds to step S78. Post-processing is necessary, for example, when performing binary classification output or multi-class classification output (for example, Figures 27 and 28), or when calculating contribution information.

[0364] Next, in step S78, the post-processing unit 43 performs post-processing on the biological-related state information M2 and outputs the biological-related state information M3. The biological-related state information M3 is stored in the storage unit 402 as output information N2. Preferably, the interpretation unit 431 calculates contribution information when the learning model TM1 estimates the biological-related state information M2. The contribution information is stored in the storage unit 402 as, for example, output information L2 or output information N2. Then the estimation process is completed and the process returns to the main routine in Figure 24.

[0365] As described above with reference to Figure 29, according to Embodiment 4, the estimation device 4 can obtain output information L2 (biologically related state information M2) with high estimation accuracy from the learning model TM1 by inputting input information K2 to the learning model TM1. Furthermore, the estimation device 4 can obtain output information N2 (biologically related state information M3) by performing post-processing on the output information L2.

[0366] In this case, the input information K2 only needs to include at least the subject's alcohol consumption estimation index information D2. Therefore, the input information K2 does not need to include the results of diagnosis and evaluation by medical professionals, the results of examinations using medical devices such as medical imaging devices, or the results of collecting and examining bodily fluids such as blood. Thus, according to Embodiment 4, by using the learning model TM1, it is possible to estimate the biological state information MX with high estimation accuracy while improving the subject's QOL.

[0367] [Learning Device 3] Next, the generation stage of the learning model TM1 will be described with reference to Figures 20, 30, and 31. Figure 30 is a block diagram showing an example configuration of the learning device 3 according to Embodiment 4. As shown in Figure 30, the learning device 3 comprises a processing unit 31, a communication unit 34, and a storage unit 35. The learning device 3 may also include an input unit 32 and an output unit 33. The hardware configuration of the processing unit 31, communication unit 34, storage unit 35, input unit 32, and output unit 33 is the same as the hardware configuration of the processing unit 400, communication unit 401, storage unit 402, input unit 403, and output unit 404 of the estimation device 4 in Figure 25, or the hardware configuration of the drinking estimation device IS in Figure 16 or Figure 17.

[0368] The storage unit 35 stores data and computer programs. The processing unit 31 includes a learning data acquisition unit 310 and a learning unit 311. For example, the processor of the processing unit 31 functions as the learning data acquisition unit 310 and the learning unit 311 by executing the computer programs stored in the storage device of the storage unit 35.

[0369] Figure 31 is a flowchart showing an example of a learning method using the learning device 3. The learning method is an example of the "learning model generation method" of this disclosure. As shown in Figure 31, the learning method includes steps S101 to S108.

[0370] First, in step S101, the learning data acquisition unit 310 acquires multiple learning datasets F1 from the learning data creation device 2 (Figure 20). Specifically, the learning data acquisition unit 310 acquires multiple learning datasets F1 from the learning database DBT (Figure 32). The storage unit 35 stores the multiple learning datasets F1. A portion of the multiple learning datasets F1 is training data, another portion is evaluation data, and yet another portion is test data.

[0371] Next, in step S102, the learning unit 311 prepares the pre-training learning model TM1. In the pre-training learning model TM1, various parameters are set to their initial values.

[0372] Next, in step S103, the learning unit 311 obtains one learning dataset F1 from the multiple learning datasets F1 stored in the storage unit 35. In this case, the learning dataset F1 is the training data.

[0373] Next, in step S104, the learning unit 311 inputs the feature information G1 contained in the learning dataset F1 to the learning model TM1 before (or during) learning. As a result, output information is output from the learning model TM1 as an estimation result according to the machine learning algorithm.

[0374] Next, in step S105, the learning unit 311 compares the ground truth labels B1 included in the learning dataset F1 with the output information output as estimation results in step S104, and performs machine learning by adjusting various parameters based on the machine learning algorithm and predetermined adjustment method. In this way, the learning unit 311 allows the learning model TM1 to learn the correlation between feature information G1 and ground truth labels B1. The predetermined adjustment method for various parameters is not particularly limited, but examples include the least squares method, maximum likelihood estimation method, EM algorithm, gradient descent method, backpropagation method, or Bayesian estimation method.

[0375] Next, in step S106, the learning unit 311 determines whether the learning termination condition has been met. The learning termination condition is, for example, that the evaluation value of the loss function based on the correct label B1 and the output information output as an estimation result has reached the target value. Alternatively, the learning termination condition is, for example, that the number of learning iterations (epochs) has reached the target number.

[0376] If it is determined in step S106 that the learning termination condition is not met (NO), the process proceeds to step S103. Steps S103 to S105 are repeated until the learning termination condition is met.

[0377] On the other hand, if it is determined in step S106 that the learning termination condition has been met (YES), the process proceeds to step S107.

[0378] Next, in step S107, the learning unit 311 adjusts the hyperparameters of the learning model TM1 based on the learning dataset F1 as evaluation data and the input values ​​input from the input unit 32 by the machine learning engineer. The adjustment method in this case is not particularly limited, but examples include grid search, random search, or Bayesian optimization.

[0379] Next, in step S108, the learning unit 311 evaluates the estimation accuracy of the learning model TM1 using the learning dataset F1 as test data. Then the learning method is completed.

[0380] As described above with reference to Figure 31, according to Embodiment 4, the learning device 3 generates a learning model TM1 that outputs output information L2 when input information K2 is input by performing learning using the learning dataset F1. That is, the learning device 3 generates a trained learning model TM1 having various trained parameters by repeatedly performing learning using a plurality of learning datasets F1. The storage unit 35 stores the trained learning model TM1.

[0381] [Learning Data Creation Device 2] Next, the generation stage of the learning dataset F1 will be described with reference to Figures 28, 32, and 33. Figure 32 is a block diagram showing an example configuration of the learning data creation device 2 according to Embodiment 4. As shown in Figure 32, the learning data creation device 2 comprises a processing unit 21, a communication unit 24, and a storage unit 25. The learning data creation device 2 may also include an input unit 22 and an output unit 23. The hardware configuration of the processing unit 21, communication unit 24, storage unit 25, input unit 22, and output unit 23 is the same as the hardware configuration of the processing unit 400, communication unit 401, storage unit 402, input unit 403, and output unit 404 of the estimation device 4 in Figure 6, or the hardware configuration of the drinking estimation device IS in Figure 16 or Figure 17.

[0382] The storage unit 25 stores data and computer programs. The storage unit 25 includes a learning database DBT. In Figure 32, the database is abbreviated as DB. The learning database DBT stores multiple learning datasets F1.

