Bioinformatics computing system

Abstract

[Subject] Provide a biological information computing system that achieves improved evaluation accuracy. [Solution] The biological information computing system outputs information related to the characteristics of carbon dioxide in the user's blood, and includes an acquisition unit, a database, and a generation unit. The acquisition unit acquires evaluation data based on the pulse wave of the user. Classification information is stored in the database, and the classification information is generated by using a pair of input data and reference data as learning data and a plurality of the learning data, wherein the input data is data based on a previously acquired learning pulse wave, and the reference data is data containing the characteristics of carbon dioxide in the blood associated with the input data. The generation unit refers to the database and generates an evaluation result containing the characteristics of carbon dioxide in the blood corresponding to the evaluation data.

Description

Technical Field

[0001] The present invention relates to a biological information computing system that outputs information related to characteristics of carbon dioxide in a user's blood. Background Art

[0002] Conventionally, as a method for estimating information related to a biological body such as the partial pressure of carbon dioxide in the blood (PaCO 2 ) of a user, a method such as that disclosed in Patent Document 1 has been proposed.

[0003] Patent document 1 discloses the following: after constructing an average frame of each of at least two volume pulse waves having different light wavelengths, this method can be used to estimate the gain between the two average frames of the two volume pulse waves, and the gain value can also be used to estimate information for obtaining blood oxygen saturation or other components present in the blood, such as hemoglobin, carbon dioxide or other substances.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent Application No. 2020-513876 Summary of the invention

[0007] Problems to be solved by the invention

[0008] Here, when measuring the value of the partial pressure of carbon dioxide in the user's blood, by monitoring the measured value at a specific period, the user's health status can be grasped in detail. However, as a problem, when monitoring the measured value, the deviation of the measured value caused by the measurement conditions, etc. becomes larger, and the evaluation accuracy decreases. In this regard, in Patent Document 1, the characteristics of carbon dioxide in the blood are estimated based on the volume pulse wave using two wavelengths of light. Therefore, there is a possibility that a large deviation will occur in the measured value due to a slight difference in the angle of light emission and incident on the sensor.

[0009] Therefore, the present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide a biological information calculation system that can improve the evaluation accuracy.

[0010] Means for solving problems

[0011] The biological information computing system of the first invention outputs information related to the characteristics of carbon dioxide in the user's blood, and the biological information computing system is characterized in that it comprises: an acquisition unit, which acquires evaluation data based on the pulse wave of the user; a database, which stores classification information, which is generated by using a pair of input data and reference data as learning data and using a plurality of the learning data, wherein the input data is data based on a previously acquired learning pulse wave, and the reference data is data containing the characteristics of carbon dioxide in the blood associated with the input data; and a generation unit, which refers to the database to generate an evaluation result containing the characteristics of carbon dioxide in the blood corresponding to the evaluation data.

[0012] The biological information calculation system of the second invention is characterized in that, in the first invention, the reference data includes status information representing the health status of the subject whose learning pulse wave is measured, the classification information includes a learning model generated by machine learning using a plurality of the learning data, and the generation unit generates a health status result representing the health status of the user associated with the evaluation data by including it in the evaluation result with reference to the learning model.

[0013] The biological information calculation system of the third invention is characterized in that, in the second invention, the plurality of learning data only include the input data and the reference data when the subject is healthy, and the health status result indicates whether the health status of the user associated with the evaluation data is healthy or other than healthy.

[0014] The biological information calculation system of the fourth invention is characterized in that, in the first invention, the classification information is a calibration model obtained by using a PLS regression analysis using the input data as an explanatory variable and the reference data as a target variable.

[0015] The biological information calculation system of the fifth invention is characterized in that, in the first invention, the acquisition unit acquires preliminary evaluation data different from the evaluation data based on the pulse wave, and the generation unit calculates the biological information of the user corresponding to the preliminary evaluation data with reference to the database.

[0016] The biological information calculation system of the sixth invention is characterized in that, in the fifth invention, preliminary classification information is stored in the database, and the preliminary classification information is generated by using a pair of preliminary input data and preliminary reference data as preliminary learning data using a plurality of the preliminary learning data, wherein the preliminary input data is data based on the learning pulse wave, and the preliminary reference data is data including the biological information associated with the preliminary input data, and the generating unit calculates the biological information corresponding to the preliminary evaluation data with reference to the preliminary classification information.

[0017] The biological information calculation system of the 7th invention is characterized in that, in the 1st invention, the acquisition unit acquires preliminary evaluation data different from the evaluation data based on the pulse wave, the classification information includes a plurality of attribute classification information calculated using different learning data respectively, and the generation unit includes: a selection unit, which refers to the preliminary evaluation data and selects the first classification information from the plurality of attribute classification information; and an attribute generation unit, which refers to the first classification information and generates the evaluation result corresponding to the evaluation data.

[0018] The biological information calculation system of the eighth invention is characterized in that, in the seventh invention, the acquisition unit acquires data corresponding to a velocity pulse wave based on the pulse wave as the evaluation data, and acquires data corresponding to an acceleration pulse wave based on the pulse wave as the preliminary evaluation data.

[0019] Effects of the Invention

[0020] According to the first to eighth inventions, the generating unit refers to the database to generate an evaluation result including the characteristics of carbon dioxide in the blood corresponding to the evaluation data. In addition, the database stores classification information calculated using a plurality of learning data. Therefore, when generating the evaluation result, it is possible to include quantitative characteristics of carbon dioxide in the blood based on data with past performance (input data and characteristics of carbon dioxide in the blood). Thus, it is possible to improve the evaluation accuracy.

[0021] In particular, according to the second invention, the generation unit can refer to the learned model and generate the health status result associated with the evaluation data by including it in the evaluation result. Therefore, the health status based on the characteristics of carbon dioxide in the blood can be reported to the user, and the health status can be easily grasped compared to the case where only the characteristics of carbon dioxide in the blood are reported. As a result, the convenience for the user can be improved.

[0022] In particular, according to the third invention, the plurality of learning data only includes input data and reference data when the subject is healthy, and the health status result indicates whether the health status of the user associated with the evaluation data is healthy. Therefore, even when it is impossible to prepare input data based on the learning pulse wave when the subject is unhealthy and the characteristics of carbon dioxide in the blood, a result indicating the health status of the user can be generated. Thus, it is possible to further improve the convenience for the user without increasing the difficulty of generating classification information.

[0023] In particular, according to the fourth invention, the classification information is a calibration model obtained by using PLS regression analysis with input data as explanatory variables and reference data as target variables. Therefore, compared with the case where the classification information is calculated using machine learning, the amount of learning data can be greatly reduced, and the calibration model can be easily updated. As a result, it is possible to facilitate the construction and update of the biological information calculation system.

