WO2022201639A1 - 生体データ評価サーバ、生体データ評価システム及び生体データ評価方法 - Google Patents
生体データ評価サーバ、生体データ評価システム及び生体データ評価方法 Download PDFInfo
- Publication number
- WO2022201639A1 WO2022201639A1 PCT/JP2021/043239 JP2021043239W WO2022201639A1 WO 2022201639 A1 WO2022201639 A1 WO 2022201639A1 JP 2021043239 W JP2021043239 W JP 2021043239W WO 2022201639 A1 WO2022201639 A1 WO 2022201639A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- period
- data
- unit
- evaluation server
- biometric data
- Prior art date
Links
- 238000011157 data evaluation Methods 0.000 title claims abstract description 131
- 238000000034 method Methods 0.000 title claims description 64
- 238000013480 data collection Methods 0.000 claims abstract description 30
- 230000006870 function Effects 0.000 claims description 117
- 210000000467 autonomic pathway Anatomy 0.000 claims description 103
- 206010003119 arrhythmia Diseases 0.000 claims description 96
- 230000001684 chronic effect Effects 0.000 claims description 84
- 230000002159 abnormal effect Effects 0.000 claims description 83
- 238000012545 processing Methods 0.000 claims description 65
- 238000011156 evaluation Methods 0.000 claims description 58
- 238000004458 analytical method Methods 0.000 claims description 56
- 230000008569 process Effects 0.000 claims description 46
- 230000002776 aggregation Effects 0.000 claims description 24
- 238000004220 aggregation Methods 0.000 claims description 24
- 238000004364 calculation method Methods 0.000 claims description 23
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000005259 measurement Methods 0.000 description 65
- 230000005856 abnormality Effects 0.000 description 60
- 238000010586 diagram Methods 0.000 description 26
- 238000000605 extraction Methods 0.000 description 20
- 238000001514 detection method Methods 0.000 description 16
- 230000006793 arrhythmia Effects 0.000 description 13
- 230000005540 biological transmission Effects 0.000 description 13
- 238000013058 risk prediction model Methods 0.000 description 11
- 230000000694 effects Effects 0.000 description 9
- 230000001133 acceleration Effects 0.000 description 8
- 238000004891 communication Methods 0.000 description 5
- 230000036541 health Effects 0.000 description 5
- 238000012935 Averaging Methods 0.000 description 4
- 101100269850 Caenorhabditis elegans mask-1 gene Proteins 0.000 description 4
- 230000037424 autonomic function Effects 0.000 description 4
- 230000007704 transition Effects 0.000 description 4
- 230000006399 behavior Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000033001 locomotion Effects 0.000 description 3
- 230000000284 resting effect Effects 0.000 description 3
- 230000003595 spectral effect Effects 0.000 description 3
- 206010003658 Atrial Fibrillation Diseases 0.000 description 2
- 208000002102 Atrial Premature Complexes Diseases 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 2
- 206010042602 Supraventricular extrasystoles Diseases 0.000 description 2
- 208000009729 Ventricular Premature Complexes Diseases 0.000 description 2
- 230000004931 aggregating effect Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000011158 quantitative evaluation Methods 0.000 description 2
- 208000000418 Premature Cardiac Complexes Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000001746 atrial effect Effects 0.000 description 1
- 206010003668 atrial tachycardia Diseases 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000004397 blinking Effects 0.000 description 1
- 230000036760 body temperature Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 230000004424 eye movement Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000009291 secondary effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000002861 ventricular Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
- G16H40/63—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for local operation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
- A61B5/02405—Determining heart rate variability
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
- A61B5/0245—Detecting, measuring or recording pulse rate or heart rate by using sensing means generating electric signals, i.e. ECG signals
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/165—Evaluating the state of mind, e.g. depression, anxiety
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
- A61B5/346—Analysis of electrocardiograms
- A61B5/349—Detecting specific parameters of the electrocardiograph cycle
- A61B5/352—Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/40—Detecting, measuring or recording for evaluating the nervous system
- A61B5/4029—Detecting, measuring or recording for evaluating the nervous system for evaluating the peripheral nervous systems
- A61B5/4035—Evaluating the autonomic nervous system
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/74—Details of notification to user or communication with user or patient ; user input means
- A61B5/746—Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms
Definitions
- the present invention relates to a biometric data evaluation server, system, and method used to evaluate the reliability of biometric data measured when promoting safe operation of transportation.