[0383] The processing unit 21 includes an alcohol consumption index calculation unit 211 and a correct label creation unit 216. The processing unit 21 may also include one or more of a statistical processing unit 212 and a condition processing unit 213. For example, the processor of the processing unit 400 functions as the alcohol consumption index calculation unit 211, the correct label creation unit 216, the statistical processing unit 212, and the condition processing unit 213 by executing a computer program stored in the storage device of the storage unit 402.

[0384] The processing of the alcohol consumption index calculation unit 211, the statistical processing unit 212, and the condition processing unit 213 is the same as the processing of the alcohol consumption index calculation unit 411, the statistical processing unit 412, and the condition processing unit 413 of the estimation device 4 (Figure 25), respectively.

[0385] For example, in the description of the alcohol consumption index calculation unit 411, statistical processing unit 412, and condition processing unit 413 of the estimation device 4, the subject can be read as the learning subject, the biological state raw data A2 as the biological state raw data A1, the input information K2 as the feature information G1, the output information L2 as the correct label B1, the alcohol consumption estimation index information D2 as the alcohol consumption estimation index information D1, the vital information H2 as the vital information H1, the behavior information J2 as the behavior information J1, the environmental information R2 as the environmental information R1, the self-reported information V2 as the reported information V1, and the biological-related state information M2 as the biological-related state information M1, thereby substituting for the description of the alcohol consumption index calculation unit 211, statistical processing unit 212, and condition processing unit 213 of the learning data creation device 2.

[0386] [Method for Creating Learning Data] Figure 33 is a flowchart showing an example of a method for creating learning data using the learning data creation device 2. As shown in Figure 33, the method for creating learning data includes steps S201 to S205.

[0387] First, in step S201, the processing unit 21 acquires the raw biological state data A1 and the correct answer information Z1 of the learning subject. The raw biological state data A1 and the correct answer information Z1 are stored in the storage unit 25.

[0388] Next, in step S202, the condition processing unit 213 performs processing (condition processing) according to specific conditions on the heart rate data, step count data, and exercise intensity data of the learning subject's biological state raw data A1. As a result, the condition processing unit 213 outputs extracted heart rate data, extracted step count data, and extracted exercise intensity data. The specific conditions are, for example, conditions related to sleep, conditions related to wakefulness, and conditions related to the time of day caused by the sun.

[0389] Next, in step S203, the statistical processing unit 210 performs statistical processing and differential processing on each of the extracted heart rate data, extracted step count data, and extracted exercise intensity data. In addition, the statistical processing unit 210 performs ratio processing on each of the extracted step count data and extracted exercise intensity data. As a result, the statistical processing unit 210 outputs a heart rate statistical index based on the extracted heart rate data, a step count statistical index based on the extracted step count data, and an exercise intensity statistical index based on the extracted exercise intensity data. These indices are stored in the storage unit 25 as feature information G1. Note that differential processing and ratio processing can be considered as statistical processing in a broad sense.

[0390] Furthermore, the statistical processing unit 210 calculates heart rate statistical indices, step count statistical indices, and exercise intensity statistical indices by performing statistical processing on heart rate data, step count data, and exercise intensity data for each of the unit periods UT. These indices are stored in the storage unit 25 as feature information G1.

[0391] The statistical processing unit 210 may perform either statistical processing on the data processed according to the specific conditions in step S202 (data for the period indicated by the specific conditions) or statistical processing on the data for the unit period UT, or it may perform both.

[0392] Next, in step S204, the alcohol consumption index calculation unit 211 calculates alcohol consumption estimation index information D1 based on extracted heart rate data and extracted step count data, or extracted heart rate data and extracted exercise intensity data (alcohol consumption index calculation process). The alcohol consumption estimation index information D1 is stored in the storage unit 25 as feature information G1.

[0393] Furthermore, the alcohol consumption index calculation unit 211 calculates alcohol consumption estimation index information D2 based on heart rate data and step count data for a unit period UT, or heart rate data and exercise intensity data for a unit period UT (alcohol consumption index calculation process). The alcohol consumption estimation index information D1 is stored in the storage unit 25 as feature information G1.

[0394] The alcohol consumption index calculation unit 211 may perform either the alcohol consumption index calculation process based on data processed according to the specific conditions in step S202 (data for the period indicated by the specific conditions), or the alcohol consumption index calculation process based on data for the unit period UT, or it may perform both.

[0395] Next, in step S205, the correct label creation unit 216 creates a correct label B1 based on the correct information Z1 and associates it with the feature information G1 (correct label creation process). The correct label B1 includes biological state information M1. The correct label B1 is stored in the storage unit 25. As an example, the correct information Z1 is a numerical value that directly indicates the state of mind, body, etc. Based on this example, the first to third examples will be explained.

[0396] As a first example, when the estimation device 4 directly performs numerical output (for example, in the case of Figure 26), the correct label creation unit 216 associates the correct information Z1 with the feature information G1 as the correct label B1 (biologically related state information M1).

[0397] As a second example, when the estimation device 4 performs binary classification output (for example, in the case of Figure 27), the correct label creation unit 216 compares the correct information Z1 with a threshold and, based on the comparison result, classifies the correct information Z1 (state of mind, body, or environment) into a first state or a second state. Then, according to the classification result, the correct label creation unit 216 sets information indicating the first state (for example, 0) or information indicating the second state (for example, 1) as the correct label B1 (biologically related state information M1).

[0398] As a third example, when the estimation device 4 performs multi-class classification (for example, in the case of Figure 28), the correct label creation unit 216 sets an evaluation corresponding to the class for the correct label B1 (biological state information M1) depending on whether or not the correct information Z1 falls within the numerical range of any of the multiple classes.

[0399] In the second and third examples, the correct answer information Z1 after binary classification or multi-class classification may be input via the input unit 22. In this case, the correct label creation unit 216 associates the correct answer information Z1 with the feature information G1 as the correct label B1, similar to the first example.

[0400] When step S205 is completed, the training data creation method is finished. Steps S201 to S205 are repeatedly executed to create multiple training datasets F1. The training data creation device 2 stores the training datasets F1 in the training database DBT. The training data creation device 2 may include one or more of the following information in the training datasets F1: environmental information R1, report information V1, and attribute information Q1.

[0401] As described above with reference to Figure 33, according to Embodiment 4, the learning data creation device 2 can create a learning dataset F1 suitable for a learning model TM1 that can estimate biological state information M2 while improving the quality of life of the subject.