[0024] In particular, according to the fifth invention, the generating unit refers to the database to generate a preliminary evaluation result including the biological information of the user corresponding to the preliminary evaluation data. Therefore, it is possible to use the preliminary evaluation result generated from a different point of view from the evaluation result, and to implement various evaluations. Thus, it is possible to implement an appropriate evaluation corresponding to the user's requirements.

[0025] In particular, according to the sixth invention, the generating unit generates a preliminary evaluation result corresponding to the preliminary evaluation data with reference to the preliminary classification information. Therefore, it is possible to use the preliminary evaluation result generated with reference to information different from the evaluation result, and to implement a more comprehensive evaluation. Thus, it is possible to easily implement an appropriate evaluation corresponding to the user's request.

[0026] In particular, according to the seventh invention, the generating unit includes: a selecting unit that selects the first classification information with reference to the preliminary evaluation data; and an attribute-based generating unit that generates an evaluation result corresponding to the evaluation data with reference to the first classification information. Therefore, it is possible to generate an evaluation result corresponding to the evaluation data based on the selection of the attribute-based classification information that is most suitable for the characteristics of the pulse wave. Thus, it is possible to further improve the evaluation accuracy.

[0027] In particular, according to the eighth invention, the acquisition unit acquires data equivalent to a velocity pulse wave based on a pulse wave as evaluation data. In addition, the acquisition unit acquires data equivalent to an acceleration pulse wave based on a pulse wave as preliminary evaluation data. Therefore, it is possible to select attribute classification information using an acceleration pulse wave that is easier to classify the characteristics of a pulse wave than a velocity pulse wave. In addition, it is possible to generate an evaluation result using a velocity pulse wave that is easier to calculate the characteristics of carbon dioxide in the blood than an acceleration pulse wave. Thus, it is possible to achieve further improvement in evaluation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a schematic diagram showing an example of the biological information calculation system in the first embodiment.

[0029] Figure 2 (a) and Figure 2 (b) is a schematic diagram showing an example of the operation of the biological information calculation system in the first embodiment.

[0030] Figure 3 (a) is a schematic diagram showing an example of classification information. Figure 3 (b) and Figure 3 (c) is a schematic diagram showing an example of processing on sensor data.

[0031] Figure 4 (a) is a schematic diagram showing an example of the structure of a biological information computing device, Figure 4 (b) is a schematic diagram showing an example of the function of the biological information computing device.

[0032] Figure 5 (a) and Figure 5 (b) is a schematic diagram showing an example of a sensor.

[0033] Figure 6 This is a flowchart showing an example of the operation of the biological information calculation system in the first embodiment.

[0034] Figure 7 This is a schematic diagram showing an example of the operation of the biological information calculation system in the second embodiment.

[0035] Figure 8 FIG. 8( a ) and FIG. 8( b ) are schematic diagrams showing an example of a process of calculating biological information corresponding to sensor data.

[0036] Fig. 9 It is a schematic diagram showing a modified example of the operation of the biological information calculation system in the second embodiment.

[0037] Fig.10 This is a schematic diagram showing an example of the operation of the biological information calculation system in the third embodiment.

[0038] Fig.11 It is a schematic diagram showing a classification example of data corresponding to the acceleration pulse wave.

[0039] Fig.12 It is a schematic diagram showing a classification example of data corresponding to a velocity pulse wave. DETAILED DESCRIPTION

[0040] Hereinafter, an example of a biological information calculation system in an embodiment of the present invention will be described with reference to the drawings.

[0041] (First Embodiment: Biological Information Calculation System 100)

[0042] Figure 1 It is a schematic diagram showing an example of the biological information calculation system 100 in the first embodiment.

[0043] The biological information computing system 100 is used to output information related to the characteristics of carbon dioxide in the user's blood. The biological information computing system 100 can generate an evaluation result including the characteristics of carbon dioxide in the blood based on the evaluation data based on the user's pulse wave. In addition, the "characteristics of carbon dioxide in the blood" means the degree of carbon dioxide contained in the user's blood. As the characteristics of carbon dioxide in the blood, for example, in addition to using the value of the partial pressure of carbon dioxide in the blood (PaCO2), the dissolved concentration of carbon dioxide in the blood, the bicarbonate / bicarbonate contained in the blood (HCO3 - ) concentration, or a value taking into account the pH of the blood according to the situation. Hereinafter, the case where the value of the partial pressure of carbon dioxide in the blood is used as the characteristic of carbon dioxide in the blood will be described.

[0044] For example Figure 1 As shown, the biological information computing system 100 includes a biological information computing device 1, and may include at least one of a sensor 5 and a server 4. The biological information computing device 1 is connected to the sensor 5 and the server 4 via a communication network 3, for example.

[0045] In the biological information computing system 100, for example, Figure 2 As shown in (a) of FIG. 1 , the biological information computing device 1 acquires sensor data generated by the sensor 5 etc. Then, the biological information computing device 1 performs pre-processing such as filtering processing on the acquired data to acquire evaluation data.

[0046] The biological information calculation device 1 refers to the database to calculate the characteristics of blood carbon dioxide corresponding to the evaluation data (for example, the value of the partial pressure of carbon dioxide in the blood). The database stores, for example, classification information for calculating the value of the partial pressure of carbon dioxide in the blood from the evaluation data.

[0047] For example Figure 3As shown in (a), a pair of input data based on a learning pulse wave acquired in the past and reference data including a value of the partial pressure of carbon dioxide in the blood associated with the input data is used as learning data, and classification information is generated using a plurality of learning data. Therefore, data for subjects with clear health conditions, etc. can be used as learning data, so that more reliable classification information can be generated than when simply accumulated data is used. In addition, by referring to the classification information, a quantitative value of the partial pressure of carbon dioxide in the blood can be calculated without the calculation basis changing over time.

[0048] The biological information computing device 1 generates an evaluation result including the calculated value of the partial pressure of carbon dioxide in the blood, and outputs the evaluation result, for example, to a display, etc. The evaluation result may include, in addition to the value of the partial pressure of carbon dioxide in the blood, information indicating an evaluation corresponding to the value of the partial pressure of carbon dioxide in the blood, such as "normal" or "abnormal".

[0049] As described above, in the biological information computing system 100 of this embodiment, it is possible to generate an evaluation result including the characteristics of carbon dioxide in the blood corresponding to the evaluation data by referring to the database. Therefore, when generating the evaluation result, it is possible to include the characteristics of carbon dioxide in the blood quantitatively based on data with past results. Thus, it is possible to improve the evaluation accuracy.

[0050] In addition, in the biological information computing system 100, for example Figure 2 As shown in (b) of FIG. 1 , the evaluation data may be obtained from the sensor 5 or the like. In this case, the sensor 5 or the like performs preprocessing to obtain the evaluation data based on the sensor data.

