Physical condition diagnosis device, wearable device, physical condition diagnosis system, model determination device, physical condition diagnosis method, and program

Abstract

A physical condition diagnosis device (1) is provided with an information acquisition unit (11), an activity body temperature estimation unit (12), and a physical condition diagnosis unit (13). The information acquisition unit (11) acquires biological information including temperature information of at least one site of a target organism to be subjected to physical condition diagnosis, and variation factor information including activity information indicating activity of the target organism that varies the temperature of the site. The activity body temperature estimation unit (12) finds, from the biological information and the variation factor information, activity body temperature data indicating temperature variation of the site due to the activity. The physical condition diagnosis unit (13) finds a physical condition index, indicating a physical condition of the target organism, on the basis of the activity body temperature data.

Description

Physical condition diagnosis device, wearable device, physical condition diagnosis system, model determination device, physical condition diagnosis method, and program

[0001] The present disclosure relates to a physical condition diagnosis device, a wearable device, a physical condition diagnosis system, a model determination device, a physical condition diagnosis method, and a program.

[0002] As an example of health management, technology has been developed to predict the health of a target organism based on its biological information. Patent Literature 1 discloses a biological rhythm prediction device that predicts the ultradian rhythm waveform of a user's biological information, such as body temperature and heart rate index. The biological rhythm prediction device disclosed in Patent Literature 1 reduces the influence of influencing factors from biological signals with an ultradian rhythm period and predicts the ultradian rhythm waveform at the time when the user's scheduled behavior will be performed, taking into account the user's scheduled behavior.

[0003] As another example, Patent Document 2 discloses a physical condition monitoring device that evaluates abnormal conditions in the human body from biological information such as blood pressure, pulse, and body temperature. The physical condition monitoring device disclosed in Patent Document 2 corrects a reference value based on the mental activity state or physical activity state of the person being measured, and determines whether or not there is an abnormality in the human body based on a comparison between the measured biological information and the corrected reference value.

[0004] JP2021-49041A JP4-71532A

[0005] The biological rhythm prediction device disclosed in Patent Document 1 can predict the ultradian rhythm waveform at the time when the scheduled action will be performed, taking into account the scheduled action of the user, but cannot diagnose the physical condition of the subject.

[0006] The physical condition monitoring device disclosed in Patent Document 2 determines whether or not there is an abnormality in the human body based on a comparison between a measured value of biological information, such as a measured body temperature, and a corrected reference value. Therefore, the physical condition monitoring device disclosed in Patent Document 2 can diagnose the abnormality from the body temperature fluctuation if the abnormality in the human body becomes a dominant factor in the body temperature fluctuation and causes a large fluctuation in body temperature. In other words, the physical condition monitoring device disclosed in Patent Document 2 has difficulty detecting an internal abnormality that is not a dominant factor in the body temperature fluctuation, making it difficult to accurately diagnose the physical condition of the target organism.

[0007] The present disclosure has been made in consideration of the above-mentioned circumstances, and aims to provide a health condition diagnosis device, a wearable device, a health condition diagnosis system, a model determination device, a health condition diagnosis method, and a program that accurately diagnose the health condition of a target organism.

[0008] To achieve the above object, the physical condition assessment device of the present disclosure includes an information acquisition unit, an activity body temperature estimation unit, and a physical condition assessment unit. The information acquisition unit acquires biological information including temperature information of at least one body part of a target organism that is the subject of physical condition assessment, and variation factor information including activity information indicating the activity of the target organism that causes the temperature of the body part to vary. The activity body temperature estimation unit obtains activity body temperature data indicating the temperature variation of the body part due to activity from the biological information and the variation factor information. The physical condition assessment unit obtains a physical condition index indicating the physical condition of the target organism based on the activity body temperature data.

[0009] The health condition diagnosis device of the present disclosure calculates a health condition index of the target organism based on activity body temperature data that indicates temperature fluctuations in parts of the body caused by the activity of the target organism, thereby making it possible to accurately diagnose the health condition of the target organism.

[0010] FIG. 1 is a block diagram of a physical condition diagnosis system according to embodiment 1; FIG. 2 is a diagram showing a hardware configuration of a physical condition diagnosis device according to embodiment 1; FIG. 3 is a flowchart showing an example of the operation of a physical condition diagnosis process performed by the physical condition diagnosis device according to embodiment 1; FIG. 4 is a timing chart showing examples of temperature information, activity information, and activity body temperature data in embodiment 1; FIG. 5 is a diagram showing an example of a display of a diagnosis result by the physical condition diagnosis device according to embodiment 1; FIG. 6 is a block diagram of a physical condition diagnosis system according to embodiment 2; FIG. 7 is a timing chart showing examples of temperature information and activity information in embodiment 2; FIG. 8 is a flowchart showing an example of the operation of an activity body temperature data calculation process performed by the physical condition diagnosis device according to embodiment 2; FIG. 9 is a timing chart showing an example of activity body temperature data in embodiment 2; FIG. 10 is a diagram showing an example of a heat transfer path within the body of a target organism according to embodiment 2; FIG. 11 is a diagram showing an example of a display of a diagnosis result by the physical condition diagnosis device according to embodiment 2;

[0011] Hereinafter, a physical condition diagnosis apparatus, a wearable device, a physical condition diagnosis system, a model determination apparatus, a physical condition diagnosis method, and a program according to embodiments of the present disclosure will be described in detail with reference to the drawings. In the drawings, the same or equivalent parts are designated by the same reference numerals.

[0012] (Embodiment 1) A physical condition diagnosis device for diagnosing the physical condition of a living organism such as a human or an animal, and a physical condition diagnosis system including the physical condition diagnosis device will be described in embodiment 1. The living organism that is the subject of a physical condition diagnosis is referred to as a target organism. The physical condition diagnosis system 100 shown in Fig. 1 includes a temperature sensor 51 that measures the temperature of at least one part of the target organism, an activity sensor 52 that generates an activity index that numerically indicates the activity of the target organism that causes the temperature of the part to fluctuate, a physical condition diagnosis device 1 that calculates a physical condition index that indicates the physical condition of the target organism from the measurement results of the temperature sensor 51 and the activity index generated by the activity sensor 52, and an output device 53 that outputs the diagnosis results of the physical condition diagnosis device 1, including the physical condition index.

[0013] The temperature sensor 51 is placed inside or outside the body of the target organism and measures the temperature of at least one part of the target organism. The temperature of a part of the target organism is the temperature of any part that can be measured or estimated, and is not limited to the temperature of the body surface, but also includes internal temperatures such as organ temperatures and core body temperature. The temperature sensor 51 includes, for example, a temperature sensor attached to the armpit, calf, or upper arm of the target organism whose physical condition is to be diagnosed, a temperature sensor attached inside the stomach of livestock whose physical condition is to be diagnosed, an infrared camera or thermograph attached at a position distant from the target organism, etc. The temperature sensor attached to the target organism is a thermistor, a resistance temperature detector, etc. In the first embodiment, the temperature sensor 51 is a temperature sensor attached to the armpit of the target organism whose physical condition is to be diagnosed.

[0014] The activity sensor 52 generates an activity index that numerically indicates the activities of the target organism, such as eating, exercise, sleep, and excretion. The activity sensor 52 includes, for example, a wearable terminal that measures the heart rate, movement speed, etc. of the target organism, a camera that can photograph the target organism, and a position sensor that can acquire the position of the target organism. In the first embodiment, the activity sensor 52 is a position sensor that detects whether the target organism, whose physical condition is to be diagnosed, is in a location where it would be located when eating. When the position sensor detects that the target organism is located in that location, it can be assumed that the target organism is eating.

[0015] The physical condition diagnosis device 1 includes an information acquisition unit 11 that acquires biological information including temperature information of at least one part of the target organism and fluctuation factor information including activity information indicating the activity of the target organism that causes the temperature of the part to fluctuate, an activity body temperature estimation unit 12 that calculates activity body temperature data that indicates the temperature fluctuation of the part due to the activity of the target organism, and a physical condition diagnosis unit 13 that calculates a physical condition index that indicates the physical condition of the target organism based on the activity body temperature data.

[0016] The information acquisition unit 11 acquires the measurement results of the temperature sensor 51 as temperature information, and acquires the activity index generated by the activity sensor 52 as activity information. The information acquisition unit 11 sends biological information including the temperature information to the activity body temperature estimation unit 12, and sends fluctuation factor information including the activity information to the activity body temperature estimation unit 12 and the physical condition diagnosis unit 13.

[0017] The active body temperature estimation unit 12 calculates active body temperature data indicating temperature fluctuations in a part of the target organism due to the activity of the target organism from the biological information and the fluctuation factor information. As an example, the active body temperature estimation unit 12 calculates active body temperature data indicating temperature fluctuations in the stomach during a meal.

[0018] The physical condition diagnosis unit 13 has a waveform processing unit 21 that performs waveform processing of the activity body temperature data to diagnose the physical condition of the target organism.

[0019] The waveform processing unit 21 calculates a physical condition index based on at least one of the phase characteristics, amplitude, frequency spectrum, and waveform of the waveform data of the activity-related body temperature data.

[0020] The output device 53 receives the physical condition index from the physical condition diagnosis unit 13. The output device 53 outputs at least one of visual information and audio information indicating the received physical condition index and a control signal to an external device according to the physical condition index.

