Bioinformation evaluation device and method for evaluating bioinformation

Abstract

A biological information evaluation device according to an embodiment of the present disclosure is provided with a signal processing circuit that generates a feature value of an observed waveform for each observation data set based on a plurality of partial observation data sets that are included in an observation data set obtained by in vivo observation for a predetermined period and have a period shorter than the observation period of the observation data. The observation object of the biological information evaluation device is, for example, a human or an animal.

Description

Technical Field

[0001] The present disclosure relates to a biological information evaluation device and a method of evaluating biological information. Background Art

[0002] Generally, it takes time and effort to establish a method for controlling a person in a desired state or increasing the number of samples in order to obtain high-quality biological information about a person in order to understand the desired state. However, in the field of psychology, it is considered necessary to collect a relatively large number of samples and also to increase the measurement period in order to show the difference between states (or between groups) in a significant manner (see, for example, non-patent document 1). In addition to humans, the objects from which biological information is obtained may also include animals, etc.

[0003] Reference List

[0004] Non-patent literature

[0005] Non-patent literature 1: European Child and Adolescent Psychiatry (2017) 26: 1511-1522 Summary of the invention

[0006] Therefore, it is extremely difficult to establish a technology that meets the application demand of understanding the difference between states. It is desirable to provide a biological information evaluation device and a method of evaluating biological information that make it possible to understand the difference between states with a small number of samples and a short measurement period.

[0007] According to an embodiment of the present disclosure, a biological information evaluation device includes a signal processing circuit that generates a characteristic value of an observed object waveform for each of the observation data based on a plurality of partial observation data contained in each of the observation data obtained through living body observation for a predetermined period of time and each having an observation period shorter than the observation period of the observation data.

[0008] According to an embodiment of the present disclosure, a method for evaluating biological information uses a signal processing circuit to perform signal processing, which includes generating a characteristic value of an observed object waveform for each of the observation data based on a plurality of partial observation data contained in each of the observation data obtained through living body observation for a predetermined period of time and each having an observation period shorter than the observation period of the observation data.

[0009] In the biological information evaluation device and the method for evaluating biological information according to the embodiment of the present disclosure, a feature value of the waveform of the observation object is generated for each of the observation data based on a plurality of partial observation data contained in each of the observation data. This makes it possible to show the difference between the states in a robust manner.

[0010] The biological information evaluation device and the method for evaluating biological information according to the embodiments of the present disclosure make full use of biological information that can see periodic changes, which makes it possible to understand the difference between states when the number of samples is small and the measurement period is short. It should be noted that the effects described in the present disclosure are not necessarily limited to those described here. Any effect described in this specification can be adopted. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 is a diagram illustrating one example of a schematic configuration of a biological information evaluation device according to the first embodiment of the present disclosure.

[0012] Figure 2 yes Figure 1 Schematic diagram of an example of a process of calculating a feature value and an evaluation value in a biological information evaluation device in.

[0013] Figure 3 yes Figure 1 A flowchart of an example of a process of calculating a feature value and an evaluation value in a biological information evaluation device in FIG.

[0014] Figure 4 is a schematic diagram of an example of a process of calculating a characteristic value and an evaluation value according to a comparative example.

[0015] Figure 5 is a flowchart of an example of a process of calculating a feature value and an evaluation value according to a comparative example.

[0016] Figure 6 yes Figure 3 Schematic diagram of a modified example of the calculation process in .

[0017] Figure 7 It is a graphic Figure 1 FIG. 5 is a diagram of a modified example of the schematic configuration of the biological information evaluation device.

[0018] Figure 8 yes Figure 7 Schematic diagram of an example of a process of calculating a feature value and an evaluation value in a biological information evaluation device in.

[0019] Fig. 9 is Figure 7 A flowchart of an example of a process of calculating a feature value and an evaluation value in a biological information evaluation device in FIG.

[0020] Fig.10 is Figure 7 Schematic diagram of a modified example of a process of calculating a feature value and an evaluation value in a biological information evaluation device in .

[0021] Fig.11 It is a graphic Figure 1 FIG. 5 is a diagram of a modified example of the schematic configuration of the biological information evaluation device.

[0022] Fig.12 yes Fig.11 Schematic diagram of an example of a learning process of a learning model in a biological information evaluation device.

[0023] Fig.13 is Fig.11 Schematic diagram of an example of a process of evaluating a state using a learning model in a bioinformation evaluation device.

[0024] Fig.14 yes Fig.11 Schematic diagram of a modified example of the learning process of the learning model in the biological information evaluation device.

[0025] Fig.15 is Fig.11 Schematic diagram of a modified example of a process for evaluating a state using a learning model in a biological information evaluation device.

[0026] Fig.16 is a diagram illustrating an example of the relationship between the length of the divided period ΔW and the accuracy of the estimation.

[0027] Fig.17 It is shown in Fig.16 Graph of an example of the accuracy of estimation in a case where the length ΔW of the divided period is ΔW2.

[0028] Fig.18 : is a diagram illustrating an example of a schematic configuration of a biological information evaluation device according to the second embodiment of the present disclosure.

[0029] Fig.19 It is a graphic Fig.18 FIG. 5 is a diagram of a modified example of the schematic configuration of the biological information evaluation device.

[0030] Fig. 20 It is a graphic Fig.18 FIG. 5 is a diagram of a modified example of the schematic configuration of the biological information evaluation device. DETAILED DESCRIPTION

[0031] Hereinafter, the mode for carrying out the present disclosure will be described in detail with reference to the accompanying drawings. Note that the description will be given in the following order.

[0032] 1. First Embodiment

[0033] 2. Modification of the First Embodiment

[0034] 3. Second embodiment

[0035] <1. First Embodiment>

[0036] [Configuration]

[0037] A biological information evaluation apparatus 1 according to a first embodiment of the present disclosure will be described. Figure 1 1 is a diagram illustrating an example of a schematic configuration of a biological information evaluation device 1. The biological information evaluation device 1 is a device that detects biological information related to a living body 100. In addition, the biological information evaluation device 1 is a device that generates a feature value of an observation object waveform contained in the detected biological information, and also generates an evaluation value related to the difference in the observation object waveform between the detected pieces of biological information based on the generated feature value. The biological information includes, for example, brain waves, pulses, and sweating. The living body 100 is typically a human being, and may be an animal. The biological information evaluation device 1 is, for example, a wearable device such as a head-mounted display.

[0038] The biological information evaluation apparatus 1 includes, for example, a detector 10 , an ADC (Analog-to-Digital Converter), an input unit 30 , a storage unit 40 , a signal processing circuit 50 , a display controller 60 , and a display 70 .