- BBI heartbeat interval
- a Lorenz plot (LP), which will be described later, is created, and the created Lorenz plot Disclosed is a medical device capable of detecting arrhythmia by distinguishing between atrial fibrillation and atrial tachycardia, which are types of arrhythmia, using plot-related feature values.
- the present invention has been made in view of the above problems, and an object of the present invention is to make it possible to evaluate the degree of arrhythmia-like abnormal value occurrence even for heartbeat interval data with missing data.
- the present invention is a biometric data evaluation server having a processor and a memory for evaluating biometric data, comprising: a data collection unit that receives heartbeat interval equivalent data from the biometric data of a subject; and a Lorenz plot generation unit that calculates a Lorenz plot in the period of and outputs it as an aggregated Lorenz plot.
- the degree of arrhythmia-like abnormal value generation can be evaluated from the generated Lorenz plot even for heartbeat interval data with missing or missing data. This makes it possible to evaluate the reliability of the autonomic nerve function index obtained from the heartbeat interval data measured during work.
- Example 1 of this invention shows an example of a structure of a biological data evaluation system.
- 4 is a flowchart showing Embodiment 1 of the present invention and showing an example of RRI data transmission processing performed by the vehicle driving data collection device.
- 4 is a flow chart showing Example 1 of the present invention and showing an example of learning processing of a chronic-occurrence discrimination model for arrhythmia-like abnormal values performed by a biological data evaluation server.
- It is a figure which shows Example 1 of this invention and shows an example of the mask area
- Example 1 of this invention shows an example of a measurement defect-like mask used by a biological data evaluation server. It is a figure which shows Example 1 of this invention and shows an example of the normal range mask utilized by a biological data evaluation server. It is a figure which shows Example 1 of this invention and shows an example of the PVC-like abnormal value mask utilized with a biological data evaluation server. It is a figure which shows Example 1 of this invention and shows an example of the 1 beat detection omission mask utilized by a biological data evaluation server. It is a figure which shows Example 1 of this invention and shows an example of the consecutive 2 beat detection omission mask utilized by a biological data evaluation server.
- Example 4 is a flowchart showing Example 1 of the present invention and showing an example of processing for evaluating the degree of chronic occurrence of arrhythmia-like abnormal values performed by a biological data evaluation server. It is a flowchart which shows Example 1 of this invention and shows an example of the determination process of the unsuitable driver for the autonomic-nervous function evaluation performed by a biological data evaluation server. It is a flowchart which shows Example 1 of this invention and shows an example of the calculation process of the autonomic-nerve function index performed by a biological data evaluation server. It is a figure which shows Example 1 of this invention and shows an example of an autonomic-nerve function evaluation result screen.
- FIG. 10 is a diagram showing the first embodiment of the present invention and showing an example of the data structure of RRI data in units of period B;
- FIG. 10 is a diagram showing the first embodiment of the present invention and showing an example of the data structure of period B unit LP data;
- FIG. 10 is a diagram showing the first embodiment of the present invention and showing an example of the data structure of period A unit aggregated LP data;
- FIG. 10 is a diagram showing the first embodiment of the present invention and showing an example of the data structure of period A unit aggregated LP data;
- FIG. 10 is a diagram showing the first embodiment of the present invention and showing an example of the data structure of aggregated LP feature amount data for each period A; It is a figure which shows Example 1 of this invention and shows an example of the data structure of abnormality degree data.
- FIG. 10 is a diagram showing Example 1 of the present invention and showing an example of a data structure of period A unit abnormality determination data; It is a figure which shows Example 1 of this invention and shows an example of the data structure of unsuitability determination data. It is a figure which shows Example 1 of this invention and shows an example of the data structure of autonomic-nerve function index data. It is a figure which shows Example 1 of this invention and shows an example of the data structure of business status data.