[0402] Here, it is preferable that the multiple training datasets F1 to be trained on the learning model TM1 consist of "information representing the mental and physical states of multiple learning subjects belonging to the same group" or "information representing the environmental state that affects the mental and physical states of multiple learning subjects belonging to the same group."

[0403] Here, it is preferable that the multiple training datasets F1 to be trained on the learning model TM1 consist of feature information G1 and correct labels B1 of multiple learning subjects belonging to the same group. A group is a group of learning subjects that satisfy predetermined conditions regarding the state of mind, body, or environment. The predetermined conditions are, for example, that there has been a past occurrence of an abnormality in the state of mind or body, such as the onset of a disease or symptoms, or that there has been a past occurrence of staying in an environment with an abnormal atmosphere.

[0404] Thus, it is preferable that the learning subjects when acquiring the learning dataset F1 belong to the same group. Furthermore, it is preferable that the subjects who acquire the input information K2 to be input into the learning model TM1 belong to the same group as the learning subjects. In other words, it is preferable that the subjects, like the learning subjects, satisfy predetermined conditions regarding the state of mind, body, or environment. According to this preferred example, for example, it is possible to explain to the subjects the validity of estimating the bio-related state information M2 using the learning model TM1, to explain the reliability of the estimation results, and to improve the level of acceptance of the estimation results. In addition, by acquiring the learning dataset F1 from learning subjects belonging to the same group as the subjects, the estimation accuracy of the bio-related state information M2 can be improved. <Modification> In the modification of Embodiment 4, the learning dataset F1 includes drinking information D100 calculated based on the drinking estimation index information D1, instead of the drinking estimation index information D1. Also, the input information K2 includes drinking information D200 calculated based on the drinking estimation index information D2, instead of the drinking estimation index information D2. The alcohol consumption information D100 and D200 include information indicating whether or not alcohol is consumed, whether or not there is a drinking habit, information indicating the frequency of drinking (e.g., daily drinking, drinking 2-3 times a week, etc.), or information indicating the amount of alcohol consumed (e.g., total amount of alcohol consumed over a certain period). In the modified example, the biological state information M1 does not include the alcohol consumption information M130, and the biological state information M2 does not include the alcohol consumption information M230.

[0405] In the modified example, the input information K2 only needs to include at least the alcohol consumption information D200. Also, the feature information G1 only needs to include at least the alcohol consumption information D100.

[0406] In the modified example, the learning model TM1 learns drinking information D100 based on drinking estimation index information D1, which can estimate drinking with high accuracy, and drinking information D200 based on drinking estimation index information D2, which can estimate drinking with high accuracy, is input to the learning model TM1. Therefore, the learning model TM1 can estimate the biological state information M2 with high accuracy.

[0407] Embodiment 4 (including modifications) of the present disclosure has been described above. In Embodiment 4 (including modifications), rheumatoid arthritis index information, blood index information, systemic connective tissue disorder index information, and non-organ-specific systemic autoimmune disease index information may be excluded from the biological state information M1 and MX. In other words, the biological state information M1 and MX do not need to include rheumatoid arthritis index information, blood index information, systemic connective tissue disorder index information, and non-organ-specific systemic autoimmune disease index information. Rheumatoid arthritis index information is an index representing the symptoms or activity of rheumatoid arthritis, or information relating to an index representing the symptoms or activity of rheumatoid arthritis. Blood index information is information that directly or indirectly indicates blood indexes. Blood indexes are indices that quantitatively indicate the components of blood, the state of the components of blood, substances in the blood, or the state of substances in the blood. Systemic connective tissue disorder index information refers to information on indicators that directly or indirectly represent the symptoms or activity of systemic connective tissue disorder, or information on indicators that directly or indirectly represent the effects resulting from the symptoms or activity of systemic connective tissue disorder. Non-organ-specific systemic autoimmune disease index information refers to information on indicators that directly or indirectly represent the symptoms or activity of non-organ-specific systemic autoimmune disease, or information on indicators that directly or indirectly represent the effects resulting from the symptoms or activity of non-organ-specific systemic autoimmune disease. Systemic connective tissue disorder and non-organ-specific systemic autoimmune disease include systemic lupus erythematosus.

[0408] Preferred embodiments and modifications of the present disclosure have been described in detail above with reference to the attached drawings, but the technical scope of the present disclosure is not limited to such examples. It is clear to any person with ordinary skill in the art of the present disclosure that various modifications or alterations can be conceived within the scope of the technical idea set forth in the claims, and these too are understood to fall within the technical scope of the present disclosure.

[0409] The apparatus or system described herein may be implemented as a single apparatus, or as a group of apparatuses (e.g., a cloud server) partially or entirely connected by a network. For example, some or all of the alcohol consumption index calculation unit 411, statistical processing unit 412, condition processing unit 413, exercise responsiveness estimation unit 414, estimation unit 42, and post-processing unit 43 in Figure 25 may be implemented by the same computer or server. For example, the alcohol consumption index calculation unit 411, statistical processing unit 412, condition processing unit 413, estimation unit 42, and post-processing unit 43 in Figure 25 may each be implemented by separate computers or servers. These points also apply to the configurations of the learning apparatus 3 in Figure 30 and the learning data creation apparatus 2 in Figure 32. For example, the relay server 44 or estimation apparatus 4 in Figure 23 may have the functions of an information provision server 47. For example, the first database 45 and the second database 46 may be implemented by a single computer or server. For example, the estimation unit 42 may have the functions of a relay server 44.

[0410] The series of processes performed by the apparatus described herein may be implemented using software, hardware, or a combination of software and hardware. Computer programs for implementing each function of the processing units 400, 31, and 21 can be created and implemented on a PC or the like. A computer-readable storage medium on which such a computer program is stored can also be provided. Examples of storage media include magnetic disks, optical disks, magneto-optical disks, and flash memory. The above-mentioned computer programs may also be distributed without using a storage medium, for example, via a network. For example, the learning database DBT may be located outside the learning data creation device 2. For example, the learning model TM1 may be located outside the estimation device 4 (for example, on an external server).

[0411] The processes described using flowcharts in this specification do not necessarily have to be executed in the order shown. Some processing steps may be executed in parallel. Additional processing steps may be adopted, and some processing steps may be omitted.