[0021] The hardware configuration of the physical condition diagnosis device 1 having the above-described configuration is shown in Figure 2. The physical condition diagnosis device 1 includes a processor 61, a memory 62, and an interface 63. The processor 61, the memory 62, and the interface 63 are connected to one another via a bus 60. The functions of each unit of the physical condition diagnosis device 1 are realized by software, firmware, or a combination of software and firmware. The software and firmware are written as programs and stored in the memory 62. The processor 61 reads and executes the programs stored in the memory 62, thereby realizing the functions of each unit described above. In other words, the memory 62 stores programs for executing the processing of each unit of the physical condition diagnosis device 1.

[0022] The memory 62 includes, for example, non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read-Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable and Programmable Read-Only Memory), magnetic disk, flexible disk, optical disk, CD (Compact Disc), minidisk, DVD (Digital Versatile Disc), etc.

[0023] The physical condition diagnosis device 1 is connected to a temperature sensor 51, an activity sensor 52, and an output device 53 via an interface 63. The interface 63 has interface modules that comply with one or more standards depending on the connected device.

[0024] When the physical condition diagnosis device 1 having the above configuration is started, it repeats the physical condition diagnosis process shown in Fig. 3 at predetermined intervals. The information acquisition unit 11 included in the physical condition diagnosis device 1 acquires biological information including temperature information measured by the temperature sensor 51, and acquires fluctuation factor information including activity information including an activity index generated by the activity sensor 52 (step S11).

[0025] An example of temperature information included in the biological information acquired by the information acquisition unit 11 is shown in graph A of FIG. 4 . The horizontal axis of graph A in FIG. 4 represents time, and the vertical axis represents armpit temperature (unit: degrees). Graph B in FIG. 4 represents an example of activity information included in the fluctuation factor information acquired by the information acquisition unit 11. The horizontal axis of graph B in FIG. 4 represents time, and the vertical axis represents the activity index generated by the activity sensor 52. The activity index indicates an H (High) level or an L (Low) level. The activity index generated by the activity sensor 52 is H level when the target organism of the physical condition diagnosis is located in a place where it eats, and is L level when the target organism is not located in a place where it eats. In this case, if the activity index acquired by the information acquisition unit 11 from the activity sensor 52 is H level, it can be assumed that the target organism is eating, as shown in graph B of FIG. 4 . If the activity index acquired by the information acquisition unit 11 from the activity sensor 52 is L level, it can be assumed that the target organism is not eating.

[0026] The information acquisition unit 11 sends the above-mentioned biological information and fluctuation factor information to the active body temperature estimation unit 12 , and sends the fluctuation factor information to the physical condition diagnosis unit 13 .

[0027] As shown in Figure 3, the activity body temperature estimation unit 12 obtains activity body temperature data indicating temperature fluctuations at parts of the body caused by the activity of the target organism from the biological information and the fluctuation factor information (step S12). The activity of the target organism includes the mental activity and physical activity of the target organism. The correspondence between the activity of the target organism and the parts whose temperature fluctuates in response to the activity can be obtained from biological knowledge. For example, the biological knowledge can be used to associate the activity of the target organism with the organs in which heat production or heat dissipation occurs due to the activity.

[0028] In the first embodiment, the active body temperature estimation unit 12 calculates the stomach temperature fluctuation of the target organism during a period in which the activity information included in the fluctuation factor information indicates that the target organism is eating. The active body temperature estimation unit 12 extracts a fluctuation component from the temperature information during periods in which the activity index shown in FIG. 4 is at level H, indicating that the target organism is eating, specifically, the period from time T0 to time T1 and the period from time T2 to time T3. While the target organism is eating, the stomach temperature rises due to digestive activity. Therefore, the fluctuation component obtained from the temperature information during the period in which the target organism is eating corresponds to the stomach temperature fluctuation.

[0029] As described above, the active body temperature estimation unit 12 extracts the fluctuation components of the temperature information while the activity index is at level H, thereby obtaining active body temperature data that indicates the temperature fluctuations of the stomach while the target organism is eating, as shown in graph C in Figure 4. The horizontal axis of graph C in Figure 4 indicates time, and the vertical axis indicates the temperature fluctuations of the stomach (unit: degrees). The active body temperature estimation unit 12 sends the obtained active body temperature data to the physical condition diagnosis unit 13.

[0030] As shown in Fig. 3, the physical condition diagnosis unit 13 included in the physical condition diagnosis device 1 calculates a physical condition index of the target organism based on the active body temperature data acquired from the active body temperature estimation unit 12 and the activity information acquired from the information acquisition unit 11 (step S13). The physical condition diagnosis unit 13 diagnoses the physical condition of the target organism by analyzing a single organ or multiple organs that contribute to a specific function. In detail, the physical condition diagnosis unit 13 calculates a physical condition index that indicates at least one of the following: the state of the target organism's physical condition, the presence or absence of a disease in the target organism, and the numerical health level of the target organism. In the first embodiment, the physical condition diagnosis unit 13 calculates the health level of the stomach and stomach metabolism as the physical condition index.

[0031] As shown in FIG. 1 , the waveform processing unit 21 included in the physical condition diagnosis unit 13 acquires activity information from the information acquisition unit 11 and acquires activity body temperature data from the activity body temperature estimation unit 12. The waveform processing unit 21 performs an FFT (Fast Fourier Transform) on the activity body temperature data to generate frequency domain data and determines whether the frequency corresponding to the peak value of the frequency domain data is within a target range. The target range can be determined based on biological knowledge. If the frequency corresponding to the peak value is within the target range, it can be assumed that the digestive function and organs related to digestion are healthy. If the frequency corresponding to the peak value is not within the target range, it can be assumed that there is a possibility of a disease in the digestive function and organs related to digestion.

[0032] In the first embodiment, the physical condition diagnosis unit 13 calculates the degree of stomach health from the ratio between the frequency corresponding to the peak value of the frequency domain data and an ideal value, which is the median of the target range. When the frequency corresponding to the peak value of the frequency domain data matches the ideal value, the degree of stomach health is highest, and as the difference between the frequency corresponding to the peak value of the frequency domain data and the ideal value increases, the degree of stomach health decreases.

[0033] Organ metabolism is closely related to heat production, and heat production can be estimated from organ temperature fluctuations. Therefore, the physical condition diagnosis unit 13 determines stomach metabolic fluctuations from stomach temperature fluctuations indicated by activity body temperature data. The physical condition diagnosis unit 13 transmits diagnosis results including physical condition indicators indicating the health of the stomach and stomach metabolic fluctuations to the output device 53. The physical condition diagnosis device 1 ends the physical condition diagnosis process when the process of step S13 in Figure 3 is completed. The physical condition diagnosis device 1 repeats the above-mentioned process at specified intervals.

[0034] The output device 53 outputs at least one of visual information and audio information indicating the physical condition index included in the received diagnostic result, and a control signal to an external device corresponding to the physical condition index. For example, the output device 53 is a display device having a screen 54 on which the diagnostic result is displayed, as shown in Fig. 5. A display area 54a of the screen 54 displays the stomach health level included in the diagnostic result, and a display area 54b of the screen 54 displays the stomach metabolic fluctuation included in the diagnostic result.

[0035] As explained above, the physical condition evaluation device 1 according to the first embodiment calculates a physical condition index of the target organism from activity body temperature data that indicates temperature fluctuations in parts of the target organism due to the activity of the target organism. Therefore, it is possible to diagnose the physical condition of the target organism by calculating a physical condition index of the target organism from temperature fluctuations in parts of the target organism before an abnormality inside the target organism becomes the dominant factor in the temperature fluctuations and the body temperature fluctuates significantly. Therefore, the physical condition evaluation device 1 according to the first embodiment can accurately diagnose the physical condition of the target organism.

[0036] (Embodiment 2) The method of calculating the physical condition index is not limited to the above example, and the physical condition index may be calculated from biological information including temperature information of multiple body parts and multiple types of activity information. A physical condition assessment device 2 according to embodiment 2 shown in Figure 6 calculates a physical condition index that indicates the physical condition of a target organism from the measurement results of multiple temperature sensors 51a, 51b and activity indices generated by multiple activity sensors 52a, 52b.

[0037] The physical condition diagnosis unit 13 provided in the physical condition diagnosis device 2 includes, in addition to the waveform processing unit 21, a matching processing unit 22 that determines a physical condition index by performing pattern matching processing between the activity body temperature data and predetermined pattern data, and a statistical processing unit 23 that determines other physical condition indexes of the target organism based on the history of the activity body temperature data and at least one of the combinations of the physical condition indexes determined by the waveform processing unit 21 and the matching processing unit 22.

[0038] The hardware configuration of the physical condition diagnosis device 2 according to embodiment 2 is similar to the hardware configuration of the physical condition diagnosis device 1 according to embodiment 1. However, the physical condition diagnosis device 2 is connected to temperature sensors 51 a and 51 b, activity sensors 52 a and 52 b, and an output device 53 via an interface 63.

[0039] The information acquisition unit 11 included in the physical condition diagnosis device 2 acquires the measurement results of the temperature sensors 51 a and 51 b as temperature information. The temperature sensor 51 a is a temperature sensor attached to the armpit of the target organism, similar to the temperature sensor 51 according to embodiment 1. The temperature sensor 51 b is a temperature sensor attached to the surface of the calf of the target organism.