[0039] The detector 10 detects biological information about the living body 100, and outputs the detected biological information as an analog signal to the ADC 20. The detector 10 obtains analog observation data 44 by observing the living body for a predetermined period (an observation period T to be described later), and outputs the obtained observation data 44 to the ADC 20. The detector 10 includes, for example, a pair of electrodes in contact with the living body 100, and a detection circuit that outputs an analog signal corresponding to a potential difference across the pair of electrodes. Note that the detector 10 may include, for example, a light source that outputs light onto the living body 100, and a light receiving circuit that detects light reflected and / or scattered within the living body 100 from the light output from the light source, and outputs an analog signal corresponding to the detected light.

[0040] The ADC 20 converts the analog observation data 44 (analog signal) input from the detector 10 into a digital signal, and outputs the digital observation data 44 (digital signal) obtained by the conversion to the signal processing circuit 50. The input unit 30 includes a device that allows a user to perform an input operation, and includes, for example, a touch screen, a button, a microphone, a switch, and a joystick. For example, the input unit 30 receives an input of a setting value 42, which will be described later.

[0041] The storage unit 40 holds a program (e.g., a processing program 41) executed by the signal processing circuit 50, a length 42 of the divided period ΔT, and a length 43 of the overlap period Δd1. The processing program 41 is a program for obtaining the observation data 44 from the detector 10 or performing predetermined signal processing on the obtained observation data 44. Here, the "predetermined processing" includes, for example, the following processing: for each observation data 44, a feature value of the observation object waveform is generated based on the obtained observation data 44, and an evaluation value about the difference of the observation object waveform between the observation data 44 is generated using the feature value of each observation data 44. The length 42 of the divided period ΔT includes a default value. The length 43 of the overlap period Δd1 includes a default value.

[0042] The storage unit 40 holds the data input from the input unit 30 into the signal processing circuit 50. For example, the storage unit 40 holds the set value of the length of the division period ΔT input from the input unit 30, and the set value of the length of the overlap period Δd1 input from the input unit 30. Note that, in the storage unit 40, the set value of the division period ΔT input from the input unit 30 is included in the length 42 of the division period ΔT. In addition, in the storage unit 40, the set value of the overlap period Δd1 input from the input unit 30 is included in the length 43 of the overlap period Δd1.

[0043] The storage unit 40 also holds data input from the detector 10 through the ADC 20 into the signal processing circuit 50. For example, the storage unit 40 holds digital observation data 44 input from the detector 10 through the ADC 20. For example, the storage unit 40 holds n pieces of observation data 44, such as Figure 1 shown.

[0044] The signal processing circuit 50 includes a processor, and executes a program (e.g., a processing program 41) held in the storage unit 40. For example, the processing program 41 is loaded into the signal processing circuit 50 to generate a feature value of an observation object waveform for each piece of observation data 44 based on the observation data 44, and also to generate an evaluation value related to a difference in the observation object waveform between the pieces of observation data 44 based on the feature value of each of the pieces of observation data 44.

[0045] The display controller 60 is a controller that controls display of the display 70. The display controller 60 generates an image signal for displaying an image including the evaluation value input from the signal processing circuit 50, and outputs it to the display 70. The display 70 displays an image corresponding to the image signal input from the display controller 60. The display 70 includes, for example, a liquid crystal panel and an organic EL (Electro Luminescence) panel.

[0046] Next, the generation of the above-mentioned characteristic value and the above-mentioned evaluation value will be described. Note that in the following description, the area area_a (the "a" at the end represents an integer in the range of 1 to n (including n)) to be described later will be given as an example of the above-mentioned characteristic value. In addition, in the following description, the effect amount d1 to be described later will be given as an example of the above-mentioned evaluation value.

[0047] Figure 2 is a diagram schematically illustrating an example of a process of calculating the above-mentioned feature value and the above-mentioned evaluation value in the biological information evaluation apparatus 1 . Figure 3 : is a flowchart of an example of a process of calculating the above-mentioned feature value and the above-mentioned evaluation value in the biological information evaluation device 1. Note that, Figure 2 The calculation process in is called Example 1.

[0048] First, the signal processing circuit 50 issues an instruction to the detector 10 to obtain biological information within a predetermined period (observation period T). Then, the detector 10 obtains analog biological information (observation data 44) within a predetermined period (observation period T) from the living body 100, and outputs the obtained analog observation data 44 to the ADC 20. The ADC 20 converts the analog observation data 44 (analog signal) input from the detector 10 into a digital signal, and outputs the digital observation data 44 (digital signal) obtained by the conversion to the signal processing circuit 50. The signal processing circuit 50 stores the digital observation data 44 input from the ADC 20 in the storage unit 40. The signal processing circuit 50 repeats this process to store n pieces of observation data 44 in the storage unit 40. At this time, the n pieces of observation data 44 may be, for example, data obtained from the same person, or may be individual data obtained from n persons.

[0049] Next, the signal processing circuit 50 obtains n pieces of observation data 44 (samples) from the storage unit 40 (step S101). The signal processing circuit 50 generates a characteristic value (area area_a) of the observation object waveform for each piece of obtained observation data 44 based on a plurality of partial observation data 45 having an observation period (division period ΔT) shorter than the observation period T of the observation data 44, and the plurality of partial observation data 45 are included in each piece of obtained observation data 44.

[0050] Specifically, the signal processing circuit 50 first derives a power spectrum PΔTa_b(t) (1≤a≤n, 1≤b≤m) for each divided time period ΔT in each of the observation data 44 (step S102). For example, the signal processing circuit 50 applies FFT (Fast Fourier Transform) to the partial observation data 45 for each divided time period ΔT in each of the observation data 44 to derive the power spectrum PΔTa_b(t). The power spectrum PΔTa_b(t) is a mathematical function of the time t within the divided time period ΔT, and may include the frequency spectrum of the observed object waveform.

[0051] In the case where a person's head is used as the living body 100, the observation object waveform includes, for example, α wave, β wave, γ wave, θ wave, δ wave, etc. The division period ΔT has a minimum necessary length that allows the analysis of the observation object waveform. In the case where a person's head is used as the living body 100 and the observation object waveform is an α wave, the division period ΔT falls within a range of, for example, about several seconds to several tens of seconds. For example, the signal processing circuit 50 sets the default value of the length 42 of the division period ΔT held in the storage unit 40 to the division period ΔT.

[0052] Furthermore, the signal processing circuit 50 sets each divided period ΔT so that the observation period (divided period ΔT) of any one of the pieces of partial observation data 45 and the observation period (divided period ΔT) of another of the pieces of partial observation data 45 partially overlap each other. In other words, the observation period (divided period ΔT) of one of the pieces of partial observation data 45 and the observation period (divided period ΔT) of another of the pieces of partial observation data 45 partially overlap. The signal processing circuit 50 sets, for example, the default value of the length 43 of the overlap period Δd1 held in the storage unit 40 to the partially overlapping portion (overlap period Δd1).