[0412] In Figures 5, 18, 24, 29, 31, and 33, the processing units SY(IS), 400, 31, and 21 execute the computer programs stored in the memory units KI(IS), 402, 35, and 25 to perform each step included in the method for calculating an alcohol consumption estimation index, the alcohol consumption estimation method, the information processing method, the estimation method, the learning method, or the learning data creation method. In Figure 19, the processing unit SY(SK) executes the computer programs stored in the memory unit KI(SK) to perform each step included in the biological examination method. In other words, the computer program causes the processing units SY(IS), 400, 31, and 21 to perform each step included in the method for calculating an alcohol consumption estimation index, the information processing method, the estimation method, the learning method, or the learning data creation method. The computer program causes the processing unit SY(SK) to perform each step included in the biological examination method. The processing units SY(IS), 400, 31, 21 and SY(SK) correspond to an example of the “computer” in this disclosure. In other words, when the computer program is executed by the processing units SY(IS), 400, 31, 21 or SY(SK), the computer program product implements each of the steps included in the method for calculating an alcohol consumption estimation index, the information processing method, the estimation method, the learning method, the learning data creation method, or the biological examination method.

[0413] In Figure 18, the alcohol consumption estimation process does not have to include either step S301 or step S302. In Figure 19, the biological examination process does not have to include either step S401 or step S402. The estimation device 4 in Figure 25 does not have to include, for example, all or part of the statistical processing unit 412 and the condition processing unit 413. The estimation process in Figure 29 does not have to include, for example, all or part of steps S72 and S73. Also, the estimation process in Figure 29 does not have to include, for example, steps S77 and S78. The learning data creation device 2 in Figure 32 does not have to include, for example, all or part of the statistical processing unit 212 and the condition processing unit 213. The learning data creation method in Figure 33 does not have to include, for example, all or part of steps S202 and S203.

[0414] The learning device 3 may generate a new learning model (distilled model) by performing learning using the input information K2 input to the learning model TM1 and the output information L2 or output information N2 output by the learning model TM1 as a learning dataset. For the learning model TM1 at the utilization stage, the input information K2 and output information L2 and N2 are information about the subject. However, in the generation of the distilled model, the input information K2 and output information L2 and N2 that constitute the learning dataset correspond to information about the learning subject for the learning device 3.

[0415] Furthermore, the effects described herein are merely descriptive or illustrative and not limiting. In other words, the technology relating to this disclosure may produce other effects that will be apparent to those skilled in the art from the description herein, in addition to or instead of the effects described herein.

[0416] <Examples of configurations> The following configurations also fall within the technical scope of this disclosure.

[0417] (Item 1) An information processing device comprising a processing unit that acquires a total heart rate change amount, which is the sum of the heart rate change amounts of a subject, wherein the processing unit calculates a first index for estimating alcohol consumption, which represents the total heart rate change amount when the subject is inactive and the subject's heart rate exceeds a reference heart rate.

[0418] (Item 2) The processing unit calculates the total heart rate change by summing the difference between the subject's heart rate in a first unit time and the subject's heart rate in a second unit time after the first unit time, which is calculated based on the subject's heart rate measured for each unit time, over a third time period longer than the unit time; the subject's inactivity is indicated by a first function that indicates whether the subject is active or not, which is defined based on the subject's physical activity information measured for each unit time; the state in which the subject's heart rate exceeds the reference heart rate is indicated by a second function that indicates whether the representative heart rate exceeds the reference heart rate in which it is presumed that the subject may be intoxicated; and the representative heart rate is a representative heart rate calculated over a fourth time period longer than the unit time, based on the subject's heart rate.

[0419] (Item 3) The information processing device described in Item 2, wherein the physical activity information indicates the number of steps taken by the subject, and the first function indicates whether or not the subject is walking.

[0420] (Item 4) The information processing device according to Item 1 or 2, wherein the processing unit calculates a second indicator for estimating alcohol consumption by performing statistical processing on the first indicator for estimating alcohol consumption within a fifth time period.

[0421] (Item 5) The statistical processing refers to the information processing device described in Item 4, which is a process for calculating a sum or a process for calculating an average.

[0422] (Item 6) The information processing device according to any one of Items 1 to 5, wherein the first indicator for estimating alcohol consumption is indicated by the total change in heart rate, which shows an increase in the heart rate of the subject.

[0423] (Item 7) An alcohol estimation device comprising at least one of the following: a first estimation unit that estimates whether or not a subject has consumed alcohol using a first indicator for alcohol estimation; and a second estimation unit that estimates when the subject consumed alcohol using the first indicator for alcohol estimation, wherein the first indicator for alcohol estimation represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

[0424] (Item 8) An alcohol estimation device comprising at least one of the following: a first estimation unit that estimates whether or not a subject has consumed alcohol using a second indicator for alcohol estimation; and a second estimation unit that estimates when the subject consumed alcohol using the second indicator for alcohol estimation, wherein the second indicator for alcohol estimation is an indicator obtained by performing statistical processing on the first indicator for alcohol estimation within a predetermined time period, the first indicator for alcohol estimation represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

[0425] (Item 9) A method for calculating an index for estimating alcohol consumption, comprising the steps of: obtaining a total heart rate change amount, which is the sum of the heart rate change amounts of a subject; and calculating a first index for estimating alcohol consumption, which represents the total heart rate change amount when the subject is inactive and the subject's heart rate exceeds a reference heart rate.

[0426] (Item 10) A method for estimating alcohol consumption, comprising at least one of the following steps: estimating whether or not a subject has consumed alcohol using a first indicator for estimating alcohol consumption; and estimating when the subject consumed alcohol using the first indicator for estimating alcohol consumption, wherein the first indicator for estimating alcohol consumption represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the changes in the subject's heart rate.

[0427] (Item 11) A method for estimating alcohol consumption, comprising at least one of the following steps: estimating whether or not a subject has consumed alcohol using a second indicator for estimating alcohol consumption; and estimating when the subject consumed alcohol using the second indicator for estimating alcohol consumption, wherein the second indicator for estimating alcohol consumption is an indicator obtained by performing statistical processing on a first indicator for estimating alcohol consumption within a predetermined time period; the first indicator for estimating alcohol consumption represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate; and the total change in heart rate represents the sum of the changes in heart rate, which are the changes in the subject's heart rate.

[0428] (Item 12) A biological examination method comprising: a step of extracting or excluding alcohol-related biological information related to a subject's alcohol consumption using at least one of a first indicator for alcohol consumption estimation and a second indicator for alcohol consumption estimation from the subject's biological information; and a step of examining the subject's biological state using the extracted alcohol-related biological information or the biological information other than the alcohol-related biological information, wherein the first indicator for alcohol consumption estimation represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate, the total change in heart rate represents the sum of the changes in the subject's heart rate, and the second indicator for alcohol consumption estimation is an indicator obtained by performing statistical processing on the first indicator for alcohol consumption estimation within a predetermined time.