[0051] <Sensor data>

[0052] The sensor data includes data representing the characteristics of the user's pulse wave, and may also include data representing characteristics other than the pulse wave (noise). The sensor data is data representing the amplitude corresponding to the measured time, and by performing filtering processing corresponding to the purpose and the generation conditions of the sensor data, data equivalent to the acceleration pulse wave, the velocity pulse wave, etc. can be obtained from the sensor data.

[0053] The sensor data can be generated by known sensors such as strain sensors, gyroscope sensors, photoplethysmography (PPG) sensors, and pressure sensors. The sensor data may be, for example, analog signals in addition to digital signals. In addition, the measurement time when generating the sensor data is, for example, a measurement time corresponding to 1 to 20 cycles of the pulse wave, and can be arbitrarily set according to conditions such as the sensor data processing method and the data communication method.

[0054] <Evaluation data>

[0055] The evaluation data indicates data for calculating the characteristics of carbon dioxide in the blood. The evaluation data indicates, for example, data corresponding to an accelerated pulse wave based on the pulse wave of the user, and indicates an amplitude corresponding to a specific cycle (for example, one cycle).

[0056] The evaluation data is obtained by processing (preprocessing) the sensor data by the biological information computing device 1 or the like. Figure 3 (b) and Figure 3 As shown in (c), by performing a plurality of processes on the sensor data, evaluation data can be obtained. The details of each process will be described later.

[0057] <Database>

[0058] The database is mainly used when calculating the characteristics of carbon dioxide in the blood corresponding to the evaluation data. In addition to storing one or more classification information, for example, a plurality of learning data used for generating the classification information may be stored in the database.

[0059] The classification information is, for example, a function indicating the correlation between the previously acquired past evaluation data (input data) and the reference data including the characteristics of the carbon dioxide in the blood. The classification information indicates, for example, a calibration model generated based on the analysis result by using regression analysis or the like with the input data as the explanatory variable and the reference data as the target variable. The classification information can be generated based on the attribute information such as the user's gender, age, and health status, in addition to being able to update the calibration model regularly.

[0060] As a regression analysis method used when generating classification information, for example, PLS (Partial Least Squares) regression analysis, SIMCA (Soft Independent Modeling of Class Analogy) regression analysis using a principal component model obtained by performing principal component analysis for each class, etc. can be used.

[0061] The classification information may include, for example, a learned model generated by machine learning using a plurality of learning data. The learned model may include, in addition to neural network models such as CNN (Convolutional Neural Network), SVM (Support vector machine) and the like. In addition, as machine learning, for example, deep learning may be used.

[0062] The input data uses the same type of data as the evaluation data, for example, past evaluation data showing clear characteristics of corresponding carbon dioxide in blood. The input data indicates data acquired based on the pulse wave (learning pulse wave) of a subject whose health status is clear.

[0063] For example, the subject is made to wear the sensor 5, etc., to generate sensor data (learning sensor data) representing the characteristics of the learning pulse wave. Then, by processing the learning sensor data, input data can be obtained. In addition, in addition to obtaining input data from the user of the biological information computing system 100, input data can also be obtained from a user different from the user. That is, the above-mentioned subject can be a user other than the user of the biological information computing system 100, or a specific or non-specific multiple user.

[0064] The input data is preferably obtained according to the same contents as the type of sensor 5 used when obtaining the evaluation data, the generation conditions of the sensor data, and the processing conditions for the sensor data. For example, by unifying the above three contents, the accuracy of calculating the characteristics of carbon dioxide in the blood can be greatly improved.

[0065] The reference data includes the characteristics of the carbon dioxide in the blood of the subject measured by the measuring device. For example, when the subject wears the sensor 5 and generates the learning sensor data, the reference data associated with the input data can be obtained by measuring the characteristics of the carbon dioxide in the blood of the subject. In this case, the timing of measuring the characteristics of the carbon dioxide in the blood is preferably the same as the timing of generating the learning sensor data, but it can also be a timing of about 1 to 10 minutes before and after, for example.

[0066] As a measuring device, for example, a known device such as a transcutaneous blood gas monitor TCM5 (manufactured by Radiometer Basel) can be used. In addition, a known device for measuring or estimating the partial pressure of carbon dioxide in the blood (PaCO2) can also be used. In addition, as a measuring device, for example, a device that can measure or estimate the bicarbonate / bicarbonate (HCO3 - ) concentration and pH in the blood. In addition, the reference data may include, in addition to the characteristics of carbon dioxide in the blood such as the value of the partial pressure of carbon dioxide in the blood, measured values ​​such as the value of oxygen saturation and the value of atmospheric pressure that can be associated with the characteristics of carbon dioxide in the blood.

[0067] <Biological information computing device 1>

[0068] The biological information computing device 1 is, for example, an electronic device such as a personal computer (PC), a mobile phone, a smart phone, a tablet terminal, a wearable terminal, and is, for example, an electronic device that can communicate via a communication network 3 based on a user's operation. In addition, the biological information computing device 1 may also have a built-in sensor 5. An example of using a PC as the biological information computing device 1 is described below.

[0069] Figure 4 (a) is a schematic diagram showing an example of the structure of the biological information computing device 1, Figure 4 (b) is a schematic diagram showing an example of the function of the biological information computing device 1.

[0070] For example Figure 4 As shown in (a) of FIG. 1 , the biological information computing device 1 includes a housing 10 , a CPU (Central Processing Unit) 101 , a ROM (Read Only Memory) 102 , a RAM (Random Access Memory) 103 , a storage unit 104 , and I / Fs 105 to 107 . The components 101 to 107 are connected via an internal bus 110 .

[0071] The CPU 101 controls the entire biological information computing device 1. The ROM 102 stores the operation code of the CPU 101. The RAM 103 is a work area used when the CPU 101 operates. The storage unit 104 stores various information such as databases and evaluation data. As the storage unit 104, for example, in addition to using a HDD (Hard Disk Drive), a data storage device such as an SSD (Solid State Drive) is also used. In addition, the biological information computing device 1 may also have a GPU (Graphics Processing Unit) not shown in the figure.

[0072] The I / F 105 is an interface for transmitting and receiving various information with the server 4, the sensor 5, etc. via the communication network 3 as needed. The I / F 106 is an interface for transmitting and receiving information with the input unit 108. As the input unit 108, for example, a keyboard is used, and the user of the biological information computing device 1 or the like inputs various information or control commands of the biological information computing device 1 or the like via the input unit 108. The I / F 107 is an interface for transmitting and receiving various information with the display unit 109. The display unit 109 displays various information or evaluation results stored in the storage unit 104. As the display unit 109, a display is used, and in the case of a touch panel type, for example, it is provided integrally with the input unit 108.