[0040] The information acquisition unit 11 acquires, as activity information, activity indices generated by the activity sensors 52 a, 52 b and type information indicating the type of activity associated with the activity sensors 52 a, 52 b. The activity sensor 52 a includes a camera that photographs the meal contents for physical condition diagnosis and a device that estimates calorie intake from the photographed meal contents. The activity sensor 52 a outputs the estimated calorie intake as the activity index, and outputs, as type information, a signal that is at an H level indicating that the subject of the physical condition diagnosis is eating if the estimated calorie intake is equal to or greater than a threshold intake amount at which the subject of the physical condition diagnosis is considered to be eating, or that is at an L level indicating that the subject of the physical condition diagnosis is not eating if the estimated calorie intake is less than the threshold intake amount.

[0041] The activity sensor 52b is a speed sensor attached to the target organism to measure the speed of the target organism. For example, the activity sensor 52b is a wearable terminal worn by the target organism for health diagnosis, and measures the speed of the target organism. The activity sensor 52b outputs, as type information, a signal that is at an H level indicating that the target organism is walking if the measured speed of the target organism is within a target walking speed range at which the target organism can be considered to be walking, and at an L level indicating that the target organism is not walking if the measured speed of the target organism is not within the target walking speed range. The activity sensor 52b outputs the measured speed of the target organism as an activity index.

[0042] When the physical condition assessment device 2 having the above configuration is started, it repeats the physical condition assessment process shown in Figure 3 at set intervals, as in embodiment 1. The information acquisition unit 11 included in the physical condition assessment device 2 acquires biological information including temperature information measured by the temperature sensors 51a and 51b, and acquires fluctuation factor information including type information and activity information including an activity index generated by the activity sensors 52a and 52b. The information acquisition unit 11 sends the temperature information and activity information to the activity body temperature estimation unit 12 and the physical condition assessment unit 13.

[0043] Graphs A and B in Fig. 7 show examples of temperature information included in the biological information acquired by the information acquisition unit 11. Graph A shows temperature information output by temperature sensor 51a. The horizontal axis of graph A represents time, and the vertical axis represents armpit temperature (unit: degrees). Graph B shows temperature information output by temperature sensor 51b. The horizontal axis of graph B represents time, and the vertical axis represents calf surface temperature (unit: degrees).

[0044] Graphs C and D in FIG. 7 show examples of activity information included in the variation factor information acquired by the information acquisition unit 11. Graph C shows type information output by the activity sensor 52a. The horizontal axis of graph C represents time, and the vertical axis represents the H / L level. Graph D shows the activity index output by the activity sensor 52a. The horizontal axis of graph D represents time, and the vertical axis represents calorie intake (unit: kcal).

[0045] Graphs E and F in FIG. 7 show examples of activity information included in the variation factor information acquired by the information acquisition unit 11. Graph E shows type information output by the activity sensor 52b. The horizontal axis of graph E represents time, and the vertical axis represents the H / L level. Graph F shows the activity index output by the activity sensor 52b. The horizontal axis of graph F represents time, and the vertical axis represents the speed (unit: m / min) of the target organism.

[0046] The active body temperature estimation unit 12 obtains active body temperature data indicating temperature fluctuations at parts of the body caused by activity of the target organism from the biological information and fluctuation factor information. Specifically, the active body temperature estimation unit 12 obtains the active body temperature data by the active body temperature data calculation process shown in Fig. 8. Specifically, the active body temperature estimation unit 12 performs the active body temperature data calculation process shown in Fig. 8 every time it obtains temperature information and activity information from the information acquisition unit 11.

[0047] The active body temperature estimation unit 12 performs an FFT on each of the temperature information acquired from the temperature sensors 51a and 51b to generate data in the frequency domain, extracts the frequency components corresponding to the periodic fluctuations of the target organism, and performs an IFFT (Inverse Fast Fourier Transform) to extract data on the periodic fluctuations in the time domain (step S21).

[0048] For example, the active body temperature estimator 12 extracts frequency components corresponding to periodic fluctuations such as circadian and ultradian from the frequency domain data obtained by performing an FFT on the armpit temperature data indicated by the temperature information acquired from the temperature sensor 51a and the calf surface temperature data indicated by the temperature information acquired from the temperature sensor 51b. An example of active body temperature data indicating periodic fluctuations is shown in graph A in Figure 9. The horizontal axis of graph A represents time, and the vertical axis represents temperature fluctuations (unit: degrees).

[0049] For example, since the human circadian rhythm is approximately 24 hours, frequency components corresponding to the circadian rhythm can be obtained by extracting frequency components corresponding to a time domain period of 23 hours or more and 25 hours or less from frequency domain data. Therefore, the active body temperature estimator 12 can extract frequency components corresponding to the circadian rhythm by filtering the frequency domain data obtained by performing an FFT on the armpit temperature data indicated by the temperature information acquired from the temperature sensor 51a using an LPF (Low Pass Filter). The active body temperature estimator 12 then performs an IFFT on the extracted frequency components to obtain data on the periodic fluctuations in the time domain.

[0050] 8, the active body temperature estimator 12 obtains the temperature information acquired from the temperature sensors 51a and 51b with the periodic fluctuation components removed (step S22). Specifically, the active body temperature estimator 12 performs HPF (High Pass Filter) filtering on the frequency domain data generated in step S21 to remove the periodic fluctuation components corresponding to the circadian.

[0051] The activity body temperature estimating unit 12 obtains activity body temperature data from the data obtained in step S22 in accordance with the activity information acquired from the activity sensors 52a and 52b (step S23).

[0052] As shown in Figure 10, the active body temperature estimation unit 12 estimates the stomach temperature of the target organism from the armpit temperature data from which the cyclical fluctuation component has been removed in step S22 and the calf surface temperature data from which the cyclical fluctuation component has been removed in step S22, based on the heat transfer pathways within the body of the target organism. Heat is transferred within the body of the target organism through heat production, such as heat generation in the digestive organs during eating and in muscles during exercise, and heat dissipation, such as sweating. Based on the heat transfer pathways within the body caused by such heat production and heat dissipation, the active body temperature estimation unit 12 estimates the stomach temperature of the target organism.

[0053] In the example of Figure 10, heat is transferred from heat sources 91 and 92 to regions 81 and 82. Heat source 91 is, for example, the stomach, which generates heat as it is active while the target organism is eating. Heat source 92 is, for example, the calf muscle, which generates heat while the target organism is exercising. Regions 81 and 82 are parts of the target organism's body and are locations that are the measurement targets of temperature sensors 51a and 51b. Region 81 is, for example, the armpit. Region 82 is, for example, the surface of the calf.

[0054] In FIG. 10 , the amount of heat transferred is represented by the thickness of the arrows. The amount of heat transferred from heat source 91 to portion 81 is greater than the amount of heat transferred from heat source 91 to portion 82. The amount of heat transferred from heat source 92 to portion 82 is greater than the amount of heat transferred from heat source 92 to portion 81. In this case, the temperature change ΔT at portion 81 due to heat generation at heat sources 91 and 92 can be expressed as a1 * temperature difference Δt1 between portion 81 and heat source 91 + a2 * temperature difference Δt2 between portion 81 and heat source 92. a1 and a2 are predetermined coefficients, and because portion 81 is more strongly affected by heat source 91, a1 > a2 holds. For example, a1 = 0.6 and a2 = 0.1. Based on the above formula, the temperatures of heat sources 91 and 92 can be estimated from the temperatures measured at portions 81 and 82.

[0055] Based on the above-mentioned heat transfer path, the active body temperature estimation unit 12 estimates the stomach temperature of the target organism from the armpit temperature data from which the cyclic fluctuation component has been removed in step S22 of Fig. 8 and the calf surface temperature data from which the cyclic fluctuation component has been removed in step S22. Based on the estimated stomach temperature, the active body temperature estimation unit 12 calculates active body temperature data that indicates stomach temperature fluctuations during the period when the type information generated by the activity sensor 52a is at the H level.

[0056] An example of activity temperature data showing stomach temperature fluctuations is shown in graph B of Figure 9. The horizontal axis of graph B represents time, and the vertical axis represents temperature fluctuations (unit: degrees).

[0057] Furthermore, the activity body temperature estimation unit 12 calculates the temperature fluctuations of the calf muscles during the period when the type information included in the activity information obtained from the activity sensor 52b is at H level from the calf surface temperature data from which the periodic fluctuation components have been removed in step S22.

[0058] An example of activity temperature data showing temperature fluctuations in the calf muscles is shown in graph C of Figure 9. The horizontal axis of graph C represents time, and the vertical axis represents temperature fluctuations (unit: degrees).

[0059] After determining the active body temperature data for each body part as described above, the active body temperature estimation unit 12 sends the determined active body temperature data to the physical condition diagnosis unit 13, and ends the active body temperature data calculation process.

[0060] The waveform processing unit 21 included in the physical condition diagnosis unit 13 determines the health level of the stomach and the metabolic fluctuations of the stomach, as in embodiment 1. The waveform processing unit 21 acquires temperature information from the information acquisition unit 11, and outputs a diagnosis result including the temperature information, for example, information on the armpit temperature of the subject organism, the health level of the stomach, and the metabolic fluctuations of the stomach to the statistical processing unit 23 and the output device 53.

[0061] The matching processing unit 22 diagnoses the physical condition of the target organism by performing pattern matching processing between activity body temperature data indicating temperature fluctuations in parts of the target organism caused by exercise and predetermined pattern data. In detail, the matching processing unit 22 performs pattern matching processing between activity body temperature data indicating temperature fluctuations in the calf muscles obtained by the activity body temperature estimation unit 12 and predetermined pattern data indicating temperature fluctuations during convulsions.