- FIG. 4 is a diagram showing the first embodiment of the present invention, and showing an example related to the definition of period A and period B and the occurrence of data loss
- FIG. 10 is a block diagram showing an example 2 of the present invention and showing an example of a main configuration when the biological data evaluation system predicts the risk of a driver's traffic accident using autonomic nerve function index data
- FIG. 12 shows the second embodiment of the present invention, and shows an example of a prediction result notification screen that is notified to the driver when an accident risk prediction result and an increase in accident risk are detected.
- Example 1 of the present invention will be described.
- FIG. 1 is a block diagram showing Embodiment 1 of the present invention and showing an example of the main configuration of a biological data evaluation system.
- the biometric data evaluation system of this embodiment includes a biometric data evaluation server 1 that processes data received from one or more vehicles 7 via a network 13 .
- the vehicle 7 collects a biometric sensor 12 for detecting biometric data of the driver, a driver ID reader 11 for identifying the driver, and collects the detected biometric data and the driver ID, and transmits the driving data to the biometric data evaluation server 1.
- the biosensors 12 include a heartbeat sensor 14 that detects RRI and an acceleration sensor 15 that detects the motion of the driver.
- a sensor that detects heartbeat based on electrocardiogram, pulse wave, heart sound, or the like can be used.
- the biosensor 12 is not limited to the above, and in addition to the heartbeat sensor 14, a sensor that detects the amount of perspiration, body temperature, blinking, eye movement, myoelectricity, electroencephalogram, or the like can be adopted.
- the biosensors 12 include wearable devices that can be worn by the driver, sensing devices attached to the inside of the vehicle such as steering wheels, seats, seat belts, etc., and an image recognition system that captures the driver's expression and behavior and analyzes the image. can be used.
- the heartbeat sensor 14 may detect heartbeat intervals other than RRI, such as PPI, which is a pulse wave interval. Alternatively, it is also possible to obtain heartbeat interval equivalent data by estimating the heartbeat interval from the image data of the driver's face. In this embodiment, any biometric data that can calculate a heartbeat interval can be used, and the data corresponding to the heartbeat interval as described above can be included.
- the driver ID reading device 11 reads a card that records the identifier of the driver.
- the driving data collection device 10 collects data from the biosensor 12 at a predetermined cycle, and transmits the data to the biometrics data evaluation server 1 via the network 13 .
- the driver ID reading device 11 is configured as a device for reading a card on which the identifier of the driver is recorded, but may be configured differently.
- the driver ID reading device 11 is composed of one portable terminal and the portable terminal is made to function as a driver ID reading unit, the driver can read the driver ID by inputting the identifier of the driver himself, or read the driver ID.
- the driver ID may be read by identifying the driver by a known face authentication technique using a camera.
- the biometric data evaluation server 1 is a computer including a processor 2, a memory 3, a storage device 4, an input/output device 5, and a communication device 6.
- the memory 3 includes a data collection unit 21, a chronic occurrence discrimination model learning unit 22, a period B unit LP generation unit 23, a period A unit aggregation LP processing unit 24, an abnormal value chronic occurrence evaluation unit 25, and an unsuitable driver determination unit.
- Each functional unit of the unit 26, the autonomic function index calculator 27, and the result display unit 28 is loaded as a program.
- Each program is executed by a processor. Details of each functional unit will be described later.
- the processor 2 operates as a functional unit that provides a predetermined function by executing processing according to the program of each functional unit.
- the processor 2 functions as the chronic abnormal value occurrence evaluation unit 25 by executing a chronic abnormal value occurrence evaluation program. The same is true for other programs.
- the processor 2 also operates as a functional unit that provides functions of multiple processes executed by each program.
- Computers and computer systems are devices and systems that include these functional units.
- the storage device 4 stores data used by each of the above functional units.
- the storage device 4 stores period B unit RRI data 41, period B unit LP data 43, period A unit aggregated LP data 45, aggregated LP feature amount data 47, anomaly degree data 49, and period A unit anomaly determination data.