[0053] Next, the signal processing circuit 50 performs component analysis in each divided time period ΔT to derive an analysis result RΔTa_b (1≤a≤n, 1≤b≤m) (step S103). For example, the signal processing circuit 50 derives the area fc(t) of the frequency band of the observed object waveform contained in each power spectrum PΔTa_b(t), and sets the derived area fc(t) as the analysis result RΔTa_b. In the case where the observed object waveform is an α wave, the signal processing circuit 50 derives, for example, the area of ​​the frequency band of the α wave under each power spectrum PΔTa_b(t). Note that although the frequency band of the α wave generally falls within the range of 8 Hz to 13 Hz, the frequency band when deriving the area fc(t) is not limited to this frequency band. Next, the signal processing circuit 50 derives the power spectrum PΔTa(1≤a≤n) of the analysis result RΔTa_b for each of the observation data 44 (step S104). For example, the signal processing circuit 50 applies FFT to the area fc(t) of each of the pieces of observation data 44 to derive the power spectrum PΔTa.

[0054] Next, the signal processing circuit 50 derives the area area_a (1≤a≤n) of the power spectrum PΔTa for each of the observation data 44 (step S105). For example, the signal processing circuit 50 derives the area per unit frequency of the power spectrum PΔTa for each of the observation data 44, and also derives the average value of these areas to set the derived area per unit frequency and the derived average value as the area area_a. Next, the signal processing circuit 50 derives the effect amount d1 based on the area area_a in each of the observation data 44. For example, the signal processing circuit 50 assigns the area area_a in each of the observation data 44 to the expressions (1) to (4) described below to derive the effect amount d1. Here, the "average value" in expression (1) represents, for example, the average value contained in the area area_a. In addition, the "individual numerical value" in expressions (3) and (4) represents, for example, the area per unit frequency contained in the area area_a.

[0055] [Expression 1]

[0056]

[0057] x 1: The average value of a sample

[0058] x 2: The mean of another sample

[0059] s: pooled standard deviation of two samples

[0060] [Expression 2]

[0061]

[0062] n1: the total number of data in a sample

[0063] n2: the total number of data in another sample

[0064] s 2 1: Variance of a sample

[0065] s 2 2: Variance of another sample

[0066] [Expression 3]

[0067]

[0068] x 1,i : Individual values ​​in a sample

[0069] [Expression 4]

[0070]

[0071] x 2,i : Individual values ​​in another sample

[0072] Next, the signal processing circuit 50 outputs the derived effect amount d1 to the display controller 60. Thereafter, the display controller 60 generates an image signal for displaying an image containing the effect amount d1, and outputs it to the display 70. The display 70 displays an image corresponding to the image signal input from the display controller 60. Based on the effect amount d1 displayed on the display 70, the user determines whether the length of the divided period ΔT is appropriate, or whether the length of the overlapping period Δd1 is appropriate. For example, it is possible to determine whether the length of the divided period ΔT is appropriate based on whether the effect amount d1 is maximized. Similarly, for example, it is possible to determine whether the length of the overlapping period Δd1 is appropriate based on whether the effect amount d1 is maximized. It can be considered that the value suitable for the length of the divided period ΔT may be different depending on the individual living body 100. Similarly, it can be considered that the value suitable for the length of the overlapping period Δd1 may be different depending on the individual living body 100.

[0073] In the case where the user judges that the length of the divided period ΔT needs to be changed, the user inputs the setting value of the length of the divided period ΔT into the input unit 30. The input unit 30 outputs the setting value of the length of the divided period ΔT input by the user to the signal processing circuit 50. The signal processing circuit 50 holds the setting value of the length of the divided period ΔT input from the input unit 30 in the storage unit 40. In addition, in the case where the user judges that the length of the overlapping period Δd1 needs to be changed, the user inputs the setting value of the length of the overlapping period Δd1 into the input unit 30. The input unit 30 outputs the setting value of the length of the overlapping period Δd1 input from the user to the signal processing circuit 50. The signal processing circuit 50 holds the setting value of the length of the overlapping period Δd1 input from the input unit 30 in the storage unit 40.

[0074] The signal processing circuit 50 determines whether to change the length of the divided period ΔT. In the case where the setting value of the length of the divided period ΔT is newly maintained in the storage unit 40 or the setting value of the length of the divided period ΔT is updated in the storage unit 40, the signal processing circuit 50 reads the setting value of the length of the divided period ΔT from the storage unit 40, and changes the length of the divided period ΔT based on the read setting value (step S107: "Yes"). The signal processing circuit 50 performs steps S101 to S106 using the divided period ΔT having the changed length. In the case where the setting value of the length of the divided period ΔT is not maintained in the storage unit 40 or the setting value of the length of the divided period ΔT is not updated in the storage unit 40, the signal processing circuit 50 moves to step S108 (step S107: "No").

[0075] The signal processing circuit 50 determines whether to change the length of the overlapping period Δd1. In the case where the setting value of the length of the overlapping period Δd1 is newly saved in the storage unit 40 or the setting value of the length of the overlapping period Δd1 is updated in the storage unit 40, the signal processing circuit 50 reads the setting value of the length of the overlapping period Δd1 from the storage unit 40, and changes the length of the overlapping period Δd1 based on the read setting value (step S108: "Yes"). The signal processing circuit 50 performs steps S101 to S107 using the overlapping period Δd1 having the changed length. In the case where the setting value of the length of the overlapping period Δd1 is not maintained in the storage unit 40 or the setting value of the length of the overlapping period Δd1 is not updated in the storage unit 40, the signal processing circuit 50 ends the processing.

[0076] Figure 4 : is a diagram schematically illustrating an example of a process of calculating the above-mentioned feature value and the above-mentioned evaluation value in a biological information evaluation apparatus according to a comparative example. Figure 5 is a flowchart of an example of a process of calculating the above-mentioned feature value and the above-mentioned evaluation value in a biological information evaluation apparatus according to a comparative example.