[0062] The matching processing unit 22 determines the degree of convulsion as a physical condition index according to the degree of match between the activity body temperature data indicating temperature fluctuations in the calf muscles and the specified pattern data. If the degree of match between the activity body temperature data indicating temperature fluctuations in the calf muscles and the pattern data is high, the degree of convulsion is high. If the degree of match between the activity body temperature data indicating temperature fluctuations in the calf muscles and the pattern data is sufficiently low, the degree of convulsion is low and it can be assumed that no convulsion has occurred. The matching processing unit 22 outputs the diagnosis result including the physical condition index indicating the degree of convulsion to the statistical processing unit 23 and the output device 53.

[0063] The statistical processing unit 23 acquires the activity body temperature data from the activity body temperature estimation unit 12, and acquires the diagnosis results including the physical condition index from the waveform processing unit 21 and the matching processing unit 22. The statistical processing unit 23 calculates other physical condition indexes based on at least one of the history of the activity body temperature data and the combination of the physical condition indexes calculated by the waveform processing unit 21 and the matching processing unit 22.

[0064] The statistical processing unit 23 generates a history of active body temperature data by storing the active body temperature data every time it is acquired from the active body temperature estimation unit 12. By performing statistical processing on the history of active body temperature data, the statistical processing unit 23 can determine whether a problematic pattern, for example, a periodic significant rise in stomach temperature during digestion, is occurring, and determine a physical condition index that indicates the presence or absence of a chronic stomach disease.

[0065] The statistical processing unit 23 determines other physical condition indices, for example, physical condition indices indicating the physical condition and health level of the target organism, based on the combination of physical condition indices determined by the waveform processing unit 21 and the matching processing unit 22. The statistical processing unit 23 outputs the diagnosis results including the determined physical condition indices to the output device 53.

[0066] As an example, the statistical processing unit 23 determines that the target organism is in good physical condition and healthy when the stomach health level determined by the waveform processing unit 21 is equal to or greater than a first threshold and the degree of convulsion determined by the matching processing unit 22 is equal to or less than a second threshold. The statistical processing unit 23 determines that the target organism is in poor physical condition and unhealthy when the stomach health level determined by the waveform processing unit 21 is less than the first threshold or the degree of convulsion determined by the matching processing unit 22 is higher than the second threshold. The first threshold is, for example, the median of the range of values ​​that the health level can take. The second threshold is, for example, the median of the range of values ​​that the degree of convulsion can take.

[0067] Furthermore, the statistical processing unit 23 determines the health level of the target organism based on the physical condition index determined by the waveform processing unit 21 and the matching processing unit 22. The health level is an index that becomes maximum when the stomach health level determined by the waveform processing unit 21 is at its upper limit and the degree of convulsion determined by the matching processing unit 22 is at its lower limit, and becomes minimum when the stomach health level determined by the waveform processing unit 21 is at its lower limit and the degree of convulsion determined by the matching processing unit 22 is at its upper limit.

[0068] The output device 53 outputs at least one of visual information and audio information indicating the diagnosis result received from the physical condition diagnosis section 13 and a control signal to an external device according to the diagnosis result. For example, the output device 53 is a display device having a screen 54 on which the diagnosis result is displayed, as shown in Fig. 11. A display area 54a of the screen 54 displays the stomach health level included in the diagnosis result of the waveform processing section 21, and a display area 54b of the screen 54 displays the stomach metabolic fluctuation included in the diagnosis result of the waveform processing section 21. A display area 54c of the screen 54 displays temperature information included in the diagnosis result of the waveform processing section 21, specifically, the armpit temperature.

[0069] Display area 54d of screen 54 displays the degree of convulsion included in the diagnosis result of matching processing unit 22. Display area 54e of screen 54 displays a determination of whether the target organism included in the diagnosis result of statistical processing unit 23 is healthy. Display area 54f of screen 54 displays the health level of the target organism included in the diagnosis result of statistical processing unit 23.

[0070] As described above, the waveform processing unit 21 and matching processing unit 22 included in the physical condition diagnosis unit 13 of the physical condition diagnosis device 2 according to embodiment 2 each calculate a physical condition index. The statistical processing unit 23 included in the physical condition diagnosis device 2 calculates another physical condition index based on at least one of a past active body temperature and a combination of the physical condition indexes calculated by the waveform processing unit 21 and the matching processing unit 22. The physical condition diagnosis device 2 can diagnose the physical condition of a target organism by calculating multiple physical condition indexes. Therefore, the physical condition diagnosis device 2 can accurately diagnose the physical condition of the target organism.

[0071] Since the statistical processing unit 23 can diagnose the physical condition based on past diagnostic results, it is possible to prevent incorrect diagnosis of the physical condition of the target organism based on temporary fluctuations in the measurement results of the temperature sensors 51a, 51b or the activity sensors 52a, 52b.

[0072] (Embodiment 3) The method of calculating the physical condition index is not limited to the above example, and the physical condition index may be calculated by applying activity body temperature data calculated based on variable factor information including environmental information indicating the state of the environment surrounding the target organism in addition to activity information to a learning model. The physical condition diagnosis device 3 according to embodiment 3 shown in Figure 12 calculates a physical condition index indicating the physical condition of the target organism from the measurement results of multiple temperature sensors 51a and 51b, activity indices generated by multiple activity sensors 52a, 52b, and 52c, and measurement results of the environmental sensor 55.

[0073] In addition to the configuration of the physical condition diagnosis device 2 according to embodiment 2, the physical condition diagnosis device 3 further includes a learning unit 14 that determines a physical condition model, which is a model for deriving a physical condition index from the activity body temperature data and the fluctuation factor information. The physical condition diagnosis unit 13 includes a waveform processing unit 21, a matching processing unit 22, a statistical processing unit 23, and a model diagnosis unit 24 that determines the physical condition index of the target organism by applying the activity body temperature data and the fluctuation factor information to the physical condition model determined by the learning unit 14.

[0074] The hardware configuration of the physical condition diagnosis device 3 according to embodiment 3 is similar to the hardware configuration of the physical condition diagnosis device 1 according to embodiment 1. However, the physical condition diagnosis device 3 is connected to temperature sensors 51 a and 51 b, activity sensors 52 a, 52 b, and 52 c, an environment sensor 55, and an output device 53 via an interface 63.

[0075] The information acquisition unit 11 included in the physical condition assessment device 3 acquires the measurement results of the temperature sensors 51 a, 51 b as temperature information, as in embodiment 2. The information acquisition unit 11 acquires, as activity information, activity indices generated by the activity sensors 52 a, 52 b and type information indicating the type of activity associated with the activity sensors 52 a, 52 b, as in embodiment 2. The information acquisition unit 11 acquires an activity index indicating a measured value of the amount of water consumed from the activity sensor 52 c, which is attached to a bottle used by the subject organism for physical condition assessment when drinking water and measures the amount of water consumed.

[0076] The information acquisition unit 11 acquires, as environmental information, measurement results of physical quantities indicating the environmental state around the target organism from the environmental sensor 55. The environmental sensor 55 is a sensor that measures brightness, temperature, humidity, odor, gas concentration, soil humidity, air pressure, radiation dose, magnetism, weather, dust, infrared rays, etc. In the third embodiment, the environmental sensor 55 measures temperature and humidity.

[0077] In the third embodiment, the temperature sensors 51 a and 51 b, the activity sensors 52 a and 52 b, and the environmental sensor 55 are realized as a wearable device 56 that can be worn by the target living thing. As an example, the target living thing wears the wearable device 56, such as a headset or eyeglass-type device, equipped with the temperature sensors 51 a and 51 b, the activity sensors 52 a and 52 b, and the environmental sensor 55, and the information acquisition unit 11 acquires the temperature information, the activity information, and the environmental information from the communication unit 57 of the wearable device 56. In detail, the communication unit 57 sends the temperature information acquired from the temperature sensors 51 a and 51 b, the activity information acquired from the activity sensors 52 a and 52 b, and the environmental information acquired from the environmental sensor 55 to the information acquisition unit 11 included in the physical condition diagnosis apparatus 3.

[0078] The active body temperature estimation unit 12 calculates the active body temperature data and also estimates the temperature of any part of the target organism. For example, the active body temperature estimation unit 12 estimates the brain temperature of the target organism based on temperature information, activity information, and environmental information, specifically, the armpit temperature, calf surface temperature, calorie intake, movement speed, amount of water consumed, and air temperature and humidity. The active body temperature estimation unit 12 sends the active body temperature data and estimation data indicating the estimated temperature to the physical condition diagnosis unit 13.

[0079] The waveform processing unit 21 included in the physical condition diagnosis unit 13 determines the degree of stomach health and stomach metabolic fluctuations, as in embodiment 1. The waveform processing unit 21 acquires temperature information from the information acquisition unit 11, and acquires active body temperature data and estimated data from the active body temperature estimation unit 12. The waveform processing unit 21 outputs diagnostic results including temperature information, for example, information on the armpit temperature of the subject organism, the degree of stomach health, stomach metabolic fluctuations, and estimated data, for example, estimated brain temperature, to the statistical processing unit 23 and the output device 53.