[0073] Figure 4 (b) is a schematic diagram showing an example of the functions of the biological information computing device 1. The biological information computing device 1 includes an acquisition unit 11, a generation unit 12, an output unit 13, and a storage unit 14, and may also include a learning unit 15. In addition, the CPU 101 uses the RAM 103 as a working area to execute a program stored in the storage unit 104, etc. Figure 4 The functions shown in (b).

[0074] <Acquisition Section 11>

[0075] The acquisition unit 11 acquires evaluation data based on the pulse wave of the user. The acquisition unit 11 acquires sensor data from the sensor 5 or the like, for example, and then processes the sensor data to acquire the evaluation data.

[0076] For example Figure 3 As shown in (b), the acquisition unit 11 performs filtering processing (filtering processing) on ​​the acquired sensor data. In the filtering process, for example, a bandpass filter of 0.5 to 5.0 Hz is used. Thus, the acquisition unit 11 extracts data (pulse wave data) corresponding to the user's pulse wave. The pulse wave data represents, for example, data corresponding to a velocity pulse wave. In addition, the pulse wave data may also represent, for example, data corresponding to an acceleration pulse wave or a volume pulse wave, and can be arbitrarily set according to the type and purpose of the sensor. In addition, the filtering range of the bandpass filter can be arbitrarily set according to the purpose.

[0077] The acquisition unit 11 performs differential processing on the pulse wave data, for example. For example, when performing differential processing on pulse wave data corresponding to a velocity pulse wave, the acquisition unit 11 acquires data (differential data) corresponding to an acceleration pulse wave. In the differential processing, in addition to the first differential, a second differential may be performed.

[0078] The acquisition unit 11 performs a segmentation process on the differential data, for example. In the segmentation process, for example, the differential data corresponding to the acceleration pulse wave of multiple cycles are segmented into data (segmented data) corresponding to each cycle of the acceleration pulse wave. Therefore, the acquisition unit 11 can acquire multiple segmented data by performing a differential process on one differential data, for example. In addition, in the segmentation process, the differential data can be segmented according to an arbitrary cycle (for example, a positive multiple of a cycle) according to the purpose.

[0079] For example, in the segmentation process, there is a case where the data amounts in each segmented data after segmentation are different. In this case, the acquisition unit 11 may also determine the segmented data with the smallest data amount and perform data reduction (trimming) on ​​the other segmented data. In this way, the data amounts in each segmented data can be unified, and the comparison of the data in each segmented data becomes easy.

[0080] In addition to the above, for example, the values ​​corresponding to the time axis of the segmented data may be normalized. In the normalization process, for example, the minimum value of the value corresponding to the time axis is set to 0 and the maximum value is set to 1. This makes it easier to compare the data in each segmented data.

[0081] For example, the acquisition unit 11 may calculate the average of a plurality of divided data on which data amount reduction or normalization has been performed, and use the average as the divided data.

[0082] The acquisition unit 11 performs normalization processing on the segmented data. In the normalization processing, normalized data (normalized data) is generated with the value corresponding to the amplitude as the object. In the normalization processing, for example, normalization is performed such that the lowest value of the amplitude is set to 0 and the highest value of the amplitude is set to 1. The acquisition unit 11 acquires the normalized data as evaluation data, for example. In this case, data equivalent to the user's acceleration pulse wave is obtained as evaluation data.

[0083] In addition to sequentially performing the above-mentioned processing, the acquisition unit 11 may also perform the following steps, for example: Figure 3 As shown in (c), the differentiation process may not be performed. In this case, data corresponding to the user's velocity pulse wave is obtained as evaluation data.

[0084] In addition, the acquisition unit 11 may, for example, perform only a part of each of the above-mentioned processes. In this case, the acquisition unit 11 may acquire any of the pulse wave data, differential data, segmented data, trimmed segmented data, and segmented data obtained by normalizing the values ​​corresponding to the time axis as evaluation data, and may set it arbitrarily according to the purpose.

[0085] <Generation section 12>

[0086] The generating unit 12 refers to the database and generates an evaluation result including the characteristics of the carbon dioxide in the blood (e.g., the value of the partial pressure of carbon dioxide in the blood) corresponding to the evaluation data. The generating unit 12, for example, refers to the classification information stored in the database to calculate the characteristics of the carbon dioxide in the blood corresponding to the evaluation data. The generating unit 12 generates the evaluation result by converting the characteristics of the carbon dioxide in the blood into a format that can be understood by the user, for example, using a display format pre-stored in the storage unit 104 or the like.

[0087] For example, when the classification information includes a learned model, the generator 12 may refer to the learned model and include the health status result indicating the health status of the user associated with the evaluation data in the evaluation result. In this case, the reference data used when generating the learned model includes status information indicating the health status of the subject whose learning pulse wave was measured. The status information includes, for example, information related to the current and past health status of the subject, such as "healthy", "possibility of ○○", "history of □□ disease in the past", etc.

[0088] In addition, the multiple learning data used when generating the learned model may also include, for example, only input data and reference data when the subject is healthy. In this case, in the evaluation result (health status result), the health status of the user associated with the evaluation data may also only indicate healthy or other than healthy. That is, the generation unit 12 may also generate the evaluation result in the following manner: when the evaluation data is similar to the learning data, "healthy" is shown as the health status result, and when the evaluation data deviates from the learning data, "other than healthy" is shown as the health status result. In addition, the threshold for classifying "healthy" and "other than healthy" can be arbitrarily set.

[0089] <Output section 13>

[0090] The output unit 13 outputs the evaluation result. The output unit 13 may output the evaluation result to the sensor 5 or the like, for example, in addition to outputting the evaluation result to the display unit 109 .

[0091] <Storage Unit 14>

[0092] The storage unit 14 retrieves various data such as a database stored in the storage unit 104 as needed. The storage unit 14 stores various data acquired or generated by each of the components 11 to 13 and 15 in the storage unit 104 as needed.

[0093] <Study Section 15>

[0094] The learning unit 15 generates classification information using, for example, a plurality of learning data. The learning unit 15 may acquire, for example, new learning data and update existing classification information.

[0095] <Communication Network 3>

[0096] The communication network 3 is the well-known Internet or the like that connects the biological information computing device 1, the server 4, and the sensor 5 via a communication line. When the biological information computing system 100 is used in a certain narrow area, the communication network 3 can be composed of a LAN (Local Area Network) or the like. In addition, the communication network 3 can also be composed of a so-called optical fiber communication network. In addition, the communication network 3 is not limited to a wired communication network, and can also be realized by a wireless communication network, and can be arbitrarily set according to the purpose.

[0097] The server 4 stores information transmitted via the communication network 3. The server 4 transmits the stored information to the biological information computing device 1 via the communication network 3 based on a request from the biological information computing device 1.