[0077] According to the biological information evaluation device (signal processing circuit) of the comparative example, n pieces of observation data 44 (samples) are obtained from the storage unit (step S201). Next, according to the biological information evaluation device (signal processing circuit) of the comparative example, a power spectrum PTa (1≤a≤n) is derived for each of the pieces of observation data 44 (step S202). Then, according to the biological information evaluation device (signal processing circuit) of the comparative example, component analysis is performed to derive an analysis result RTa (1≤a≤n) for each of the pieces of observation data 44 (step S203). For example, according to the biological information evaluation device (signal processing circuit) of the comparative example, the area fca(t) of the frequency band of the observation object waveform contained in each power spectrum PTa is derived, and the average value fca_avg of the area fca(t) is also derived to set the derived area fca(t) and the derived average value fca_avg as the analysis result RTa. Next, the signal processing circuit 50 derives the effect amount d2 based on the analysis result RTa in each of the pieces of observation data 44. For example, the signal processing circuit 50 assigns the analysis result RTa in each of the observation data 44 to the above-mentioned expressions (1) to (4) to derive the effect amount d2. Here, the "average value" in expression (1) represents, for example, the average value fca_avg contained in the analysis result RTa. In addition, the "individual numerical value" in expressions (3) and (4) represents, for example, the area fca(t) contained in the area area_a. In this way, the processing in the biological information evaluation device (signal processing circuit) according to the comparative example is completed.

[0078] [Effect]

[0079] Next, the effects of the biological information evaluation device 1 will be described.

[0080] In the field of psychology, it is considered necessary to collect a relatively large number of samples and also increase the measurement period in order to show the significance of the difference between states. However, it usually takes time and effort to establish a method to control a person in a desired state or to increase the number of samples in order to obtain high-quality biological information about a person in order to understand the desired state. Therefore, it is extremely difficult to establish a technology that meets the application requirements of understanding the difference between states.

[0081] In contrast, the biological information evaluation device 1 according to the present embodiment generates a characteristic value (e.g., area area_a) of the observation object waveform for each of the pieces of observation data 44 based on the plurality of partial observation data 45 included in each of the pieces of observation data 44. This makes it possible to reveal the existence of differences between states. Therefore, the differences between states (e.g., effect size d1) can be understood in a case where the number of samples is small and the measurement period is short.

[0082] In fact, the evaluation value (e.g., effect size d1) generated based on the characteristic value (e.g., area area_a) of each of the pieces of observation data 44 and related to the difference in the observation object waveform between the pieces of observation data 44 is actually larger than the evaluation value (e.g., effect size d2) according to the comparative example. Therefore, with the biological information evaluation device 1 according to the present embodiment, the number of samples can be reduced. In addition, even if the measurement period is shortened, the existence of the difference between the states can be revealed.

[0083] Furthermore, in the case of the present embodiment, the observation period ΔT of the partial observation data 45 is set to have the minimum necessary length that allows the analysis of the waveform of the observation object. By this setting, sufficient pieces of partial observation data 45 required for analysis are ensured. Therefore, the length of the observation period T of the observation data 44 can be reduced to a length that allows the application requirements to be met. Therefore, with the biological information evaluation device 1 according to the present embodiment, even if the measurement period is shortened, the existence of the difference between the states can be revealed.

[0084] In addition, in the present embodiment, the length of the observation period ΔT is changed based on the setting value of the observation period ΔT received by the input unit 30. Using the observation period ΔT having the changed length, a plurality of partial observation data 45 are obtained from each of the observation data 44. Based on the obtained plurality of partial observation data 45, a characteristic value (e.g., area area_a) of the observation object waveform is generated for each of the observation data 44. This makes it possible to set the length of the observation period ΔT to a length that maximizes the difference between the states (e.g., the effect amount d1). Therefore, with the biological information evaluation device 1 according to the present embodiment, the number of samples can be reduced, and even if the measurement period is shortened, the existence of the difference between the states can be exhibited.

[0085] In addition, in the present embodiment, an image signal for displaying an image containing an evaluation value (e.g., effect size d1) is generated, and the image is displayed based on the generated image signal. This enables the user to judge whether the length of the divided period ΔT is appropriate based on the displayed evaluation value (e.g., effect size d1). For example, it is possible to judge whether the length of the divided period ΔT is appropriate based on whether the evaluation value (e.g., effect size d1) is maximized. Therefore, using the biological information evaluation device 1 according to the present embodiment, the number of samples can be reduced, and even if the measurement period is shortened, the existence of the difference between the states can be shown.

[0086] In addition, in the present embodiment, the observation period ΔT of any part of the observation data 45 partially overlaps with the observation period of another part of the observation data 45. This can further increase the evaluation value (e.g., effect size d1). In this case, the number of samples can be reduced, and even if the measurement period is shortened, the existence of the difference between the states can be revealed.

[0087] Furthermore, in the present embodiment, the length of the overlapping period Δd1 is changed based on the setting value of the overlapping period Δd1 received by the input unit 30. Using the overlapping period Δd1 having the changed length, a plurality of partial observation data 45 are obtained from each of the observation data 44. Based on the obtained plurality of partial observation data 45, a characteristic value (e.g., area area_a) of the observation object waveform is generated for each of the observation data 44. This makes it possible to set the length of the overlapping period Δd1 to a length that allows the difference between the states (e.g., the effect amount d1) to be maximized. Therefore, with the biological information evaluation device 1 according to the present embodiment, the number of samples can be reduced, and even if the measurement period is shortened, the presence of the difference between the states can be exhibited.

[0088] In addition, in the present embodiment, an image signal for displaying an image containing an evaluation value (e.g., effect size d1) is generated based on the generated image signal to display the image. This enables the user to judge whether the length of the overlapping period Δd1 is appropriate based on the displayed evaluation value (e.g., effect size d1). For example, it is possible to judge whether the length of the overlapping period Δd1 is appropriate based on whether the evaluation value (e.g., effect size d1) is maximized. Therefore, using the biological information evaluation device 1 according to the present embodiment, the number of samples can be reduced, and even if the measurement period is shortened, the existence of the difference between the states can be shown.

[0089] It should be noted that in this embodiment, for example, Figure 6 As shown, any divided period ΔT and another divided period ΔT may not overlap with each other. Even with this configuration, effects similar to those of the above-described embodiment can be obtained.

[0090] <2. Modification of the First Embodiment>

[0091] [Revision A]

[0092] Figure 74 is a diagram illustrating a modification of the schematic configuration of the biological information evaluation device 1 according to the first embodiment described above. In this modification, the storage unit 40 holds a length 46 of the division period ΔW and a length 47 of the overlap period Δd2 in addition to the program (e.g., the processing program 41) executed by the signal processing circuit 50, the length 42 of the division period ΔT, and the length 43 of the overlap period Δd1. The length 46 of the division period ΔW includes a default value. The length 47 of the overlap period Δd2 includes a default value. The value of the division period ΔW is greater than the value of the division period ΔT, and preferably, its value is greater than the value of ΔW1 to be described later.

[0093] Figure 8 : is a diagram schematically illustrating an example of a process of calculating the above-mentioned feature value and the above-mentioned evaluation value in the biological information evaluation apparatus 1 according to the present modification example. Fig. 9 : is a flowchart of an example of a process of calculating the above-mentioned feature value and the above-mentioned evaluation value in the biological information evaluation apparatus 1 according to the present modification example.