[0080] The matching processing unit 22 acquires temperature information from the information acquisition unit 11 and acquires activity body temperature data from the activity body temperature estimation unit 12. Similar to the second embodiment, the matching processing unit 22 determines the degree of convulsion. The matching processing unit 22 also determines the dehydration risk by performing pattern matching between a data pattern corresponding to the temperature and humidity included in the environmental information and a pattern of armpit temperature included in the temperature information. The matching processing unit 22 adjusts the dehydration risk based on at least one of the exercise intensity estimated from the speed of the target organism indicated by the activity sensor 52b and the amount of water intake of the target organism acquired from the activity sensor 52c. For example, high-intensity exercise in a hot place increases the risk of dehydration, while drinking water in a cold place reduces the risk of dehydration.

[0081] The matching processing unit 22 also determines the presence or absence of heatstroke symptoms by pattern matching a data pattern corresponding to the temperature and humidity included in the environmental information with a pattern of armpit temperature included in the temperature information. The matching processing unit 22 determines the risk of heatstroke from the dehydration risk, the armpit temperature included in the temperature information, and the amount of water intake of the target organism acquired from the activity sensor 52c. The matching processing unit 22 transmits diagnostic results including data on the amount of water intake and physical condition indicators indicating the risk of dehydration, the presence or absence of heatstroke symptoms, and the risk of heatstroke to the statistical processing unit 23 and the output device 53.

[0082] The learning unit 14 acquires fluctuation factor information from the information acquisition unit 11 and acquires activity body temperature data from the activity body temperature estimation unit 12. The learning unit 14 acquires physical condition indices from each of the waveform processing unit 21 and the matching processing unit 22. The learning unit 14 learns the correspondence between the activity body temperature data, fluctuation factor information, and physical condition indices.

[0083] The learning unit 14 generates a physical condition model, which is a neural network model for deriving a physical condition index of the target organism from the active body temperature data and the fluctuation factor information. Specifically, the learning unit 14 learns first learning data including the active body temperature data, the fluctuation factor information, and the physical condition index calculated by the waveform processing unit 21. The learning unit 14 uses the active body temperature data and the fluctuation factor information as input values ​​and the physical condition index calculated by the waveform processing unit 21, specifically, stomach health and gastric metabolic fluctuations, as output values ​​to generate a first physical condition model, which is a neural network model having an input layer, an intermediate layer, and an output layer. The first physical condition model is realized, for example, by a recurrent neural network (RNN), a long short-term memory (LSTM), a general-purpose artificial intelligence model, or the like. The learning unit 14 adjusts the weights between the input layer and the intermediate layer, the weights between the intermediate layers, and the weights between the intermediate layer and the output layer of the first physical condition model based on the first learning data.

[0084] The learning unit 14 learns second learning data including the activity body temperature data, the variation factor information, and the physical condition index determined by the matching processing unit 22. The learning unit 14 uses the activity body temperature data and the variation factor information as input values ​​and the physical condition index determined by the matching processing unit 22, specifically, the degree of convulsion, the presence or absence of heatstroke symptoms, and the risk of heatstroke, as output values ​​to generate a second physical condition model which is a neural network model having an input layer, an intermediate layer, and an output layer. The learning unit 14 adjusts the weights between the input layer and the intermediate layer, the weights between the intermediate layers, and the weights between the intermediate layer and the output layer for the second physical condition model based on the second learning data.

[0085] The learning unit 14 learns third learning data including the activity body temperature data, the fluctuation factor information, at least one of the physical condition indices determined by the waveform processing unit 21 and the matching processing unit 22, and the physical condition indices determined by the statistical processing unit 23. The learning unit 14 uses the activity body temperature data, the fluctuation factor information, and at least one of the physical condition indices determined by the waveform processing unit 21 and the matching processing unit 22 as input values, and uses the physical condition indices determined by the statistical processing unit 23, specifically, the physical condition and health level, as output values, to generate a third physical condition model which is a neural network model having an input layer, an intermediate layer, and an output layer. The learning unit 14 adjusts the weights between the input layer and the intermediate layer, the weights between the intermediate layers, and the weights between the intermediate layer and the output layer for the third physical condition model based on the third learning data.

[0086] The learning unit 14 sends a physical condition model including at least one of the first physical condition model, the second physical condition model, and the third physical condition model to the model diagnosis unit 24.

[0087] The model diagnosis unit 24 obtains a physical condition index of the target organism by applying the fluctuation factor information obtained from the information acquisition unit 11 and the active body temperature data obtained from the active body temperature estimation unit 12 to the physical condition model obtained from the learning unit 14. In detail, the model diagnosis unit 24 obtains a physical condition index, specifically, the health level of the stomach and the metabolic fluctuations of the stomach, by applying the active body temperature data and the fluctuation factor information to the first physical condition model.

[0088] The model diagnosis unit 24 applies the activity body temperature data and the fluctuation factor information to the second physical condition model to obtain physical condition indicators, specifically, the degree of convulsion, the presence or absence of heatstroke symptoms, and the risk of heatstroke.

[0089] The model diagnosis unit 24 obtains a physical condition index, specifically, the quality of physical condition and the degree of health, by applying the activity body temperature data, fluctuation factor information, and the physical condition index obtained by the waveform processing unit 21 and the matching processing unit 22 to the third physical condition model.

[0090] The output device 53 outputs at least one of visual information and audio information indicating the diagnosis result received from the physical condition diagnosis section 13 and a control signal to an external device according to the diagnosis result. For example, the output device 53 is a display device having a screen 54 on which the diagnosis result is displayed, as shown in Fig. 13. As in the first embodiment, the gastric health level included in the diagnosis result of the waveform processing section 21 is displayed in a display area 54a of the screen 54, and the gastric metabolic fluctuation included in the diagnosis result of the waveform processing section 21 is displayed in a display area 54b of the screen 54.

[0091] As in the second embodiment, the display area 54c of the screen 54 displays temperature information, specifically, the armpit temperature, included in the diagnosis result of the waveform processing unit 21. The display area 54d of the screen 54 displays the degree of convulsion included in the diagnosis result of the matching processing unit 22. The display area 54e of the screen 54 displays a determination of whether the target organism, included in the diagnosis result of the statistical processing unit 23, is healthy. The display area 54f of the screen 54 displays the health level of the target organism, included in the diagnosis result of the statistical processing unit 23.

[0092] A display area 54g of the screen 54 displays the presence or absence of heatstroke symptoms included in the diagnosis result of the matching processing unit 22. A display area 54h of the screen 54 displays the heatstroke risk included in the diagnosis result of the matching processing unit 22. A display area 54i of the screen 54 displays the amount of water intake included in the diagnosis result of the matching processing unit 22. A display area 54j of the screen 54 displays estimated data of brain temperature included in the diagnosis result of the waveform processing unit 21.

[0093] As explained above, the model diagnosis unit 24 included in the physical condition diagnosis device 3 according to the third embodiment obtains other physical condition indices by applying the active body temperature data and the fluctuation factor information, or the active body temperature data, the fluctuation factor information, and the physical condition indices, to the physical condition model, which is a neural network model. This allows the physical condition of the target organism to be diagnosed with high accuracy.

[0094] The present disclosure is not limited to the above-described embodiments. The embodiments may be combined as desired. For example, the physical condition assessment device 1 according to the first embodiment and the physical condition assessment device 2 according to the second embodiment may include a learning unit 14, as in the third embodiment.

[0095] The physical condition diagnosis device may acquire the physical condition model from an external device. The physical condition diagnosis system 101 shown in Fig. 14 includes temperature sensors 51a and 51b, activity sensors 52a, 52b, and 52c, an environmental sensor 55, a physical condition diagnosis device 4, an output device 53, and a model determination device 31 that determines a physical condition model to be used for physical condition diagnosis. The configuration of the physical condition diagnosis device 4 is the same as that of the physical condition diagnosis device 3, except that the learning unit 14 is removed. The model determination device 31 includes a physical condition model determination unit 32 that performs processing similar to that of the learning unit 14 included in the physical condition diagnosis device 3 according to the third embodiment.

[0096] The physical condition model determination unit 32 acquires fluctuation factor information from the information acquisition unit 11, acquires active body temperature data from the active body temperature estimation unit 12, and acquires physical condition indices from the waveform processing unit 21, matching processing unit 22, and statistical processing unit 23 of the physical condition diagnosis unit 13. Like the learning unit 14 provided in the physical condition diagnosis device 3 according to embodiment 3, the physical condition model determination unit 32 learns the correspondence between the active body temperature data, fluctuation factor information, and physical condition indices indicating the physical condition of the target organism, and determines a physical condition model for deriving the physical condition indices from the active body temperature data and fluctuation factor information. The physical condition model determination unit 32 sends the determined physical condition model to the model diagnosis unit 24 of the physical condition diagnosis unit 13.

[0097] The model diagnosis unit 24 acquires the physical condition model from the physical condition model determination unit 32. As in the third embodiment, the model diagnosis unit 24 applies the fluctuation factor information acquired from the information acquisition unit 11 and the active body temperature data acquired from the active body temperature estimation unit 12 to the physical condition model acquired from the physical condition model determination unit 32 to obtain a physical condition index of the target organism.

[0098] The model determination device 31 may determine a temperature model for deriving activity body temperature data from biological information and variation factor information in addition to the physical condition model. The physical condition diagnosis system 101 shown in Figure 15 includes temperature sensors 51a and 51b, activity sensors 52a, 52b, and 52c, an environmental sensor 55, a physical condition diagnosis device 5, an output device 53, and a model determination device 31 that determines a physical condition model used in physical condition diagnosis and a temperature model for deriving activity body temperature data. The configuration of the physical condition diagnosis device 5 is the same as that of the physical condition diagnosis device 3, except that the learning unit 14 is removed.