[0098] <Sensor 5>

[0099] The sensor 5 generates sensor data. For example, Figure 5 As shown in (a) of FIG. 5 , the sensor 5 includes a detection unit 6 . The sensor 5 is attached to a position where the pulse wave of the user can be detected via the detection unit 6 , and is fixed to a wristband 55 , for example.

[0100] The detection unit 6 uses a known detection device that can detect the pulse wave of the user. As the detection unit 6, for example, a strain sensor such as a fiber Bragg grating (FBG) sensor, a gyro sensor, one or more electrodes for measuring a pulse wave signal, a photoplethysmography (PPG) sensor, a pressure sensor, and at least one of a light detection module are used. For example, a plurality of detection units 6 may be configured.

[0101] In addition, the sensor 5 may also be embedded in clothing. In addition, the user wearing the sensor 5 may be a pet such as a dog or a cat, for example, a livestock such as a cow or a pig, or a farmed animal such as a fish, in addition to a human being.

[0102] For example Figure 5 As shown in (b) of FIG. 8 , the sensor 5 includes an acquisition unit 50 , a communication I / F 51 , a memory 52 , and a command unit 53 , and each of the components is connected via an internal bus 54 .

[0103] The acquisition unit 50 measures the pulse wave of the user via the detection unit 6 and generates sensor data. The acquisition unit 50 transmits the generated sensor data to the communication I / F 51 or the memory 52, for example.

[0104] The communication I / F 51 transmits various data such as sensor data to the biological information computing device 1 and the server 4 via the communication network 3. In addition, the communication I / F 51 is equipped with a line control circuit for connecting to the communication network 3, a signal conversion circuit for performing data communication between the biological information computing device 1 and the server 4, etc. The communication I / F 51 performs conversion processing on various commands from the internal bus 54 and sends them to the communication network 3 side. When receiving data from the communication network 3, it performs a predetermined conversion processing on it and sends it to the internal bus 54.

[0105] The memory 52 stores various data such as sensor data transmitted from the acquisition unit 50. The memory 52 transmits the stored various data such as sensor data to the communication I / F 51 by receiving a command from another terminal device connected via the communication network 3, for example.

[0106] The command unit 53 includes operation buttons and a keyboard for acquiring sensor data, and includes a processor such as a CPU. When the command unit 53 receives a command to acquire sensor data, it notifies the acquisition unit 50 of the command. The acquisition unit 50 that receives the notification acquires the sensor data. Figure 3 (b) and Figure 3 As shown in (c) of FIG. 5 , the command unit 53 may also implement processing for acquiring evaluation data based on sensor data.

[0107] Here, as an example of acquiring sensor data, a case of using an FBG sensor will be described.

[0108] The FBG sensor forms a diffraction grating structure at predetermined intervals in one optical fiber. The FBG sensor has the following features, for example: the length of the sensor portion is 10 mm, the wavelength resolution is ±0.1 pm, the wavelength range is 1550 ± 0.5 nm, the diameter of the optical fiber is 145 μm, and the core diameter is 10.5 μm. The FBG sensor can be used as the above-mentioned detection unit 6 to perform measurements in a state in which it is in contact with the user's skin.

[0109] As a light source for the optical fiber, for example, an ASE (Amplified Spontaneous Emission) light source with a wavelength range of 1525 to 1570 nm is used. The outgoing light from the light source is incident on the FBG sensor via a circulator. The reflected light from the FBG sensor is guided to the Mach-Zehnder interferometer via the circulator, and the output light from the Mach-Zehnder interferometer is detected by a photodetector. The Mach-Zehnder interferometer is an instrument for generating interference light by separating a light beam into two optical paths with an optical path difference through a beam splitter and superimposing them into one again through a beam splitter. In order to form an optical path difference, for example, the length of one optical fiber can also be lengthened. Coherent light generates interference fringes according to the optical path difference, so by measuring the pattern of the interference fringes, the change in strain generated in the FBG sensor, that is, the pulse wave, can be detected. The acquisition unit 50 generates sensor data based on the detected pulse wave. In this way, the sensor data is acquired.

[0110] In addition, the optical fiber sensor system that detects the waveform of the pulse wave by detecting the strain of the FBG sensor includes, in addition to the light source that causes the light beam to enter the FBG sensor, a unit that is a wideband ASE light source, an optical system such as a circulator, a Mach-Zehnder interferometer, and a beam splitter, a light receiving sensor provided in the photodetector, and an analysis unit that analyzes the wavelength shift. In the optical fiber sensor system, the light source and the frequency band light can be selected and used according to the characteristics of the FBG sensor used, and various methods can be adopted for the analysis method such as the detection method.

[0111] (Embodiment 1: Operation of the biological information calculation system 100)

[0112] Next, an example of the operation of the biological information calculation system 100 in this embodiment will be described. Figure 6 This is a flowchart showing an example of the operation of the biological information calculation system 100 in this embodiment.

[0113] The biological information computing system 100 is executed, for example, by a biological information computing program installed in the biological information computing device 1. That is, when a user operates the biological information computing device 1 or the sensor 5, the biological information computing program installed in the biological information computing device 1 can obtain an evaluation result including characteristics of carbon dioxide in the blood (for example, the value of the partial pressure of carbon dioxide in the blood) based on the sensor data.

[0114] The operation of the biological information calculation system 100 includes the acquisition step S110 and the generation step S120 , and may include, for example, the output step S130 .

[0115] <Acquisition Step S110>

[0116] In the acquisition step S110, evaluation data based on the pulse wave of the user is acquired. For example, the acquisition unit 50 of the sensor 5 measures the pulse wave of the user via the detection unit 6 and generates sensor data. The acquisition unit 50 transmits the sensor data to the biological information computing device 1 via the communication I / F 51 and the communication network 3. The acquisition unit 11 of the biological information computing device 1 receives the sensor data from the sensor 5.

[0117] The acquisition unit 11 performs, for example, Figure 3 The acquisition unit 11 stores the acquired evaluation data in the storage unit 104 via the storage unit 14, for example. In addition, the conditions such as the frequency of acquisition of sensor data from the sensor 5 by the acquisition unit 11 can be arbitrarily set according to the application.

[0118] <Generation Step S120>

[0119] Next, in the generation step S120, the evaluation result including the characteristics of the carbon dioxide in the blood (e.g., the value of the partial pressure of carbon dioxide in the blood) corresponding to the evaluation data is generated with reference to the database. For example, the generation unit 12 refers to the classification information and calculates the value of the partial pressure of carbon dioxide in the blood corresponding to the evaluation data. The generation unit 12 generates an evaluation result including the calculated value. The generation unit 12 stores the generated evaluation result in the storage unit 104, for example, via the storage unit 14. In addition, as the value of the partial pressure of carbon dioxide in the blood, in addition to showing a specific value, for example, an error range (e.g., "○○±2 mmHg", etc.) can also be calculated.