[0094] The signal processing circuit 50 first obtains n pieces of observation data 44 (samples) from the storage unit 40 (step S301). Then, the signal processing circuit 50 performs the above-mentioned steps S102 and S103 to derive the analysis result RΔTa_b (1≤a≤n, 1≤b≤k) in each divided period ΔT (step S302). For example, the signal processing circuit 50 sets the above-mentioned area fc(t) as the analysis result RΔTa_b. In addition, the signal processing circuit 50 generates the analysis result RΔTa(t) as a mathematical function of the time t within the observation period T based on the derived analysis result RΔTa_b.

[0095] Next, for the analysis result RΔTa(t), the signal processing circuit 50 derives the power spectrum PΔWa_b(t) of each divided period ΔW (step S302). For example, for each analysis result RΔTa(t), the signal processing circuit 50 applies FFT to part of the analysis result in each divided period ΔW to derive the power spectrum PΔWa_b(t). The power spectrum PΔWa_b(t) is a mathematical function of the time t within the divided period ΔW, and may include the frequency spectrum of the observed object waveform. For example, the signal processing circuit 50 sets the default value of the length 46 of the divided period ΔW maintained in the storage unit 40 to the divided period ΔW.

[0096] Next, the signal processing circuit 50 performs component analysis on each divided time period ΔW to derive the analysis result RΔWa_b (1≤a≤n, 1≤b≤k) (step S303). For example, the signal processing circuit 50 derives the area fc1(t) of the frequency band of the observed object waveform contained in each power spectrum PΔWa_b(t) to set the derived area fc1(t) as the analysis result RΔWa_b. Next, the signal processing circuit 50 derives the power spectrum PΔWa(1≤a≤n) of the analysis result RΔWa_b for each of the observation data 44 (step S304). For example, the signal processing circuit 50 applies FFT to the area fc1(t) in each of the observation data 44 to derive the power spectrum PΔWa.

[0097] Next, the signal processing circuit 50 derives the area area_a (1≤a≤n) of the power spectrum PΔWa in each of the observation data 44 (step S305). For example, the signal processing circuit 50 derives the area per unit frequency of the power spectrum PΔWa and the average value of these areas in each of the observation data 44 to set the derived area per unit frequency and the average value of the area as the area area_a. Next, based on the area area_a in each of the observation data 44, the signal processing circuit 50 derives the effect amount d1. For example, the signal processing circuit 50 assigns the area area_a in each of the observation data 44 to the above-mentioned expressions (1) to (4) to derive the effect amount d1.

[0098] Next, the signal processing circuit 50 outputs the derived effect amount d1 to the display controller 60. Then, the display controller 60 generates an image signal for displaying an image containing the effect amount d1, and outputs it to the display 70. The display 70 displays an image corresponding to the image signal input from the display controller 60. The user determines whether the lengths of the division periods ΔT and ΔW are appropriate, or whether the lengths of the overlapping periods Δd1 and Δd2 are appropriate, based on the effect amount d1 displayed on the display 70. For example, it is possible to determine whether the lengths of the division periods ΔT and ΔW are appropriate based on whether the effect amount d1 is maximized. Similarly, for example, it is possible to determine whether the lengths of the overlapping periods Δd1 and Δd2 are appropriate based on whether the effect amount d1 is maximized. It can be considered that the value of each length suitable for the division periods ΔT and ΔW may be different depending on the individual living bodies 100. Similarly, it can be considered that the value of each length suitable for the overlapping periods Δd1 and Δd2 may be different depending on the individual living bodies 100.

[0099] In the case where the user judges that the length of the divided period ΔT or ΔW needs to be changed, the user inputs the setting value of the length of the divided period ΔT or ΔW into the input unit 30. The input unit 30 outputs the setting value of the length of the divided period ΔT or ΔW input from the user to the signal processing circuit 50. The signal processing circuit 50 holds the setting value of the length of the divided period ΔT or ΔW input from the input unit 30 in the storage unit 40. In addition, in the case where the user judges that the length of the overlapping period Δd1 or Δd2 needs to be changed, the user inputs the setting value of the length of the overlapping period Δd1 or Δd2 into the input unit 30. The input unit 30 outputs the setting value of the length of the overlapping period Δd1 or Δd2 input from the user to the signal processing circuit 50. The signal processing circuit 50 holds the setting value of the length of the overlapping period Δd1 or Δd2 input from the input unit 30 in the storage unit 40.

[0100] The signal processing circuit 50 determines whether to change the length of the divided period ΔT or ΔW. In the case where the setting value of the length of the divided period ΔT or ΔW is newly retained in the storage unit 40, or in the case where the setting value of the length of the divided period ΔT or ΔW of the storage unit 40 is updated, the signal processing circuit 50 reads the setting value of the length of the divided period ΔT or ΔW from the storage unit 40, and changes the length of the divided period ΔT or ΔW based on the read setting value (step S307: "Yes"). The signal processing circuit 50 performs steps S301 to S306 using the divided period ΔT or ΔW with the changed length. In the case where the setting value of the length of the divided period ΔT or ΔW is not retained in the storage unit 40, or in the case where the setting value of the length of the divided period ΔT or ΔW is not updated in the storage unit 40, the signal processing circuit 50 moves to step S308 (step S307: "No").

[0101] The signal processing circuit 50 determines whether to change the length of the overlapping period Δd1 or Δd2. In the case where the setting value of the length of the overlapping period Δd1 or Δd2 is newly maintained in the storage unit 40, or in the case where the setting value of the length of the overlapping period Δd1 or Δd2 is updated in the storage unit 40, the signal processing circuit 50 reads the setting value of the length of the overlapping period Δd1 or Δd2 from the storage unit 40, and changes the length of the overlapping period Δd1 or Δd2 based on the read setting value (step S308: "Yes"). The signal processing circuit 50 performs steps S301 to S307 using the overlapping period Δd1 or Δd2 with the changed length. In the case where the setting value of the length of the overlapping period Δd1 or Δd2 is not maintained in the storage unit 40, or in the case where the setting value of the length of the overlapping period Δd1 or Δd2 is not updated in the storage unit 40, the signal processing circuit 50 ends the processing.

[0102] In the present modification, a power spectrum PΔTa_b is derived for each partial observation data 45. Component analysis is performed on each derived power spectrum PΔTa_b, and for each piece of data having a plurality of division periods ΔW longer than the division period ΔT included in the data (analysis result RΔTa) obtained based on the component analysis, a power spectrum RΔWa_b is derived. Component analysis is performed on each derived power spectrum RΔWa_b. Based on the component analysis, an effect amount d1 is generated for each piece of observation data 44. This makes it possible to obtain an effect similar to that of the above-described embodiment.