[0099] The temperature model determination unit 33 acquires biological information and variation factor information from the information acquisition unit 11 and acquires activity body temperature data from the activity body temperature estimation unit 12. The temperature model determination unit 33 learns the correspondence between biological information, variation factor information, and activity body temperature data. The temperature model determination unit 33 determines a temperature model, which is a neural network model that receives biological information and variation factor information as input and outputs activity body temperature data. Specifically, the temperature model determination unit 33 learns temperature learning data consisting of biological information, variation factor information, and activity body temperature data, and generates a temperature model, which is a neural network model having an input layer, an intermediate layer, and an output layer, using the biological information and variation factor information as input values ​​and the activity body temperature data as output values. The temperature model determination unit 33 adjusts the weights between the input layer and the intermediate layer, the weights between the intermediate layers, and the weights between the intermediate layer and the output layer of the temperature model based on the temperature learning data. The temperature model determination unit 33 sends the determined temperature model to the activity body temperature estimation unit 12.

[0100] The active body temperature estimation unit 12 acquires a temperature model from the temperature model determination unit 33. The active body temperature estimation unit 12 obtains active body temperature data by applying the biological information and fluctuation factor information acquired from the information acquisition unit 11 to the temperature model acquired from the temperature model determination unit 33. The active body temperature estimation unit 12 sends the obtained active body temperature data to the physical condition diagnosis unit 13.

[0101] The learning unit 14 included in the physical condition diagnosis device 3 may obtain a temperature model and obtain active body temperature data based on the obtained temperature model, similar to the temperature model determination unit 33. In this case, the active body temperature estimation unit 12 obtains active body temperature data by applying the biological information and fluctuation factor information acquired from the information acquisition unit 11 to the temperature model acquired from the learning unit 14.

[0102] The temperature model may be a neural network model that receives biological information, fluctuation factor information, and environmental information as inputs and outputs activity body temperature data.

[0103] The target organism may be any organism, such as a human, livestock, pet, etc. The activity of the target organism is not limited to activities that promote heat production, such as eating and exercise, as long as it causes a temperature change in a part of the target organism, but also includes activities that promote a decrease in body temperature, such as sleep and meditation.

[0104] The information acquisition unit 11 may acquire biological information from data input to the physical condition assessment device 1-5. As an example, the information acquisition unit 11 may acquire information on the blood glucose level of the target organism as biological information. Fluctuations in blood glucose levels that occur in cycles shorter than eight hours can be considered to be fluctuations caused by eating and digestion. In this case, the physical condition assessment unit 13 can determine whether there is a problem with nutrient absorption capacity, for example, from fluctuations in blood glucose levels and fluctuations in active body temperature, which corresponds to stomach temperature.

[0105] The information acquisition unit 11 may acquire activity information from data input to the physical condition diagnosis device 1-5. As an example, the information acquisition unit 11 may acquire type information indicating the state of activity from the behavior schedule of the target organism.

[0106] The information acquisition unit 11 may acquire measurement results from a temperature sensor located near an artery and a temperature sensor located near a vein. Temperature changes occurring upstream of the blood flow appear as a phase difference between the measurement results of the temperature sensor located near the artery and the measurement results of the temperature sensor located near the vein. In this case, the activity body temperature estimation unit 12 may calculate activity body temperature data from the phase difference. In particular, the activity body temperature estimation unit 12 can obtain temperature information excluding the influence of blood flow by removing the component of the activity body temperature data calculated from the phase difference from the temperature information.

[0107] The information acquisition unit 11 may acquire an index indicating exercise intensity as activity information. If the target organism is healthy, the amplitude of the activity body temperature data increases as the exercise intensity increases, and the amplitude of the activity body temperature data decreases as the exercise intensity decreases. In this case, the physical condition diagnosis unit 13 may determine the target organism's physical condition index, for example, health level, based on whether the exercise intensity and the amplitude of the activity body temperature data are linked. In particular, the physical condition diagnosis unit 13 may determine the target organism's health level based on the phase difference, frequency spectrum deviation, amplitude deviation, etc. between the waveform data of the index indicating exercise intensity and the waveform data of the activity body temperature data.

[0108] The method of obtaining the activity body temperature data by the activity body temperature estimation unit 12 is not limited to the above example. As an example, the activity body temperature estimation unit 12 may compare the armpit temperature with a waveform pattern determined according to temperature fluctuations during meals, and extract the armpit temperature waveform that matches the waveform pattern as the activity body temperature data.

[0109] The activity-related body temperature estimation unit 12 may extract frequency components corresponding to the period during which activity is being performed from the frequency-domain data, and perform IFFT on the extracted frequency components to obtain activity-related body temperature data in the time domain.

[0110] The active body temperature estimation unit 12 may remove the cyclical fluctuation component by subtracting a reference waveform corresponding to the cyclical fluctuation component from the waveform data indicated by the temperature information. As another example, the active body temperature estimation unit 12 may obtain the cyclical fluctuation component corresponding to the circadian by a cosinor method, which calculates the parameters of an approximate cosine wave from the armpit temperature using the least squares method.

[0111] The activity body temperature estimation unit 12 may calculate the variation from a reference value, such as the most recent measurement value or the average value of measurement values ​​over a recent fixed period, specifically the variation in armpit temperature from the reference value, as activity body temperature data.

[0112] The heat transfer path shown in Figure 10 is an example, and the active body temperature estimation unit 12 may estimate the active body temperature based on the fact that the influence of heat exchange with the outside is dominant in areas close to the surface of the target organism's body, that the influence of heat exchange with adjacent organs is dominant, and that the degree of influence from the upstream side closer to the heart varies depending on the thickness of the blood vessels.

[0113] The physical condition diagnosis unit 13 is not limited to the above example. As an example, the physical condition diagnosis unit 13 may acquire multiple pieces of activity body temperature data that are different in at least one of target body parts and fluctuation factors, and diagnose the physical condition of the target organism in the time domain based on the correlation between waveform data of the multiple pieces of activity body temperature data.

[0114] The physical condition diagnosis unit 13 may output a diagnosis result including at least one of biological information, fluctuation factor information including activity information and environmental information, and active body temperature data, and a physical condition index. As an example, if the physical condition index indicating liver function obtained by the waveform processing unit 21 indicates a decline in liver function, the physical condition diagnosis unit 13 may output a diagnosis result including the physical condition index indicating liver function, active body temperature data of organs such as the liver and bladder, and estimated data indicating the temperatures of organs such as the liver and bladder.

[0115] The waveform processing unit 21 may diagnose the physical condition of the target organism based on the frequency domain data obtained by performing IFFT on the activity body temperature data, and based on whether the frequency component that takes a peak value is within a reference range, whether a peak value exists, whether the amplitude of a specified frequency component is within a target range, etc.

[0116] The waveform processing unit 21 may diagnose that an abnormality has occurred in the organ corresponding to the activity when the amplitude of the waveform data of the activity body temperature data is not within a target amplitude range based on the identification information corresponding to the activity body temperature data.

[0117] The waveform processing unit 21 can diagnose the physical condition of the target organism based on activity body temperature data corresponding to an arbitrarily determined period, such as a fixed period from midnight to midnight, a period corresponding to activity information, or the period from the rise of the waveform data of the activity body temperature data to the next rise.

[0118] The waveform processing unit 21 may acquire multiple types of activity body temperature data and determine a physical condition index of the target organism based on at least one of the phase, amplitude, frequency spectrum, and waveform of the waveform data of each activity body temperature data. In this case, the waveform processing unit 21 may output a diagnosis result including multiple physical condition indexes to the statistical processing unit 23, and the statistical processing unit 23 may determine other physical condition indexes of the target organism based on the multiple physical condition indexes determined by the waveform processing unit 21.

[0119] As an example, the waveform processing unit 21 may calculate a physical condition index of the target organism based on the frequency spectrum of the activity body temperature data related to eating and the activity body temperature data related to excretion. In this case, the statistical processing unit 23 can calculate a physical condition index indicating the health of an organ related to both eating and excretion, or an organ related to only eating or excretion, from the physical condition index related to eating and the physical condition index related to excretion.

[0120] The matching processing unit 22 may determine a physical condition index from the activity body temperature data and the fluctuation factor information. As an example, the matching processing unit 22 may determine the severity of a cramp by performing a pattern matching process between the activity body temperature data indicating temperature fluctuations in the calf muscles determined by the activity body temperature estimation unit 12 and pattern data indicating temperature fluctuations during a specified cramp, and may also determine the cause of the cramp from the fluctuation factor information. As an example, when the fluctuation factor information indicates that no exercise has been performed in the recent period, the cause of the cramp can be considered to be stiffness due to lack of exercise. As another example, when the fluctuation factor information indicates that exercise has been performed in the recent period, the cause of the cramp can be considered to be dehydration. The matching processing unit 22 outputs a diagnosis result including a physical condition index indicating the severity of the cramp and the cause of the cramp to the output device 53.

[0121] The statistical processing unit 23 may combine physical condition indices related to each organ, and estimate the disease that the target organism is suffering from from the combination of disorders appearing in each organ.