[0429] (Item 13) A biological examination device comprising: an extraction / exclusion unit that extracts or excludes alcohol-related biological information related to a subject's alcohol consumption using at least one of a first indicator for alcohol consumption estimation and a second indicator for alcohol consumption estimation from the subject's biological information; and an examination unit that examines the subject's biological state using the extracted alcohol-related biological information or the biological information other than the alcohol-related biological information, wherein the first indicator for alcohol consumption estimation represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a standard heart rate, the total change in heart rate represents the sum of the heart rate changes which are the changes in the subject's heart rate, and the second indicator for alcohol consumption estimation is an indicator obtained by performing statistical processing on the first indicator for alcohol consumption estimation within a predetermined time.

[0430] (Item 14) Estimation device for estimating biological state information of a subject, comprising: an estimation unit that inputs input information to a learning model and obtains output information from the learning model; and a storage unit that stores the output information, wherein the output information includes the biological state information, the biological state information includes information representing the physical and mental state of the subject, information representing the state of the environment that affects the physical and mental state of the subject, information representing the state of the body affected by the physical and mental state of the subject, or information regarding the subject's alcohol consumption, the input information includes at least an index information for estimating alcohol consumption, the index information for estimating alcohol consumption includes information regarding the total change in heart rate in a state inactive of the subject and in a state in which the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

[0431] (Item 15) The total heart rate change is information calculated by summing up the heart rate change, which is the difference between the subject's heart rate in a first unit time and the subject's heart rate in a second unit time after the first unit time, calculated based on the subject's heart rate measured for each unit time, over a third time period longer than the unit time; the subject's inactivity is indicated by a first function that indicates whether the subject is active or not, defined based on the subject's physical activity information measured for each unit time; the state in which the subject's heart rate exceeds the reference heart rate is indicated by a second function that indicates whether the representative heart rate exceeds the reference heart rate in which it is presumed that the subject may be intoxicated; and the representative heart rate is a representative heart rate calculated over a fourth time period longer than the unit time, based on the subject's heart rate, as described in Item 14.

[0432] (Item 16) The estimation device described in Item 15, wherein the physical activity information indicates the number of steps taken by the subject, and the first function indicates whether or not the subject is walking.

[0433] (Item 17) An estimation device according to any one of Items 14 to 16, wherein the alcohol consumption estimation index information includes a first index for alcohol consumption estimation that shows the change in total heart rate when the subject is inactive and the subject's heart rate exceeds the reference heart rate.

[0434] (Item 18) The estimation device according to any one of Items 14 to 16, wherein the alcohol consumption estimation index information includes a second alcohol consumption estimation index obtained by performing statistical processing on a first alcohol consumption estimation index within a predetermined time, and the first alcohol consumption estimation index indicates the total change in heart rate when the subject is inactive and the subject's heart rate exceeds the reference heart rate.

[0435] (Item 19) An estimation device according to any one of Items 14 to 18, wherein the indicator information for estimating alcohol consumption is indicated by the total change in heart rate, which indicates an increase in the heart rate of the subject.

[0436] (Item 20) The estimation device described in any of Items 14 to 19, wherein the biological state information includes information representing the physical and mental state of the subject, the information representing the physical state includes information representing the mental state, information regarding the intake of specific components, or information regarding the quality of life, and the information regarding the quality of life indicates the quality of life related to the physical state, mental state, or intake of specific components.

[0437] (Item 21) The estimation device described in Item 20, wherein the information representing the physical and mental state of the subject includes information representing the physical state, and the information representing the physical state includes information relating to diseases or symptoms of the nervous and cerebrovascular systems, information relating to diseases or symptoms of the circulatory system, information relating to diseases or symptoms of the respiratory system, information relating to diseases or symptoms of the digestive system and hepatobiliary and pancreatic regions, information relating to diseases or symptoms of the endocrine and metabolic systems, information relating to diseases or symptoms of the renal and urinary systems, information relating to diseases or symptoms of sex hormones and the reproductive system, information relating to diseases or symptoms of the skin and sensory organs, information relating to diseases or symptoms related to immunity and infectious diseases, information relating to tumor symptoms, information relating to pre-disease conditions of the body, or information representing the physical state that can be indicated by the results of a biopsy.

[0438] (Item 22) The estimation device according to any one of Items 14 to 21, wherein the input information further includes one or more pieces of information from among the subject's vital information, the subject's behavioral information, the subject's environment, the subject's self-reported mental and physical state, and the subject's attribute information.

[0439] (Item 23) The estimation device according to any one of Items 14 to 22, wherein the alcohol consumption estimation index information is alcohol consumption estimation index information for a period indicated by specific conditions, the specific conditions are conditions relating to sleep, conditions relating to wakefulness, conditions relating to the time of day caused by the sun, or a combination of two or more of these conditions, and the period indicated by the specific conditions does not include the period of sleep.

[0440] (Item 24) The estimation device according to any one of Items 14 to 22, wherein the alcohol consumption estimation indicator information is alcohol consumption estimation indicator information for a period indicated by specific conditions, and the period indicated by the specific conditions includes at least one of a predetermined period before falling asleep and a predetermined period after waking up.

[0441] (Item 25) The estimation device according to any one of Items 14 to 24, wherein the learning model is constructed by learning using a learning dataset, the learning dataset includes at least indicator information for estimating the drinking of a learning subject and information on the biological state of the learning subject, the indicator information for estimating the drinking of a learning subject includes information on the total change in heart rate when the learning subject is inactive and the learning subject's heart rate exceeds the reference heart rate, the total change in heart rate represents the sum of the heart rate changes which are the changes in the learning subject's heart rate, and the biological state information includes information representing the physical and mental state of the learning subject, information representing the state of the environment that affects the physical and mental state of the learning subject, information representing the biological state affected by the physical and mental state of the learning subject, or information on the drinking of the learning subject.