[0103] It should be noted that in the present modification, any divided period ΔT and another divided period ΔT may not overlap with each other, for example, as Fig.10 As shown. In addition, any divided period ΔW and another divided period ΔW may not overlap with each other. In these cases, similar effects to those of the above-mentioned modification examples can also be obtained.

[0104] [Revision B]

[0105] In the above-mentioned modification A, the biological information evaluation device 1 may cause the storage unit 40 to hold the learning model 48, for example, as Fig.11 As shown. The learning model 48 is used, for example, to perform Fig.12 The learning process or execution shown Fig.13 The estimation process is shown.

[0106] Here, SΔTa_b(t) represents the area of ​​the frequency band of the observation object waveform contained in the power spectrum PΔTa_b(t) (1≤a≤h, 1≤b≤j), which is derived from each of the j divided time periods ΔT in each of the h observation data 44 used for learning (see Fig.12 ). In addition, SΔWa_b(t) represents the area of ​​the frequency band of the observation object waveform contained in the power spectrum PΔWa_b(t) (1≤a≤h, 1≤b≤j) derived from each of the j divided time periods ΔW obtained by dividing each of the h observation data 44 for learning (see Fig.12 ). E(t) represents the emotional state in the observation period T in each of the h observation data 44 used for learning. Note that the number of observation data 44 used for learning (h pieces) is equal to the number of observation data 44 when estimating emotions, which will be described later (n pieces). In addition, the number of divisions of the observation data 44 when learning (k pieces) may be equal to or different from the number of divisions of the observation data 44 when estimating emotions, which will be described later (j pieces). In addition, in both the area SΔTa_b(t) and the area SΔWa_b(t), the data exists simultaneously in the observation period T.

[0107] In the area SΔTa_b(t) and the area SΔWa_b(t), the learning model 48 represents a model obtained by learning (such as machine learning) that uses 2n pieces of data each having the same "b" as explanatory variables and also uses the emotional state E(t) in the period corresponding to "b" among the emotional states E(t) as an object variable (see Fig.14 ). In other words, the learning model 48 is a model for estimating one emotional state from 2n pieces of data.

[0108] The signal processing circuit 50 derives a power spectrum PΔTa_b(t) (1≤a≤n, 1≤b≤k) for each of the k divided time periods ΔT in each of the n pieces of observation data 44 that have been actually measured, to derive an area RΔTa_b(t) of the frequency band of the observation object waveform included in the derived power spectrum PΔTa_b(t) (see Fig.13 ). In addition, the signal processing circuit 50 derives a power spectrum PΔWa_b(t) (1≤a≤n, 1≤b≤k) for each of the k divided time periods ΔW in each of the n pieces of observation data 44 that have been actually measured, so as to derive an area RΔWa_b(t) of the frequency band of the observation object waveform included in the derived power spectrum PΔWa_b(t) (see Fig.13 ).

[0109] In the learning model 48, in the case where 2n pieces of data having the same "b" are input into the area RΔTa_b and the area RΔWa_b (each of which is derived in the signal processing circuit 50), the emotional state Out_b corresponding to each of the 2n pieces of data is output to the signal processing circuit 50 (see Fig.13 ).

[0110] Here, it is assumed that i (i≤k) emotional states in the emotional state Out_b match the actual emotional state when the actual emotional states corresponding to each of the area RΔTa_b and the area RΔWa_b are equal. At this time, the estimated accuracy is i / k.

[0111] In the case of the present modification, the learning model 48 is used to estimate the emotional state Out_b. Using the effect size d1 obtained in the above embodiment, it is possible to estimate, for example, two states: a high alert level and a low alert level. On the other hand, using the learning model, in addition to high or low alert, the present modification is also able to estimate, for example, an intermediate level state, such as which degree of alert level is high or which degree of alert level is low. Therefore, in the present modification, a higher estimation accuracy than the above embodiment can be obtained.

[0112] In addition, the present modification uses the learning model 48 to estimate the emotional state Out_b. Therefore, compared with the number of data used in the above-mentioned embodiment, when estimating emotions, it is possible to use a reduced number of (n) observation data 44 for estimation, and it is possible to use only one piece of data for estimation. This makes it possible to estimate emotions even in a case where it is impossible to obtain a large number of (n) observation data 44 when estimating emotions.

[0113] [Revision C]

[0114] In the above-mentioned modification B, the biological information evaluation device 1 can use a model obtained by learning (such as machine learning) that uses the area SΔTa_b(t) for each "b" as an explanatory variable and also uses the emotional state E(t) in the period corresponding to "b" in the emotional state E(t) as an object variable, for example, as Fig.14 In other words, in the present modification, the learning model 48 is a model for estimating one emotional state based on n pieces of data.

[0115] The signal processing circuit 50 derives a power spectrum PΔTa_b(t) (1≤a≤n, 1≤b≤k) for each of the k divided time periods ΔT in each of the n pieces of observation data 44 that have been actually measured, so as to derive an area RΔTa_b(t) of the frequency band of the observation object waveform included in the derived power spectrum PΔTa_b(t) (see Fig.15 ).

[0116] In the learning model 48, the area RΔTa_b derived by the signal processing circuit 50 is input for each "b", and the emotional state Out_b corresponding to each of the n input data is output to the signal processing circuit 50 (see Fig.15 ).

[0117] Here, it is assumed that when the actual emotional states corresponding to each area RΔTa_b are all equal, i (i≤k) emotional states in the emotional state Out_b match the actual state. At this time, the accuracy of the estimation is i / k.

[0118] In this modification, the learning model 48 is used to estimate the emotional state Out_b. Using the effect size d1 obtained in the above embodiment, it is possible to estimate, for example, two states: a high alert level and a low alert level. On the other hand, using the learning model, in addition to whether the alert level is high or low, this modification is also able to estimate, for example, intermediate level states, such as which degree of alert level is high or which degree of alert level is low. Therefore, in this modification, a higher estimation accuracy than the above embodiment can be obtained.

[0119] Next, refer to Fig.16 and Fig.17 , the effects of the biological information evaluating apparatus 1 according to the above-mentioned modifications B and C will be described.

[0120] Fig.16 : is a diagram illustrating an example of the relationship between the length of the divided period ΔW and the accuracy of estimation in a model in which the divided period ΔT, the overlapping period Δd1 , and the overlapping period Δd2 are each fixed to a certain value. Fig.17 The length of the divided time period ΔW is fixed to be greater than Fig.16 Graph showing an example of estimation accuracy for a large value (ΔW2) of ΔW1 in the model shown. Fig.16 and Fig.17 In, examples of estimation accuracy according to Modifications B and C are given. Fig.16 and Fig.17 It is clear that in Modification B, when the length of the divided period ΔW is equal to or greater than ΔW1, the state can be estimated with higher accuracy than Modification C. Note that, Fig.16 and Fig.17 The values ​​in are only examples. Depending on the application, higher values ​​can be obtained. Fig.16 and Fig.17 The numerical value of the value in .