[0122] The statistical processing unit 23 may calculate a physical condition index by combining the activity body temperature data and the fluctuation factor information. For example, when the activity body temperature data for a recent fixed period shows an upward trend and the fluctuation factor information indicates that no bowel movements have occurred in the recent fixed period, the statistical processing unit 23 may calculate a physical condition index indicating constipation.

[0123] The method for generating a physical condition model by the learning unit 14 is not limited to the above example. The learning unit 14 may learn by associating the activity body temperature information, activity information, the physical condition index determined by the matching processing unit 22, and the physical condition index determined by the waveform processing unit 21, and may obtain a first physical condition model that receives the activity body temperature information, activity information, and the physical condition index determined by the matching processing unit 22 as inputs and outputs the physical condition index determined by the waveform processing unit 21.

[0124] The learning unit 14 may generate the physical condition model and the temperature model by performing supervised learning based on multidimensional function fitting. Supervised learning means that a large amount of learning data, which is a data set of inputs and results, is provided to a learning device, and the learning device learns features in the large data set and generates a model that estimates results from inputs.

[0125] Similarly, the physical condition model determination unit 32 and the temperature model determination unit 33 may perform supervised learning based on multidimensional function fitting to generate the physical condition model and the temperature model, respectively.

[0126] The diagnosis results sent from the physical condition diagnosis unit 13 to the output device 53 may include at least one of biological information and variation factor information, or activity body temperature data, in addition to the physical condition indicators calculated by each unit of the physical condition diagnosis unit 13. The variation factor information may include both activity information and environmental information, or may include only one of activity information and environmental information. The activity information included in the variation factor information may include both type information and activity indicators, or may include only one of type information and activity indicators.

[0127] The statistical processing unit 23 may store each physical condition index obtained from the waveform processing unit 21 and the matching processing unit 22, and may determine other physical condition indexes of the target organism from the previously determined physical condition indexes. As one example, the statistical processing unit 23 may determine that there is a possibility of PTSD if the degree of convulsions determined by the matching processing unit 22 is high throughout the year regardless of the season, and the frequency of high convulsions is equal to or greater than a threshold frequency. As another example, the statistical processing unit 23 may determine that there is a possibility of cold sensitivity rather than PTSD if the timing of high convulsions determined by the matching processing unit 22 is concentrated in winter.

[0128] The configuration of the physical condition diagnosis unit 13 is not limited to the above example, and may be any configuration that can determine the physical condition index of the target organism. As one example, the physical condition diagnosis unit 13 may have only the matching processing unit 22. As another example, the physical condition diagnosis unit 13 may have the matching processing unit 22 and the model diagnosis unit 24.

[0129] The output device 53 may output a control signal to an external device according to the diagnosis result. For example, when the target organism is a livestock, the output device 53 outputs a control signal according to the diagnosis result to external devices such as devices for adjusting the environment of the livestock barn, such as an air conditioner, a fan, a feeding device, etc.

[0130] As another example, when the target living being is a human exercising using exercise equipment, output device 53 outputs a control signal to the exercise equipment according to a physical condition index included in the diagnosis result. For example, when the diagnosis result indicates that the health level, which is an example of a physical condition index, is declining, output device 53 transmits a control signal to the exercise equipment to reduce the exercise load.

[0131] As another example, when the target organism is a human, the output device 53 outputs a control signal according to the diagnosis result to an ordering device that orders medicines, supplements, etc. For example, when the diagnosis result indicates the possibility of a disease, the physical condition diagnosis unit 13 sends a control signal to the ordering device instructing it to order medicines, supplements, etc. according to the disease.

[0132] Implementation examples of the physical condition diagnosis systems 100, 101 are not limited to the above examples, and any implementation is possible as long as it is possible to diagnose the physical condition of a target organism. As an example, part or all of the physical condition diagnosis device 3 shown in Fig. 12 may be mounted on a wearable device 56 together with temperature sensors 51a, 51b, activity sensors 52a, 52b, and an environmental sensor 55. Similarly, part or all of the physical condition diagnosis devices 1, 2, 4, and 5 may also be mounted on a wearable device 56.

[0133] The wearable device 56 may transmit and receive information to and from the physical condition assessment apparatus 3. In particular, the wearable device 56 may acquire a physical condition index from the physical condition assessment unit 13. A communication unit 57 included in the wearable device 56 shown in Fig. 16 acquires the physical condition index from the physical condition assessment unit 13 included in the physical condition assessment apparatus 3. The wearable device 56 has an equipment control unit 58 that controls the exercise equipment used by the target living thing in accordance with the physical condition index received by the communication unit 57.

[0134] In detail, the communication unit 57 receives the physical condition index from at least one of the waveform processing unit 21, the matching processing unit 22, the statistical processing unit 23, and the model diagnosis unit 24 included in the physical condition diagnosis unit 13. The communication unit 57 sends the received physical condition index to the equipment control unit 58. The equipment control unit 58 controls the exercise equipment based on the exercise load corresponding to the physical condition index calculated by the physical condition diagnosis unit 13 included in the physical condition diagnosis device 3 and received by the communication unit 57 for the target organism exercising using the exercise equipment. For example, when the health level, which is an example of the physical condition index received by the communication unit 57, decreases, the equipment control unit 58 transmits a control signal to the exercise equipment to reduce the exercise load.

[0135] 17 has an output unit 59 that outputs information about exercise load corresponding to the physical condition index received by the communication unit 57 by screen display, audio, etc. For example, the output unit 59 suggests a lower exercise load when the health level declines, and suggests a higher exercise load when the health level improves. The exercise load may be, for example, the running pace, gradient, weight of exercise equipment, etc.

[0136] 16 and 17 may acquire the diagnosis result from the output device 53. Alternatively, the output device 53 may be implemented in the wearable device 56.

[0137] The above-described hardware configuration and flowchart are merely examples, and can be changed and modified as desired. The core control processing section, including the processor 61, memory 62, and interface 63, can be implemented using a standard computer system rather than a dedicated system. For example, a computer program for executing the above-described operations may be stored and distributed on a computer-readable recording medium (such as a flexible disk, a CD-ROM (Compact Disc-Read Only Memory), or a DVD-ROM (Digital Versatile Disc-Read Only Memory)), and the computer program may be installed on a computer to configure the physical condition assessment device 1-5 that executes the above-described processes. Alternatively, the computer program may be stored in a storage device of a server device on a communication network, and downloaded by a standard computer system to configure the physical condition assessment device 1-5.

[0138] When the functions of the physical condition diagnosis device 1-5 are realized by sharing the functions between an OS (Operating System) and an application program, or by cooperation between the OS and the application program, only the application program portion may be stored in a recording medium or storage device.

[0139] It is also possible to superimpose a computer program on a carrier wave and distribute it via a communication network. For example, the computer program may be posted on a bulletin board system (BBS) on the communication network and distributed via the communication network. The computer program may then be started and executed under the control of an OS in the same way as other application programs, thereby performing the above-described processing.

[0140] The hardware configuration of the physical condition diagnosis device 1-5 is not limited to the above example. The physical condition diagnosis device 1-5 may be realized by a processing circuit 64, as shown in FIG. 18 . The processing circuit 64 is connected to a temperature sensor 51, an activity sensor 52, an output device 53, and the like via an interface circuit 65. When the processing circuit 64 is dedicated hardware, the processing circuit 64 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. Each component of the physical condition diagnosis device 1-5 may be realized by a separate processing circuit 64, or all components of the physical condition diagnosis device 1-5 may be realized by a common processing circuit 64.

[0141] Some of the functions of the physical condition assessment device 1-5 may be implemented by dedicated hardware, and other functions may be implemented by software or firmware. For example, the information acquisition unit 11 may be implemented by a processing circuit 64 shown in Fig. 18, and the active body temperature estimation unit 12 and the physical condition assessment unit 13 may be implemented by a processor 61 shown in Fig. 2 reading and executing a program stored in a memory 62.

[0142] The present disclosure allows various embodiments and modifications without departing from the broad spirit and scope of the present disclosure. Furthermore, the above-described embodiments are intended to illustrate the present disclosure and do not limit the scope of the present disclosure. That is, the scope of the present disclosure is defined by the claims, not the embodiments. Various modifications made within the scope of the claims and the meaning of equivalent disclosures are considered to be within the scope of the present disclosure.

[0143] 1, 2, 3, 4, 5 Physical condition diagnosis device, 11 Information acquisition unit, 12 Activity body temperature estimation unit, 13 Physical condition diagnosis unit, 14 Learning unit, 21 Waveform processing unit, 22 Matching processing unit, 23 Statistical processing unit, 24 Model diagnosis unit, 31 Model determination device, 32 Physical condition model determination unit, 33 Temperature model determination unit, 51, 51a, 51b Temperature sensor, 52, 52a, 52b, 52c Activity sensor, 53 Output device, 54 Screen, 54a, 54b, 54c, 54d, 54e, 54f, 54g, 54h, 54i, 54j Display area, 55 Environmental sensor, 56 Wearable device, 57 Communication unit, 58 Equipment control unit, 59 Output unit, 60 Bus, 61 Processor, 62 Memory, 63 Interface, 64 Processing circuit, 65 Interface circuit, 81, 82 Parts, 91, 92 Heat source, 100, 101 Physical condition diagnosis system.