[0442] (Item 26) A learning model that causes a computer to function to estimate biological state information of a subject, wherein the computer functions to take input information and output output information, the output information includes the biological state information, the biological state information includes information representing the physical and mental state of the subject, information representing the state of the environment that affects the physical and mental state of the subject, information representing the state of the body affected by the physical and mental state of the subject, or information regarding the subject's alcohol consumption, the input information includes at least an index information for estimating alcohol consumption, the index information for estimating alcohol consumption includes information regarding the total change in heart rate in a state inactive of the subject and in a state in which the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

[0443] (Item 27) An estimation method for estimating biological state information of a subject, comprising the steps of: inputting input information into a learning model; and obtaining output information from the learning model into which the input information has been input, wherein the output information includes the biological state information, the biological state information includes information representing the physical and mental state of the subject, information representing the state of the environment affecting the physical and mental state of the subject, information representing the state of the body affected by the physical and mental state of the subject, or information regarding the subject's alcohol consumption, the input information includes at least an index information for estimating alcohol consumption, the index information for estimating alcohol consumption includes information regarding the total change in heart rate in a state inactive of the subject and in a state in which the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

[0444] (Item 28) A computer program that causes a computer to execute the estimation method described in Item 27.

[0445] (Item 29) The process includes the steps of acquiring a training dataset and generating a learning model that outputs output information when input information is input by performing training using the training dataset, wherein the training dataset includes at least indicator information for estimating the drinking of a training subject and biological state information of the training subject, the indicator information for estimating the drinking of a training subject includes information on the total change in heart rate when the training subject is inactive and the training subject's heart rate exceeds a reference heart rate, the total change in heart rate of the training subject represents the sum of the heart rate changes which are the changes in the training subject's heart rate, the biological state information of the training subject includes information representing the physical and mental state of the training subject, information representing the state of the environment that affects the physical and mental state of the training subject, information representing the biological state affected by the physical and mental state of the training subject, or information regarding the drinking of the training subject, and the output information includes the biological state information of the subject. A method for generating a learning model, wherein the biological state information of the subject includes information representing the physical and mental state of the subject, information representing the state of the environment that affects the physical and mental state of the subject, information representing the state of the body affected by the physical and mental state of the subject, or information regarding the subject's alcohol consumption; the input information includes at least indicator information for estimating the subject's alcohol consumption; the indicator information for estimating the subject's alcohol consumption includes information regarding the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate; and the total change in heart rate of the subject represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

[0446] (Item 30) A computer program that causes a computer to execute the learning model generation method described in Item 29.

[0447] (Item 31) (1) A total heart rate change SumΔHR(t) calculated by summing the difference between the subject's heart rate HR(t-1) in a first unit time (t-1) and the subject's heart rate HR(t) in a second unit time (t) following the first unit time (t-1), based on the subject's heart rate HR(t) measured for each unit time (t), over a third time ((t-Δt3) to t) that is longer than the unit time (t), (2) A function L_ST(t) that indicates whether the subject is walking or not, defined based on the number of steps ST(t) measured for each unit time (t), (3) A first index for estimating alcohol consumption, defined using a function L_DR(t) that indicates whether the representative heart rate repHR(t) exceeds a reference heart rate criHR which is presumed to indicate the possibility of alcohol consumption, and which is defined from a representative heart rate repHR(t) which is a representative heart rate selected within a fourth time interval ((t-Δt4) to t) that is longer than the unit time (t), based on the heart rate HR(t), and .

[0448] (Item 32) A second indicator F(d) for estimating alcohol consumption, defined by summing or averaging the first indicator f(t) for estimating alcohol consumption described in Item 31 on a daily basis.

[0449] (Item 33) (S1) A step of calculating the heart rate HR(t) and the number of steps ST(t) of the subject for each unit time (t); (S2) A step of calculating a heart rate change amount ΔHR(t) which is the difference between the heart rate HR(t−1) of the subject within the first unit time (t−1) and the heart rate HR(t) of the subject within the second unit time (t) following the first unit time (t−1); (S3) A step of calculating a total heart rate change amount SumΔHR(t) by summing the heart rate change amount ΔHR(t) within a third time ((t−Δt3) to t) longer than the unit time (t); (S4) A step of defining a function L_ST(t) indicating whether the subject is walking or not based on the number of steps ST(t); (S5) A step of selecting a representative heart rate repHR(t) which is a representative heart rate within each of a fourth time ((t−Δt4) to t) longer than the unit time (t) based on the heart rate HR(t); (S6) A step of defining a function L_DR(t) indicating whether the representative heart rate repHR(t) exceeds a reference heart rate criHR at which there is a possibility of the subject being under the influence of alcohol; (S7) A step of defining a first index f(t) for alcohol consumption estimation using the total heart rate change amount SumΔHR(t), the function L_ST(t), and the function L_DR(t). A method for calculating an index for alcohol consumption estimation including the above steps.

[0450] (Item 34) A step of further defining a second index F(d) for alcohol consumption estimation by totaling or averaging the first index f(t) for alcohol consumption estimation on a daily basis, included in the method for calculating an index for alcohol consumption estimation according to Item 33.

[0451] (Item 35) A step of estimating whether the subject has consumed alcohol using the first index f(t) for alcohol consumption estimation according to Item 31; A step of estimating in which time zone of the day the time zone when the subject has consumed alcohol is using the first index f(t) for alcohol consumption estimation. An alcohol consumption estimation method including the above steps.

[0452] (Item 36) A drinking estimation method including: a step of estimating whether or not the subject has consumed alcohol using the second index F(d) for drinking estimation according to Item 32; and a step of estimating on which day the subject has consumed alcohol using the second index F(d) for drinking estimation.

[0453] (Item 37) A drinking estimation apparatus including: a first estimation unit that estimates whether or not the subject has consumed alcohol using the first index f(t) for drinking estimation according to Item 31; and a second estimation unit that estimates in which time zone of a day the subject has consumed alcohol using the first index f(t) for drinking estimation.

[0454] (Item 38) A drinking estimation apparatus including: a first estimation unit that estimates whether or not the subject has consumed alcohol using the second index F(d) for drinking estimation according to Item 32; and a second estimation unit that estimates on which day the subject has consumed alcohol using the second index F(d) for drinking estimation.

[0455] (Item 39) A biological examination method including: a step of extracting or excluding drinking-related biological information related to the subject's drinking using at least one of the first index f(t) for drinking estimation according to Item 31 and the second index F(d) for drinking estimation according to Item 32 from the biological information of the subject according to Item 31; and a step of examining the biological state of the subject using the extracted drinking-related biological information or the biological information obtained other than the drinking-related biological information.

[0456] (Item 40) A biological examination apparatus including: an extraction / exclusion unit that extracts or excludes drinking-related biological information related to the subject's drinking using at least one of the first index f(t) for drinking estimation according to Item 31 and the second index F(d) for drinking estimation according to Item 32 from the biological information of the subject according to Item 31; and an examination unit that examines the biological state of the subject using the extracted drinking-related biological information or the biological information obtained other than the drinking-related biological information.