[0121] <3. Second Embodiment>

[0122] [Configuration]

[0123] Next, a biological information evaluation device 2 according to a second embodiment of the present disclosure will be described. Fig.18 , Fig.19 and Fig. 20 1 and 2 are diagrams each illustrating an example of a schematic configuration of a biological information evaluation apparatus 2 including the biological information evaluation apparatus 1 according to the above-described first embodiment and modifications A, B, and C of the first embodiment.

[0124] The biological information evaluation device 2 includes, for example, a biological information observation device 210 and a biological information analysis device 220 .

[0125] The biological information observation device 210 includes, for example, a detector 10, an ADC 20, a signal processing circuit 211, and a communication section 212. The signal processing circuit 211 includes a processor. For example, the signal processing circuit 211 controls the detector 10 to acquire observation data 44, and also controls the transmission of the acquired observation data 44 through the communication section 212. The communication section 212 transmits the observation data 44 to the biological information analysis device 220 according to the control of the signal processing circuit 211.

[0126] The biological information analysis device 220 is, for example, a mobile terminal, a smart phone, or a tablet computer having a communication function. The biological information analysis device 220 includes, for example, a communication section 221, an input unit 30, a storage unit 40, a signal processing circuit 50, a display controller 60, and a display 70. The communication section 221 receives the observation data 44 transmitted from the biological information observation device 210, and outputs the received observation data 44 to the signal processing circuit 50. For example, near-field wireless communication is used for communication between the communication section 212 and the communication section 221. Near-field wireless communication is implemented using, for example, ISO / IEC14443 (international standard for near-field RFID), ISO / IEC18092 (international standard for wireless communication called NFC), ISO / IEC15693 (international standard for RFID), Bluetooth (registered trademark), etc. Note that the communication between the communication section 212 and the communication section 221 can be implemented, for example, through a wired LAN (local area network), a wireless LAN (such as Wi-Fi), or a telephone line.

[0127] In this embodiment, as in the first embodiment and the modifications A, B, and C of the first embodiment, the existence of the difference between the states can be exhibited with a small number of samples and a short measurement period. In addition, with this embodiment, as in the first embodiment and the modifications A, B, and C of the first embodiment described above, the number of samples can be reduced, and the existence of the difference between the states can be exhibited even if the measurement period is shortened.

[0128] It should be noted that the effects described in this specification are merely examples. The effects of the present disclosure are not limited to the effects described in this specification. The present disclosure may have effects other than those described in this specification.

[0129] For example, the above-described series of processing can be executed by software, and alternatively, can be executed by hardware.

[0130] In addition, the first and second embodiments and modifications A, B and C of the first embodiment can be applied to applications in which brain waves are measured to understand the state in which the alert level (stress) is high and the state in which the alert level (stress) is low. However, they can be applied to other applications. For example, the first and second embodiments and modifications A, B and C according to the first embodiment can be applied to applications in which the difference between states in games, e-sports, movies, music, health care, medicine, education, driving of mobile bodies such as cars, or other applications is understood. For example, in the case of a game, the scene or difficulty can be changed by understanding the difference in the user's immersion, the difference in the user's excitement, etc.

[0131] In addition, in the case of the first and second embodiments and the modifications A, B and C of the first embodiment, the object of measurement is a living body. However, the object of measurement is not limited to a living body. The present disclosure can be applied to a field where the waveform of the observed object is known, and the observation data 44 containing the known waveform of the observed object can be obtained using the detector 10. In addition, in the case of the first and second embodiments and the modifications A, B and C of the first embodiment, the observed object is a human being. However, the observed object may be a livestock or an animal in a zoo, an aquarium, etc. In the case where the observed object is a livestock or an animal in a zoo, an aquarium, etc., the first and second embodiments and the modifications of the first embodiment may be used, for example, to detect the stress of the animal, etc.

[0132] Furthermore, the present disclosure may take, for example, the following configurations.

[0133] (1) A biological information evaluation device comprising:

[0134] A signal processing circuit generates a characteristic value of an observation object waveform for each of the observation data, based on a plurality of partial observation data contained in each of the observation data obtained through a living body observation for a predetermined period of time and each having an observation period shorter than the observation period of the observation data.

[0135] (2) The biological information evaluation apparatus according to (1), wherein the signal processing circuit generates an evaluation value regarding a difference in an observation object waveform between the pieces of observation data based on a feature value of each of the pieces of observation data.

[0136] (3) The biological information evaluation apparatus according to (2), wherein the evaluation value is an effect size.

[0137] (4) The biological information evaluation apparatus according to any one of (1) to (3), wherein an observation period of each of the pieces of partial observation data has a minimum necessary length that allows analysis of a waveform of the observation object.

[0138] (5) The biological information evaluation device according to (2) or (3), further comprising:

[0139] an input unit receiving an input of a setting value regarding the length of an observation period of each of the pieces of partial observation data, wherein

[0140] The signal processing circuit:

[0141] changing the length of the observation period of each of the pieces of partial observation data based on the setting value received at the input unit,

[0142] obtaining the plurality of partial observation data from each of the pieces of observation data using the observation period having the changed length, and

[0143] A characteristic value of the waveform of the observation object is generated for each of the observation data based on the obtained plurality of partial observation data.

[0144] (6) The biological information evaluation device according to (5), further comprising:

[0145] a display controller that generates an image signal for displaying an image including the evaluation value; and

[0146] The display displays an image based on the image signal from the display controller.

[0147] (7) The biological information evaluating device according to any one of (1) to (6), wherein an observation period of one of the pieces of partial observation data and an observation period of another of the pieces of partial observation data partially overlap with each other.

[0148] (8) The biological information evaluation device according to (7), further comprising:

[0149] an input unit that receives an input of a setting value regarding the length of an overlapping period in which an observation period of one of the pieces of partial observation data overlaps an observation period of another of the pieces of partial observation data, wherein

[0150] The signal processing circuit:

[0151] changing the length of the overlap period based on the setting value received at the input unit;

[0152] obtaining the plurality of partial observation data from each of the pieces of observation data using overlapping periods having varying lengths, and

[0153] A characteristic value of the observation object waveform is generated for each of the observation data based on the obtained plurality of partial observation data.

[0154] (9) The biological information evaluation device according to (8), further comprising:

[0155] a display controller that generates an image signal for displaying an image including the evaluation value; and

[0156] The display displays an image based on the image signal from the display controller.

[0157] (10) The biological information evaluation device according to any one of (1) to (9), further comprising:

[0158] The detector obtains each of the observation data through in vivo observation.