[0120] For example, the generator 12 may calculate the values ​​of the blood carbon dioxide partial pressure corresponding to a plurality of evaluation data at different measurement times and include each value in the evaluation result. In this case, the user can easily understand the temporal change and average value of the blood carbon dioxide partial pressure.

[0121] <Output Step S130>

[0122] Next, for example, in the output step S130 , the evaluation result may be output. For example, the output unit 13 outputs the evaluation result to the display unit 109 .

[0123] Thus, the operation of the biological information calculation system 100 is completed. In addition, the frequency and order of executing each step can be arbitrarily set according to the application.

[0124] According to this embodiment, the generating unit refers to the database and generates an evaluation result including the characteristics of carbon dioxide in the blood corresponding to the evaluation data. In addition, the database stores classification information calculated using a plurality of learning data. Therefore, when generating the evaluation result, it is possible to include quantitative characteristics of carbon dioxide in the blood based on data with past performance (input data and characteristics of carbon dioxide in the blood). Thus, it is possible to improve the evaluation accuracy.

[0125] In addition, according to this embodiment, the generation unit can refer to the learned model and include the health status result associated with the evaluation data in the evaluation result for generation. Therefore, the health status based on the characteristics of carbon dioxide in the blood can be reported to the user, and the health status can be easily grasped compared to the case where only the characteristics of carbon dioxide in the blood are reported. As a result, the convenience for the user can be improved.

[0126] In addition, according to the present embodiment, the plurality of learning data only includes input data and reference data when the subject is healthy, and the health status result indicates whether the health status of the user associated with the evaluation data is healthy. Therefore, even when it is impossible to prepare input data based on the learning pulse wave when unhealthy and the characteristics of carbon dioxide in the blood, a result indicating the health status of the user can be generated. Thus, it is possible to further improve the convenience for the user without increasing the difficulty of generating classification information.

[0127] In addition, according to the present embodiment, the classification information is a calibration model obtained by using PLS regression analysis with input data as explanatory variables and reference data as target variables. Therefore, compared with the case where the classification information is calculated using machine learning, the amount of learning data can be greatly reduced, and the calibration model can be easily updated. Thus, the construction and update of the biological information computing system 100 can be facilitated.

[0128] (Second Embodiment: Biological Information Calculation System 100)

[0129] Next, an example of the biological information calculation system 100 in the second embodiment is described. The above embodiment differs from the second embodiment in that preliminary evaluation data based on the pulse wave is acquired in the second embodiment. In addition, description of the same contents as those in the above embodiment is omitted.

[0130] The biological information computing device 1 in this embodiment is, for example, Figure 7 As shown, two types of processing are performed on one sensor data generated based on the pulse wave. Thus, the biological information computing device 1 obtains evaluation data and preliminary evaluation data different from the evaluation data. Alternatively, a plurality of preliminary evaluation data may be obtained.

[0131] The biological information computing device 1 calculates biological information corresponding to the preliminary evaluation data, and generates, for example, a preliminary evaluation result including the biological information. For example, in addition to the evaluation result, the biological information computing device 1 can also output the preliminary evaluation result. In addition, for example, in the biological information computing device 1, the evaluation result can also include the biological information.

[0132] As biological information, for example, at least one of the pulse rate and the respiratory rate can be calculated. In this case, for example, Figure 8 (a) and Figure 8 As shown in (b) of FIG. 1 , the content of the preprocessing performed in the acquisition step S110 and the content of the database referred to in the generation step S120 may be different. Hereinafter, an example of calculating the pulse rate and the respiratory rate as biological information will be described.

[0133] <Calculation of pulse rate>

[0134] For example, in the case of counting pulses, Figure 8 As shown in (a) of FIG. 1 , the acquisition unit 11 performs the above-mentioned filtering process on the sensor data to extract the pulse wave data.

[0135] Then, the acquisition unit 11 performs peak position calculation processing on the pulse wave data. In the peak position calculation processing, multiple peaks (maximum values ​​of amplitude) included in the pulse wave data are detected, and the sampling order (equivalent to the time from the start of measurement) is determined. In this way, the acquisition unit 11 acquires the peak position data included in the pulse wave data.

[0136] Then, the acquisition unit 11 performs a peak interval average calculation process on the peak position data. In the peak interval average calculation process, the interval between peaks included in the peak position data (the difference in the order in which adjacent peaks are sampled) is calculated, for example, the average value of the peak interval is calculated. Then, the acquisition unit 11 divides the peak interval or the average value of the peak interval by the sampling rate of the sensor data, and acquires data representing the peak interval equivalent to seconds as preliminary evaluation data.

[0137] Then, the generator 12 refers to the database and calculates the pulse rate corresponding to the preliminary evaluation data. At this time, the generator 12 refers to the pulse rate classification information stored in the database. The pulse rate classification information, for example, represents a function of dividing 60 [seconds] by the peak interval. Therefore, the generator 12 can calculate the pulse rate (= 71 [bpm]) corresponding to the preliminary evaluation data (peak interval = 0.85 [seconds]), for example. Thus, the generator 12 can generate an evaluation result or a preliminary evaluation result including the pulse rate.

[0138] <Calculation of breathing number>

[0139] For example, in the case of counting the number of breaths, Figure 8 As shown in (b) of FIG. 1 , the acquisition unit 11 performs the above-mentioned filtering process on the sensor data and extracts the pulse wave data.

[0140] Then, the acquisition unit 11 performs Fourier transform processing on the pulse wave data. In the Fourier transform processing, for example, the pulse wave data representing sampling time versus amplitude is transformed into frequency data representing frequency versus intensity. Thus, the acquisition unit 11 acquires frequency data corresponding to the pulse wave data.

[0141] Then, the acquisition unit 11 performs a maximum frequency detection process on the frequency data. In the maximum frequency detection process, the frequency of the maximum intensity between 0.15 and 0.35 Hz in the frequency data is determined. Thus, the acquisition unit 11 acquires the value of the determined frequency as preliminary evaluation data.

[0142] After that, the generation unit 12 refers to the database and calculates the respiratory rate corresponding to the preliminary evaluation data. At this time, the generation unit 12 refers to the classification information for the respiratory rate stored in the database. The classification information for the respiratory rate, for example, represents a function of multiplying the determined frequency by 60 [seconds]. Therefore, the generation unit 12 can, for example, calculate the respiratory rate (= 13.5 [bpm]) corresponding to the preliminary evaluation data (determined frequency = 0.225 Hz). Thus, the generation unit 12 can, for example, generate an evaluation result or a preliminary evaluation result including the respiratory rate.