[0457] This disclosure provides an information processing device, an alcohol consumption estimation device, a method for calculating an index for alcohol consumption estimation, an alcohol consumption estimation method, a biological examination method, a biological examination device, an estimation device, a learning model, an estimation method, a learning model generation method, and a computer program, and has industrial applicability.

[0458] SKS...Biometric testing system, WT...Wearable terminal, IS...Alcohol consumption estimation device, NW...Network, SY(IS)...Processing unit, KI(IS)...Memory unit, 2...Learning data creation device, 3...Learning device, 4...Estimation device, 42...Estimation unit, 402...Memory unit, 411...Alcohol consumption index calculation unit, 412...Statistical processing unit, 413...Condition processing unit, 431...Interpretation unit, TM1...Learning model

Claims

1. An information processing device comprising a processing unit that acquires a total heart rate change amount, which is the sum of the heart rate change amounts of a subject, wherein the processing unit calculates a first index for estimating alcohol consumption, which represents the total heart rate change amount when the subject is inactive and the subject's heart rate exceeds a reference heart rate.

2. The processing unit calculates the total heart rate change by summing the difference between the subject's heart rate in a first unit time and the subject's heart rate in a second unit time after the first unit time, calculated based on the subject's heart rate measured for each unit time, over a third time period longer than the unit time; the subject's inactivity is indicated by a first function that indicates whether the subject is active or not, defined based on the subject's physical activity information measured for each unit time; the state in which the subject's heart rate exceeds the reference heart rate is indicated by a second function that indicates whether the representative heart rate exceeds the reference heart rate in which it is presumed that the subject may be intoxicated; and the representative heart rate is a representative heart rate calculated over a fourth time period longer than the unit time, based on the subject's heart rate.

3. The information processing device according to claim 2, wherein the physical activity information indicates the number of steps taken by the subject, and the first function indicates whether or not the subject is walking.

4. The information processing apparatus according to claim 1 or 2, wherein the processing unit calculates a second indicator for estimating alcohol consumption by performing statistical processing on the first indicator for estimating alcohol consumption within a fifth time period.

5. The information processing apparatus according to claim 4, wherein the statistical processing refers to a process of calculating a sum or a process of calculating an average.

6. The information processing device according to claim 1 or 2, wherein the first indicator for estimating alcohol consumption is indicated by the total change in heart rate, which indicates an increase in the heart rate of the subject.

7. An alcohol estimation device comprising at least one of the following: a first estimation unit that estimates whether or not a subject has consumed alcohol using a first indicator for alcohol estimation; and a second estimation unit that estimates when the subject consumed alcohol using the first indicator for alcohol estimation, wherein the first indicator for alcohol estimation represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

8. An alcohol estimation device comprising at least one of the following: a first estimation unit that estimates whether or not a subject has consumed alcohol using a second indicator for alcohol consumption estimation; and a second estimation unit that estimates when the subject consumed alcohol using the second indicator for alcohol consumption estimation, wherein the second indicator for alcohol consumption estimation is an indicator obtained by performing statistical processing on the first indicator for alcohol consumption estimation within a predetermined time period, the first indicator for alcohol consumption estimation represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

9. A method for calculating an index for estimating alcohol consumption, comprising the steps of: obtaining a total heart rate change amount, which is the sum of the heart rate change amounts of a subject; and calculating a first index for estimating alcohol consumption, which represents the total heart rate change amount when the subject is inactive and the subject's heart rate exceeds a reference heart rate.

10. A method for estimating alcohol consumption, comprising at least one of the following steps: estimating whether or not a subject has consumed alcohol using a first indicator for estimating alcohol consumption; and estimating when the subject consumed alcohol using the first indicator for estimating alcohol consumption, wherein the first indicator for estimating alcohol consumption represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the changes in the subject's heart rate.

11. A method for estimating alcohol consumption, comprising at least one of the following steps: estimating whether or not a subject has consumed alcohol using a second indicator for estimating alcohol consumption; and estimating when the subject consumed alcohol using the second indicator for estimating alcohol consumption, wherein the second indicator for estimating alcohol consumption is an indicator obtained by performing statistical processing on a first indicator for estimating alcohol consumption within a predetermined time period; the first indicator for estimating alcohol consumption represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate; and the total change in heart rate represents the sum of the changes in heart rate, which are the changes in the subject's heart rate.

12. A biological examination method comprising: a step of extracting or excluding alcohol-related biological information related to a subject's alcohol consumption using at least one of a first indicator for alcohol consumption estimation and a second indicator for alcohol consumption estimation from the subject's biological information; and a step of examining the subject's biological state using the extracted alcohol-related biological information or the biological information other than the alcohol-related biological information, wherein the first indicator for alcohol consumption estimation represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate, the total change in heart rate represents the sum of the changes in the subject's heart rate, and the second indicator for alcohol consumption estimation is an indicator obtained by performing statistical processing on the first indicator for alcohol consumption estimation within a predetermined time.

13. A biological testing device comprising: an extraction / exclusion unit that extracts or excludes alcohol-related biological information relating to a subject's alcohol consumption using at least one of a first indicator for alcohol consumption estimation and a second indicator for alcohol consumption estimation from the subject's biological information; and an inspection unit that examines the subject's biological state using the extracted alcohol-related biological information or the biological information other than the alcohol-related biological information, wherein the first indicator for alcohol consumption estimation represents the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a standard heart rate, the total change in heart rate represents the sum of the heart rate changes which are the changes in the subject's heart rate, and the second indicator for alcohol consumption estimation is an indicator obtained by performing statistical processing on the first indicator for alcohol consumption estimation within a predetermined time.

14. Estimation device for estimating biological state information of a subject, comprising: an estimation unit that inputs input information to a learning model and obtains output information from the learning model; and a storage unit that stores the output information, wherein the output information includes the biological state information, the biological state information includes information representing the physical and mental state of the subject, information representing the state of the environment that affects the physical and mental state of the subject, information representing the state of the body affected by the physical and mental state of the subject, or information regarding the subject's alcohol consumption, the input information includes at least index information for estimating alcohol consumption, the index information for estimating alcohol consumption includes information regarding the total change in heart rate in a state inactive of the subject and in a state in which the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

15. The estimation device according to claim 14, wherein the total change in heart rate is calculated by summing the heart rate change, which is the difference between the subject's heart rate in a first unit time and the subject's heart rate in a second unit time after the first unit time, calculated based on the subject's heart rate measured for each unit time, over a third time period longer than the unit time; the subject's inactivity is indicated by a first function that indicates whether the subject is active or not, defined based on the subject's physical activity information measured for each unit time; the state in which the subject's heart rate exceeds the reference heart rate is indicated by a second function that indicates whether the representative heart rate exceeds the reference heart rate in which it is presumed that the subject may be intoxicated; and the representative heart rate is a representative heart rate calculated over a fourth time period longer than the unit time, based on the subject's heart rate.