[0159] (11) The biological information evaluation apparatus according to any one of (1) to (10), wherein

[0160] The signal processing circuit:

[0161] deriving a first power spectrum for each of the partial observation data,

[0162] performing a component analysis on the derived first power spectrum; and

[0163] Generate feature values ​​for each of the observations based on component analysis.

[0164] (12) The biological information evaluation apparatus according to any one of (1) to (10), wherein

[0165] The signal processing circuit:

[0166] deriving a first power spectrum for each of the partial observation data,

[0167] Perform component analysis for the derived first power spectrum,

[0168] deriving a second power spectrum for each of the plurality of pieces of data contained in the data obtained based on the component analysis and having a second period longer than the first period, performing component analysis on the derived second power spectrum, and

[0169] Generate feature values ​​for each of the observations based on component analysis.

[0170] (13) The biological information evaluation device according to claim 1, wherein

[0171] The signal processing circuit:

[0172] deriving a first power spectrum for each of the partial observation data,

[0173] performing component analysis on the derived first power spectrum to obtain first data based on the component analysis;

[0174] deriving a second power spectrum for each of a plurality of pieces of data included in the observation data and having a second time period longer than the first time period,

[0175] performing component analysis for the derived second power spectrum to obtain second data based on the component analysis, and

[0176] An emotional state is obtained based on the first data and the second data.

[0177] (14) The biological information evaluation device according to claim 1, wherein

[0178] The signal processing circuit:

[0179] deriving a power spectrum for each of the partial observations, performing a component analysis on the derived power spectrum, and

[0180] Emotional states were obtained from the data obtained based on component analysis.

[0181] (15) A method for evaluating biological information by performing signal processing using a signal processing circuit, comprising:

[0182] Based on a plurality of pieces of partial observation data included in each of the pieces of observation data obtained by living body observation for a predetermined period and each having an observation period shorter than that of the observation data, a feature value of an observation object waveform is generated for each of the pieces of observation data.

[0183] This application claims priority based on Japanese Patent Application No. 2018-078807 filed with the Japan Patent Office on April 17, 2018, the entire contents of which are incorporated herein by reference.

[0184] It should be understood that those skilled in the art will make various modifications, combinations, sub-combinations and alterations depending on design requirements and other factors, and they are within the scope of the appended claims or the equivalents thereof.

Claims

1. A biological information evaluation device, include: a signal processing circuit that generates a characteristic value of an observation object waveform for each of the pieces of observation data based on a plurality of pieces of partial observation data that are included in each of the pieces of observation data obtained by the living body observation for a predetermined period and each having a first divided time period that is shorter than the observation time period of the observation data; as well as Input unit, wherein the signal processing circuit generates an evaluation value regarding a difference in an observation object waveform between the pieces of observation data based on a feature value of each of the pieces of observation data, wherein the first divided period has a minimum necessary length to allow analysis of the waveform of the observation object, wherein one of the first divided time periods and another of the first divided time periods partially overlap each other, In which, the input unit is configured to receive a setting value for the length of the first divided time period or a setting value for the length of the overlapping time period input by the user when the user of the biological information evaluation device determines, based on the evaluation value, that the length of the first divided time period is inappropriate or the length of the overlapping time period in which one of the first divided time periods overlaps with another of the first divided time periods is inappropriate.

2. The biological information evaluation device according to claim 1, wherein the evaluation value is an effect size.

3. The biological information evaluation device according to claim 1, The signal processing circuit changing the length of the first divided period based on a setting value of the length of the first divided period received at the input unit, obtaining the plurality of partial observation data from each of the observation data with the changed length, and A characteristic value of the observation object waveform is generated for each of the pieces of observation data based on the obtained plurality of partial observation data.

4. The biological information evaluation device according to claim 3, further comprising: include: a display controller that generates an image signal for displaying an image including the evaluation value; as well as The display displays an image based on the image signal from the display controller.

5. The biological information evaluation device according to claim 1, The signal processing circuit changing the length of the overlap period based on a setting value of the length of the overlap period received at an input unit; obtaining the plurality of partial observation data from each of the pieces of observation data using overlapping periods having varying lengths, and A characteristic value of the observation object waveform is generated for each of the pieces of observation data based on the obtained plurality of partial observation data.

6. The biological information evaluation device according to claim 5, further comprising: include: a display controller that generates an image signal for displaying an image including the evaluation value; as well as The display displays an image based on the image signal from the display controller.

7. The biological information evaluation device according to claim 1, further comprising The detector obtains each of the observation data through in vivo observation.

8. The biological information evaluation device according to claim 1, wherein The signal processing circuit deriving a first power spectrum for each of the partial observation data, performing a component analysis on the derived first power spectrum; and Generate feature values ​​for each of the observations based on component analysis.

9. The biological information evaluation device according to claim 1, wherein The signal processing circuit deriving a first power spectrum for each of the partial observation data, Perform component analysis for the derived first power spectrum, deriving a second power spectrum for each of a plurality of pieces of data contained in the data obtained based on the component analysis and having a second divided period longer than the first divided period, performing a component analysis on the derived second power spectrum, and Generate feature values ​​for each of the observations based on component analysis.

10. The biological information evaluation device according to claim 1, wherein The signal processing circuit deriving a first power spectrum for each of the partial observation data, performing component analysis on the derived first power spectrum to obtain first data based on the component analysis; deriving a second power spectrum for each of a plurality of pieces of data that are included in the observation data and have a second divided period that is longer than the first divided period, performing component analysis for the derived second power spectrum to obtain second data based on the component analysis, and An emotional state is obtained based on the first data and the second data.

11. The biological information evaluation device according to claim 1, wherein The signal processing circuit Derive a power spectrum for each of the partial observations, Perform component analysis on the derived power spectrum, and Emotional states were obtained from the data obtained based on component analysis.

12. A method of evaluating biological information by performing signal processing using a signal processing circuit, include: generating a feature value of an observation object waveform for each of the pieces of observation data based on a plurality of pieces of partial observation data contained in each of the pieces of observation data obtained by the living body observation for a predetermined period and each having a first divided time period shorter than the observation time period of the observation data, wherein the method further comprises generating an evaluation value regarding a difference in an observation object waveform between the pieces of observation data based on a feature value of each of the pieces of observation data, wherein the first divided period has a minimum necessary length to allow analysis of the waveform of the observation object, wherein one of the first divided time periods and another of the first divided time periods partially overlap each other, In which, when the user judges that the length of the first divided time period is inappropriate based on the evaluation value or the length of the overlapping period in which one of the first divided time periods overlaps with another of the first divided time periods is inappropriate, the setting value of the length of the first divided time period or the setting value of the length of the overlapping period input by the user is received.

Images