[0143] In this way, in the biological information computing device 1 , it is possible to set pre-processing for sensor data according to the content of biological information included in the preliminary evaluation result, and to arbitrarily set the content of a database to be referred to.

[0144] Furthermore, as biological information, in addition to the above-mentioned pulse rate and respiration rate, for example, information that can be calculated based on the pulse wave, such as pressure, blood sugar level, blood pressure, oxygen saturation, blood vessel age, and degree of diabetes, can be used.

[0145] In addition, preliminary classification information for calculation may be stored in advance in a database according to the content of the biological information to be calculated. In this case, a pair of preliminary input data representing the above-mentioned learning pulse wave and preliminary reference data containing biological information associated with the preliminary input data is used as preliminary learning data, and a plurality of preliminary learning data are prepared. Then, for example, the learning unit 15 generates preliminary classification information using the plurality of preliminary learning data, and stores the preliminary classification information in the database.

[0146] In addition, the biometric information included in the preliminary reference data can be obtained, for example, by measuring the subject using a measuring device for measuring the necessary biometric information, similarly to the reference data described above. In addition, the learning method of the preliminary classification information can use the same method as the classification information described above.

[0147] In addition to the above, in the biological information computing device 1, for example Fig. 9 As shown, a comprehensive evaluation result can be generated based on the evaluation result and the preliminary evaluation result, and outputted, etc. Therefore, in the biological information calculation system 100, preliminary evaluation results generated from a different viewpoint from the evaluation result can be used, and various evaluations can be performed.

[0148] In the biological information computing device 1, for example, by implementing Figure 3 (b) and Figure 3 The two types of processing shown in (c) of the present invention are used to obtain evaluation data and preliminary evaluation data based on sensor data. That is, when data equivalent to the acceleration pulse wave of the user is obtained as evaluation data, data equivalent to the velocity pulse wave of the user is obtained as preliminary evaluation data. In this case, by using the evaluation result generated for the evaluation data and the preliminary evaluation result generated for the preliminary evaluation result, a more accurate evaluation can be achieved.

[0149] According to this embodiment, in addition to the effects of the above-mentioned embodiment, the generation unit refers to the database to generate a preliminary evaluation result including the biological information of the user corresponding to the preliminary evaluation data. Therefore, it is possible to use the preliminary evaluation result generated from a different point of view from the evaluation result, and to implement various evaluations. Thus, it is possible to implement an appropriate evaluation corresponding to the user's requirements.

[0150] In addition, according to the present embodiment, the generation unit generates a preliminary evaluation result corresponding to the preliminary evaluation data with reference to the preliminary classification information. Therefore, it is possible to use the preliminary evaluation result generated with reference to information different from the evaluation result, and to implement a more comprehensive evaluation. Thus, it is possible to easily implement an appropriate evaluation corresponding to the user's requirements.

[0151] (Third Embodiment: Biological Information Calculation System 100)

[0152] Next, an example of the biological information calculation system 100 in the third embodiment is described. The difference between the above embodiment and the third embodiment is that in the third embodiment, classification information suitable for the evaluation data is selected from a plurality of classification information. In addition, the description of the same contents as the above embodiment is omitted.

[0153] The biological information computing device 1 in this embodiment is, for example, Fig.10 As shown, referring to the preliminary evaluation data, specific classification information (e.g., the first classification information) is selected from a plurality of attribute classification information (selection unit). In addition, referring to the selected first classification information, the biological information computing device 1 calculates the characteristics of carbon dioxide in the blood corresponding to the evaluation data (e.g., the value of the partial pressure of carbon dioxide in the blood), and generates an evaluation result (attribute generation unit).

[0154] A plurality of attribute classification information is calculated using different learning data. For example, when using data corresponding to the acceleration pulse wave of the subject as input data for learning data, for example, Fig.11 Such 7 types (A to G) of input data are prepared to generate 7 types of attribute classification information.

[0155] When such a plurality of attribute classification information is stored in the database, for example, the acquisition unit 11 acquires the evaluation data and the preliminary evaluation data corresponding to the acceleration pulse wave of the user. Then, the generation unit 12 refers to the preliminary evaluation data and selects the first classification information. Thereafter, the generation unit 12 refers to the first classification information and calculates the characteristics of the carbon dioxide in the blood corresponding to the evaluation data (for example, the value of the partial pressure of carbon dioxide in the blood). Therefore, the classification information most suitable for the user can be selected from each attribute classification information.

[0156] In addition, for example, when using data corresponding to the velocity pulse wave of the subject as input data for learning data, for example, Fig.12 Such two types (group 1 and group 2) of input data are prepared, and two types of attribute classification information are generated.

[0157] Here, about Fig.11 The data equivalent to the accelerated pulse wave shown in FIG. 1 is easy to perform detailed classification based on characteristics, but on the other hand, when calculating the characteristics of carbon dioxide in the blood, there is a possibility of false detection of the peak value, which may lead to a decrease in accuracy. In addition, compared with the data equivalent to the accelerated pulse wave, it is difficult to Fig.12 The data corresponding to the velocity pulse wave shown is classified in detail based on the characteristics, but since there are few false detections of peaks, etc., the characteristics of carbon dioxide in the blood can be calculated with high accuracy.

[0158] Based on the above, as selection data used to select specific classification information, a plurality of attribute classification information may include, for example, Fig.11 Such data corresponding to the acceleration pulse wave is used, and data corresponding to the velocity pulse wave is used as learning data when generating attribute classification information.

[0159] In this case, as an acquisition unit, for example, the acquisition unit 11 acquires data equivalent to the velocity pulse wave as evaluation data based on sensor data of the user's pulse wave. In addition, the acquisition unit 11 acquires data equivalent to the acceleration pulse wave as preliminary evaluation data based on the sensor data.

[0160] Next, as a selection unit, for example, the generation unit 12 refers to the preliminary evaluation data, determines the selection data (first selection data) most similar to the preliminary evaluation data among a plurality of selection data including the data equivalent to the acceleration pulse wave, and selects the first classification information associated with the first selection data. Then, as an attribute generation unit, the generation unit 12 refers to the first classification information and generates an evaluation result corresponding to the evaluation data. Thereby, further improvement in evaluation accuracy can be achieved.

[0161] Here, an example of the data used in the above selection data and the like will be described.