- 51, autonomic nerve function index data 53, work status data 42, history data 44, chronic occurrence correct data 46, inappropriate determination data 48, feature extraction model 50, chronic occurrence determination model 52, inappropriate driver and the determination model 54 are stored.
- LP is an abbreviation for Lorenz plot (hereinafter also referred to as LP, also referred to as Poincaré plot), and the same notation is used in the latter part.
- the input/output device 5 includes input devices such as a mouse, keyboard, touch panel, or microphone, and output devices such as a display and speakers.
- the communication device 6 communicates with the vehicle via the network 13 .
- the biological data measured from the driver who controls the vehicle 7 is exemplified, but the form is not limited to this embodiment.
- the target may be a driver who operates a moving object such as an airplane or a train.
- the target may be not only drivers but also general employees who are not limited to drivers who measure biometric data during work, or people who lead daily lives without being limited to working.
- FIG. 2 is a flowchart showing an example of RRI data transmission processing performed by the vehicle driving data collection device 10.
- the first period A which is a target period for evaluating the degree of occurrence of chronic arrhythmia-like abnormal values, is set to, for example, one business day, and the second period B having a window width equal to or less than the period A. , for example, in units of 2 minutes.
- period A and period B may be determined by an interval defining a time width as in this embodiment, or may be determined by the number of data points. For example, period A may be set to be equal to one business day and period B may be set to 120 data points.
- the period B is preferably equal to or shorter than the first period A and equal to or longer than the time width required for heart rate variability analysis. Heart rate variability analysis is difficult with only one beat, and it takes about 10 seconds. Also, the period B may be determined based on information on the loss of RRI data received from the input device or the time width of the loss.
- the driving data collection device 10 when it receives a login input 71 from the driver, it starts RRI data measurement (S10). For example, as the login input 71, an event in which a card on which the identifier of the driver is recorded using the driver ID reader 11 may be used as the login input 71.
- FIG. 10 As the login input 71, an event in which a card on which the identifier of the driver is recorded using the driver ID reader 11 may be used as the login input 71.
- the driving data collection device 10 performs writing (S12) and transmission processing (S13) of the measured RRI data up to step S11 where the logout input 72 from the driver is received.
- the measured RRI data is divided in advance for each period B independently of the measurement situation, and the RRI data is written for each period B.
- An example of sequentially transmitting to the biometric data evaluation server 1 is shown.
- the data collection unit 21 receives the RRI data and stores it in the period B unit RRI data 41 .
- the RRI data in units of period A is measured and transmitted to the biological data evaluation server 1 in units of period A.
- the data collection unit 21 of the data evaluation server 1 may divide the RRI data of the period A unit into the period B units and store them in the period B unit RRI data 41 .
- the driving data collecting device 10 does not necessarily transmit the RRI data divided into the period B units for each period B, but monitors the measurement status and collects the RRI data only when the lack of the RRI data corresponds to a certain condition. It may be cut out in units of period B and transmitted to the biometric data evaluation server 1 .
- the driving data collection device 10 executes the process of writing the RRI data in units of period B in a data transmission format (S12). After the completion of the period B unit writing process (S12), the driving data collection device 10 transmits the written period B unit RRI data 41 to the biological data evaluation server 1, and obtains the measurement information related to the period B unit RRI data 41. Stored in the history data 44 of the biometric data evaluation server 1 .
- the driving data collection device 10 executes the above process for each period B until it receives the logout input 72 from the driver (S11).
- an event timer may be provided in the driving data collection device 10 to count the elapse of two minutes, which is an example of the period B, and execute step S11.
- an event in which the driver ID reading device 11 is used for the first time to read the card on which the driver's identifier is recorded may be used as the logout input 72 after the login input 71 is performed.
- the history data 44 stores information related to each data of the period B unit RRI data 41 measured through the driving data collection device 10 .
- a driver ID and a vehicle ID for linking the period B unit RRI data 41 with the driver and the vehicle 7 are stored.
- the time when the data was transmitted to the biological data evaluation server 1 and the transmitted file name may be stored.