Claims

1. A physical condition diagnosis device comprising: an information acquisition unit that acquires bioinformation including temperature information of at least one part of a target organism that is the subject of physical condition diagnosis, and fluctuation factor information including activity information indicating the activity of the target organism that causes the temperature of the part to fluctuate; an activity body temperature estimation unit that calculates activity body temperature data indicating the temperature fluctuation of the part due to the activity from the bioinformation and the fluctuation factor information; and a physical condition diagnosis unit that calculates a physical condition index indicating the physical condition of the target organism based on the activity body temperature data.

2. The physical condition diagnostic device according to claim 1, wherein the information acquisition unit acquires the activity information including an activity index that indicates the activity of the target organism in numerical form.

3. The physical condition diagnosis device according to claim 1 or 2, wherein the information acquisition unit acquires the activity information including type information indicating a type of the activity of the target organism.

4. A physical condition diagnosis device as claimed in any one of claims 1 to 3, wherein the information acquisition unit acquires the variable factor information including the activity information and environmental information indicating the environmental state around the target organism.

5. A physical condition diagnostic device as described in any one of claims 1 to 4, wherein the activity body temperature estimation unit calculates the activity body temperature data based on the biological information, the variable factor information, and a heat transfer pathway within the body of the target organism that is determined according to heat production and heat dissipation within the body of the target organism.

6. A physical condition diagnosis device as described in any one of claims 1 to 5, wherein the activity body temperature estimation unit acquires a temperature model, which is a model obtained from the correspondence between the biological information, the fluctuation factor information, and the activity body temperature data, for deriving the activity body temperature data from the biological information and the fluctuation factor information, and applies the biological information and the fluctuation factor information acquired by the information acquisition unit to the temperature model to obtain the activity body temperature data indicating the temperature fluctuation of the part caused by the activity indicated by the activity information included in the fluctuation factor information.

7. A physical condition diagnosis device as claimed in any one of claims 1 to 6, wherein the activity body temperature estimation unit determines a plurality of activity body temperature data, each of which corresponds to at least one of the body part and the activity, different from each other from the biological information and the variable factor information.

8. The physical condition diagnostic device according to claim 7, wherein the physical condition diagnostic section determines the physical condition index based on a plurality of pieces of activity body temperature data.

9. A physical condition diagnostic device according to any one of claims 1 to 8, wherein the physical condition diagnostic unit determines the physical condition index indicating at least one of the good or bad physical condition of the target organism, the presence or absence of a disease of the target organism, and the degree of health of the target organism expressed as a numerical value.

10. A physical condition diagnostic device according to any one of claims 1 to 9, wherein the physical condition diagnostic unit determines the physical condition index based on at least one of the biological information and the variable factor information, and the active body temperature data.

11. A physical condition diagnostic device according to any one of claims 1 to 10, wherein the physical condition diagnostic unit has a waveform processing unit that determines the physical condition index based on at least one of the phase characteristics, amplitude, frequency spectrum, and waveform of the waveform data of the activity body temperature data.

12. A physical condition diagnostic device as claimed in any one of claims 1 to 11, wherein the physical condition diagnostic unit has a matching processing unit that determines the physical condition index by performing pattern matching between a pattern determined according to the physical condition of the target organism and the activity body temperature data.

13. The physical condition diagnostic device according to claim 11, wherein the physical condition diagnostic unit has a matching processing unit that determines the physical condition index by performing pattern matching between a pattern determined according to the physical condition of the target organism and the activity body temperature data, and the physical condition diagnostic unit has a statistical processing unit that determines other physical condition indexes based on the history of the activity body temperature data and at least one of the combinations of the physical condition indexes determined by the waveform processing unit and the matching processing unit.

14. A physical condition diagnostic device as described in any one of claims 1 to 13, wherein the physical condition diagnostic unit outputs a physical condition diagnostic result of the target organism including at least one of the biometric information, the fluctuation factor information, and the active body temperature data, and the physical condition index.

15. A physical condition diagnosis device as described in any one of claims 1 to 14, wherein the physical condition diagnosis unit further comprises a model diagnosis unit which obtains a physical condition model for deriving the physical condition index from the activity body temperature data and the fluctuation factor information, which is a model obtained from a correspondence with the activity body temperature data, the fluctuation factor information, and the physical condition index, and obtains the physical condition index of the target organism by applying the activity body temperature data obtained by the activity body temperature estimation unit and the fluctuation factor information obtained by the information acquisition unit to the physical condition model.

16. A physical condition diagnosis device as described in any one of claims 1 to 15, further comprising a learning unit that determines a physical condition model based on the correspondence between the activity body temperature data, the fluctuation factor information, and the physical condition index, for deriving the physical condition index from the activity body temperature data and the fluctuation factor information.

17. A wearable device which can be worn by a target living organism that is the subject of a physical condition diagnosis, and which transmits and receives information to and from a physical condition diagnosis device described in any one of claims 1 to 16, comprising: a temperature sensor which measures the temperature of at least one part of the target living organism; an activity sensor which generates an activity index which numerically indicates the activity of the target living organism that causes the temperature of the part to change; a communication unit which transmits the measurement value of the temperature sensor and the activity index generated by the activity sensor to the physical condition diagnosis device, and receives from the physical condition diagnosis device the physical condition index determined by the physical condition diagnosis unit provided in the physical condition diagnosis device; and an equipment control unit which controls the exercise equipment based on the exercise load corresponding to the physical condition index determined by the physical condition diagnosis unit provided in the physical condition diagnosis device and received by the communication unit, for the target living organism exercising using the exercise equipment.

18. A wearable device which can be worn by a target living organism that is the subject of a physical condition diagnosis, and which transmits and receives information to and from a physical condition diagnosis device described in any one of claims 1 to 16, comprising: a temperature sensor which measures the temperature of at least one part of the target living organism; an activity sensor which generates an activity index which numerically indicates the activity of the target living organism that causes the temperature of the part to change; a communication unit which transmits the measurement value of the temperature sensor and the activity index generated by the activity sensor to the physical condition diagnosis device, and receives from the physical condition diagnosis device the physical condition index determined by the physical condition diagnosis unit provided in the physical condition diagnosis device; and an output unit which outputs information about the exercise load of the target living organism that is exercising, corresponding to the physical condition index determined by the physical condition diagnosis unit provided in the physical condition diagnosis device and received by the communication unit.

19. A physical condition diagnosis system comprising: a temperature sensor that measures the temperature of at least one part of a target organism that is the subject of a physical condition diagnosis; and a physical condition diagnosis device described in any one of claims 1 to 16, wherein the information acquisition unit of the physical condition diagnosis device acquires the measurement result of the temperature sensor as the temperature information.

20. The physical condition diagnosis system of claim 19, further comprising an activity sensor that generates an activity index that numerically indicates the activity of the target organism that causes the temperature of the area to fluctuate, and the information acquisition unit of the physical condition diagnosis device acquires the activity information including type information indicating the type of activity associated with the activity sensor and the activity index generated by the activity sensor.

21. A health diagnosis system as described in claim 19 or 20, further comprising an environmental sensor that measures a physical quantity indicating the environmental state surrounding the target organism, and the information acquisition unit provided in the health diagnosis device acquires the variable factor information including the measurement results of the environmental sensor.

22. A health diagnosis system as claimed in any one of claims 19 to 21, further comprising an output device that acquires the health index determined by the health diagnosis unit provided in the health diagnosis device, and outputs at least one of visual information and audio information indicating the acquired health index, and a control signal to an external device according to the health index.

23. A physical condition diagnosis system comprising: a physical condition diagnosis device according to any one of claims 1 to 16; and a wearable device according to claim 17 or 18, wherein the information acquisition unit of the physical condition diagnosis device acquires the measurement result of the temperature sensor of the wearable device as the temperature information, and acquires the activity information including type information indicating the type of activity associated with the activity sensor of the wearable device and the activity index generated by the activity sensor.

24. A model determination device comprising: a temperature model determination unit that learns the correspondence between biological information including temperature information of at least one part of a target organism that is the subject of physical condition diagnosis, fluctuation factor information including activity information indicating the activity of the target organism that fluctuates the temperature at the part, and activity body temperature data indicating the temperature fluctuation of the part due to the activity, and determines a temperature model for deriving the activity body temperature data from the biological information and the fluctuation factor information.

25. The model determination device described in claim 24, further comprising a physical condition model determination unit that learns the correspondence between the activity body temperature data, the fluctuation factor information, and a physical condition index indicating the physical condition of the target organism, and determines a physical condition model for deriving the physical condition index from the activity body temperature data and the fluctuation factor information.

26. A method for diagnosing physical condition, comprising: acquiring biological information including temperature information of at least one part of a target organism that is the subject of physical condition diagnosis, and variable factor information including activity information indicating the activity of the target organism that causes the temperature of said part to fluctuate; determining activity body temperature data indicating the temperature fluctuation of said part caused by said variable factor from said biological information and said variable factor information; and determining a physical condition index indicating the physical condition of the target organism based on said activity body temperature data.

27. A program that causes a computer to function as: an information acquisition unit that acquires biological information including temperature information of at least one part of a target organism that is the subject of a health diagnosis, and fluctuation factor information including activity information indicating the activity of the target organism that causes the temperature of the part to fluctuate; an activity body temperature estimation unit that calculates activity body temperature data indicating the temperature fluctuation of the part caused by the activity from the biological information and the fluctuation factor information; and a health diagnosis unit that calculates a health condition index indicating the health condition of the target organism based on the activity body temperature data.

Images