16. The estimation device according to claim 15, wherein the physical activity information indicates the number of steps taken by the subject, and the first function indicates whether or not the subject is walking.

17. The estimation device according to claim 14 or 15, wherein the alcohol consumption estimation index information includes a first index for alcohol consumption estimation that indicates the change in total heart rate when the subject is inactive and the subject's heart rate exceeds the reference heart rate.

18. The estimation device according to claim 14 or 15, wherein the alcohol consumption estimation index information includes a second alcohol consumption estimation index obtained by performing statistical processing on a first alcohol consumption estimation index within a predetermined time, and the first alcohol consumption estimation index indicates the total change in heart rate when the subject is inactive and the subject's heart rate exceeds the reference heart rate.

19. The estimation device according to claim 14 or 15, wherein the indicator information for estimating alcohol consumption is indicated by the total change in heart rate, which indicates an increase in the heart rate of the subject.

20. The estimation device according to claim 14 or 15, wherein the biological state information includes information representing the physical and mental state of the subject, the information representing the physical state includes information representing the mental state, information regarding the intake of specific components, or information regarding the quality of life, and the information regarding the quality of life indicates the quality of life related to the physical state, mental state, or intake of specific components.

21. The estimation device according to claim 20, wherein the information representing the physical and mental state of the subject includes information representing the physical state, and the information representing the physical state includes information relating to diseases or symptoms of the nervous system and the cranial nervous system, information relating to diseases or symptoms of the circulatory system, information relating to diseases or symptoms of the respiratory system, information relating to diseases or symptoms of the digestive system and the hepatobiliary and pancreatic region, information relating to diseases or symptoms of the endocrine and metabolic system, information relating to diseases or symptoms of the renal and urinary system, information relating to diseases or symptoms of sex hormones and the reproductive system, information relating to diseases or symptoms of the skin and sensory organs, information relating to diseases or symptoms related to immunity and infectious diseases, information relating to tumor symptoms, information relating to pre-disease conditions of the body, or information representing the physical state that can be indicated by the results of a biological examination.

22. The estimation device according to claim 14 or claim 15, wherein the input information further includes one or more pieces of information from among the subject's vital information, the subject's behavioral information, the subject's environment, the subject's self-reported information regarding the subject's physical and mental state, and the subject's attribute information.

23. The estimation device according to claim 14 or 15, wherein the alcohol consumption estimation index information is alcohol consumption estimation index information for a period indicated by specific conditions, the specific conditions are conditions relating to sleep, conditions relating to wakefulness, conditions relating to the time of day caused by the sun, or a combination of two or more of these conditions, and the period indicated by the specific conditions does not include the period of sleep.

24. The estimation device according to claim 14 or claim 15, wherein the alcohol consumption estimation index information is alcohol consumption estimation index information for a period indicated by specific conditions, and the period indicated by the specific conditions includes at least one of a predetermined period before falling asleep and a predetermined period after waking up.

25. The estimation device according to claim 14 or 15, wherein the learning model is constructed by learning using a learning dataset, the learning dataset includes at least an index information for estimating the drinking of a learning subject and information on the biological state of the learning subject, the index information for estimating the drinking of a learning subject includes information on the total change in heart rate when the learning subject is inactive and the learning subject's heart rate exceeds the reference heart rate, the total change in heart rate represents the sum of the heart rate changes which are the changes in the learning subject's heart rate, and the biological state information includes information representing the physical and mental state of the learning subject, information representing the state of the environment that affects the physical and mental state of the learning subject, information representing the biological state affected by the physical and mental state of the learning subject, or information on the drinking of the learning subject.

26. A learning model that causes a computer to function to estimate biological state information of a subject, wherein the computer functions to take input information and output output information, the output information includes the biological state information, the biological state information includes information representing the physical and mental state of the subject, information representing the state of the environment affecting the physical and mental state of the subject, information representing the state of the body affected by the physical and mental state of the subject, or information regarding the subject's alcohol consumption, the input information includes at least index information for estimating alcohol consumption, the index information for estimating alcohol consumption includes information regarding the total change in heart rate in a state of inactivity of the subject and in a state where the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

27. An estimation method for estimating biological state information of a subject, comprising the steps of: inputting input information into a learning model; and obtaining output information from the learning model into which the input information has been input, wherein the output information includes the biological state information, the biological state information includes information representing the physical and mental state of the subject, information representing the state of the environment affecting the physical and mental state of the subject, information representing the state of the body affected by the physical and mental state of the subject, or information regarding the subject's alcohol consumption, the input information includes at least an index information for estimating alcohol consumption, the index information for estimating alcohol consumption includes information regarding the total change in heart rate in a state inactive of the subject and in a state in which the subject's heart rate exceeds a reference heart rate, and the total change in heart rate represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

28. A computer program that causes a computer to execute the estimation method described in claim 27.

29. The process includes the steps of acquiring a training dataset and generating a learning model that outputs output information when input information is input by performing training using the training dataset, wherein the training dataset includes at least indicator information for estimating the drinking of a training subject and biological state information of the training subject, the indicator information for estimating the drinking of a training subject includes information on the total change in heart rate when the training subject is inactive and the training subject's heart rate exceeds a reference heart rate, the total change in heart rate of the training subject represents the sum of the heart rate changes which are the changes in the training subject's heart rate, the biological state information of the training subject includes information representing the physical and mental state of the training subject, information representing the state of the environment that affects the physical and mental state of the training subject, information representing the biological state affected by the physical and mental state of the training subject, or information regarding the drinking of the training subject, and the output information includes the biological state information of the subject. A method for generating a learning model, wherein the biological state information of the subject includes information representing the physical and mental state of the subject, information representing the state of the environment that affects the physical and mental state of the subject, information representing the state of the body affected by the physical and mental state of the subject, or information regarding the subject's alcohol consumption; the input information includes at least indicator information for estimating the subject's alcohol consumption; the indicator information for estimating the subject's alcohol consumption includes information regarding the total change in heart rate when the subject is inactive and the subject's heart rate exceeds a reference heart rate; and the total change in heart rate of the subject represents the sum of the heart rate changes, which are the changes in the subject's heart rate.

30. A computer program that causes a computer to execute the learning model generation method described in claim 29.

Images