[0162] For example, as Fig.11 shown, there are inflection points a to e in the acceleration pulse wave. For example, in the case of performing the following normalization, a method of classifying the acceleration pulse wave into 7 patterns can be used based on the magnitude relationship between the values of each inflection point and their differences. In this normalization, the maximum peak value in the acceleration pulse wave is set as point a, and each inflection point is sequentially set as point b, point c, point d, and point e starting from point a. Point a is set to 1, and the minimum value point b or point d is set to 0. First, when the value of the inflection point is b < d, it is classified into pattern A or B. If b < d and c ≥ 0.5, it is classified into A, otherwise it is classified into B. Next, when the value of the inflection point is b ≒ d, it is classified into pattern C or D. When b ≒ d and c ≒ 0, it is classified into pattern D, otherwise it is classified into pattern C. Finally, when b > d, it can be classified into any of patterns E, F, and G. If b > d and b < c, it is classified into pattern E, if b ≒ c, it is classified into pattern F, and if b > c, it is classified into pattern G.

[0163] For example, the generation unit 12 determines which pattern the preliminary evaluation data, for example, conforms to Fig.11 and determines the first selection data. For example, if the inflection point b of the input preliminary evaluation data is smaller than the inflection point d and the inflection point c ≥ 0.5, then pattern A is used as the first selection data. Thereby, it is possible to calculate the characteristics of carbon dioxide in the blood with high accuracy by referring to the classification information suitable for the characteristics of the evaluation data.

[0164] According to this embodiment, in addition to the effects of the above-mentioned embodiment, the generation unit includes: a selection unit that refers to the preliminary evaluation data and selects the first classification information; and an attribute-based generation unit that refers to the first classification information and generates an evaluation result corresponding to the evaluation data. Therefore, it is possible to generate an evaluation result corresponding to the evaluation data based on the selection of the attribute-based classification information that is most suitable for the characteristics of the pulse wave. As a result, it is possible to further improve the evaluation accuracy.

[0165] In addition, according to the present embodiment, the acquisition unit acquires data equivalent to the velocity pulse wave based on the pulse wave as evaluation data. In addition, the acquisition unit acquires data equivalent to the acceleration pulse wave based on the pulse wave as preliminary evaluation data. Therefore, it is possible to select attribute classification information using the acceleration pulse wave that is easier to classify the characteristics of the pulse wave than the velocity pulse wave. In addition, it is possible to generate an evaluation result using the velocity pulse wave that is easier to calculate the characteristics of carbon dioxide in the blood than the acceleration pulse wave. Thus, it is possible to achieve further improvement in evaluation accuracy.

[0166] In addition, in each of the above-mentioned embodiments, as the characteristic of carbon dioxide in blood, the situation of using the value of partial pressure of carbon dioxide in blood is mainly described, but the situation of using at least any one of the concentration of carbon dioxide in blood, the concentration of bicarbonate / bicarbonate contained in blood and the pH of blood is also the same, so the description is omitted. In particular, when the value of partial pressure of carbon dioxide in blood is used as the characteristic of carbon dioxide in blood, there is a tendency to easily collect reference data compared with other values. Therefore, it is possible to achieve a significant reduction in the time and expense spent when generating classification information.

[0167] The embodiments of the present invention have been described, but the embodiments are shown as examples and are not intended to limit the scope of the invention. Such new embodiments can be implemented in various other ways, and various omissions, substitutions, and changes can be made without departing from the scope of the main purpose of the invention. The embodiments and their variations are included in the scope and main purpose of the invention, and are included in the invention described in the claims and their equivalents.

[0168] Description of symbols

[0169] 1: Biological information computing device

[0170] 3: Communication network

[0171] 4: Server

[0172] 5: Sensor

[0173] 6: Detection Department

[0174] 10: Shell

[0175] 11: Acquisition

[0176] 12: Generation Department

[0177] 13: Output section

[0178] 14: Storage

[0179] 15: Learning Department

[0180] 50: Acquisition

[0181] 51: Communication I / F

[0182] 52: Memory

[0183] 53: Command Department

[0184] 54: Internal bus

[0185] 55: Wristband

[0186] 100: Biological Information Computing System

[0187] 101: CPU

[0188] 102: ROM

[0189] 103: RAM

[0190] 104: Preservation Department

[0191] 105: I / F

[0192] 106: I / F

[0193] 107: I / F

[0194] 108: Input section

[0195] 109: Display unit

[0196] 110: Internal bus

[0197] S110: Acquisition Steps

[0198] S120: Generation Steps

[0199] S130: Output step

Claims

1. A biological information computing system that outputs information related to characteristics of carbon dioxide in a user's blood, The biological information computing system is characterized by comprising: an acquisition unit that acquires evaluation data based on the pulse wave of the user; A database storing classification information generated by using a pair of input data and reference data as learning data and using a plurality of the learning data, wherein The input data is data based on a learning pulse wave acquired in advance and is the same type of data as the evaluation data, and the reference data is data including characteristics of carbon dioxide in the blood associated with the input data; as well as a generating unit for generating an evaluation result including a characteristic of the carbon dioxide in the blood corresponding to the evaluation data by referring to the database, The acquisition unit acquires preliminary evaluation data different from the evaluation data based on the pulse wave, The classification information includes a plurality of attribute-based classification information calculated using different learning data, respectively. The generating unit comprises: A selection unit for selecting a first classification information from among the plurality of attribute classification information with reference to the preliminary evaluation data; as well as an attribute-based generating unit for generating the evaluation result corresponding to the evaluation data with reference to the first classification information; The acquisition unit acquires data reflecting a velocity pulse wave based on the pulse wave as the evaluation data, and acquires data reflecting an acceleration pulse wave based on the pulse wave as the preliminary evaluation data.

2. The biological information computing system according to claim 1, characterized in that: The reference data includes status information indicating the health status of the subject for whom the learning pulse wave is measured. The classification information includes a learned model generated by machine learning using a plurality of the learning data. The generating unit refers to the learned model and generates a health status result indicating the health status of the user associated with the evaluation data by including it in the evaluation result.

3. The biological information computing system according to claim 2, characterized in that: The plurality of learning data only include the input data and the reference data when the subject is healthy, The health status result indicates whether the health status of the user associated with the evaluation data is healthy or other than healthy.

4. The biological information computing system according to claim 1, characterized in that: The classification information is a calibration model obtained using a partial least squares regression analysis using the input data as an explanatory variable and the reference data as a target variable.

5. The biological information computing system according to claim 1, characterized in that: The acquisition unit acquires preliminary evaluation data different from the evaluation data based on the pulse wave, The generating unit refers to the database and calculates the biological information of the user corresponding to the preliminary evaluation data.

6. The biological information computing system according to claim 5, characterized in that: The database stores preliminary classification information generated by using a plurality of preliminary learning data by using a pair of preliminary input data and preliminary reference data as preliminary learning data, wherein the preliminary input data is data based on the learning pulse wave and is of the same type as the preliminary evaluation data, and the preliminary reference data is data including the biological information associated with the preliminary input data, The generating unit calculates the biological information corresponding to the preliminary evaluation data with reference to the preliminary classification information.

Images