- the state of the driver during period B may be stored. The driver state is information indicating under what measurement conditions the period B unit RRI data 41 was measured.
- the driver status does not need to be explicitly stored in the history data 44.
- the period B unit writing process S12 by the driving data collection device 10 is performed only while the vehicle is running, and the RRI data measured when the vehicle is not running is discarded, biometric data Since it is self-evident that all the RRI data for the period B unit transmitted to the evaluation server 1 were measured while the vehicle was running, it is not necessary to store the driver's state.
- the writing process for the period B unit occurs every two minutes regardless of the measurement situation. 41 is generated, that is, the RRI data is lost. Further, for example, in the data transmission process (S13), the transmission process may fail due to a communication failure or the like, and the period B unit RRI data 41 may be lost.
- the loss of RRI data in the period A unit is caused by the absence of the period B unit RRI data 41, or by the absence of effective RRI data even though data exists. Occurs when
- step S14 of the RRI data measurement end processing Upon receiving the logout input 72, the driving data collection device 10 transitions to step S14 of the RRI data measurement end processing.
- business status data 42 including business information related to the period A of the driver is generated and transmitted to the biometric data evaluation server 1 .
- the driving data collection device 10 when the driving data collection device 10 starts measuring the RRI data, it transmits the RRI data in units of period B to the biometric data evaluation server 1, and the biometric data evaluation server 1 stores the RRI data 41 in units of period B. is accumulated for each driver.
- FIG. 3 is a flowchart showing an example of learning processing of the arrhythmia-like abnormal value chronic occurrence determination model 52 performed by the biological data evaluation server 1 .
- the chronic onset discrimination model learning unit 22 refers to the chronic onset correct data 46 that defines whether or not an arrhythmia-like abnormal value is chronically occurring in the RRI data measured in a certain period A, The presence or absence of chronic occurrence correct data 46 to be learned is determined (S21).
- the chronic-onset discrimination model learning unit 22 proceeds to step S22 and performs the following processing for each period A stored in the chronic-onset correct data 46.
- the chronic occurrence correct data 46 to be learned can be set in advance from the input/output device 5 .
- the chronic occurrence discrimination model learning unit 22 determines whether an arrhythmia-like abnormal value is chronically occurring in the period B unit RRI data 41 measured in the period A.
- Aggregate LP feature quantity data 47 used for evaluating whether or not is generated.
- the aggregated LP feature data 47 is input to the chronic occurrence determination model 52 to determine whether or not the arrhythmia-like abnormal value (abnormal value of the arrhythmia feature value) is chronic, as described later. data.
- the chronic onset discrimination model learning unit 22 calls the period B unit LP generation unit 23, and the period B unit LP generation unit 23 divides the period B unit RRI data 41 into the period B unit for the RRI data within the period A. is read (S23).
- period B unit LP generation unit 23 when reading the period B unit RRI data 41, assumes that the effective RRI data in the read period B unit RRI data 41 is extremely small or is empty.
- the period B unit RRI data 41 may be inspected to determine the presence or absence of data loss. In this case, subsequent processing is not performed for the period B unit RRI data 41 determined to correspond to data loss, and the period B is treated as data loss.
- the period B unit LP generator 23 may also read the history data 44 when reading the period B unit RRI data 41 . In this case, referring to the driver state stored in the history data 44, if the driver state does not match the predetermined driver state, the reading of the period B unit RRI data 41 may be stopped and subsequent processing relating to the data may be skipped. good.
- the corresponding period B unit RRI data 41 may not be read and excluded from the learning process of the chronic onset discrimination model.
- the measurement conditions of the period B unit RRI data 41 used for learning of the chronic onset discrimination model 52 are unified, and an improvement in the evaluation accuracy of the arrhythmia-like abnormal value under the measurement conditions is expected.
- period B unit LP generator 23 uses the read period B unit RRI data 41 to generate the period B unit Lorenz plot and stores it as period B unit LP data 43 (S24).
- an LP plotted with T[t] on the x-axis and T[t+dt] on the y-axis is shown as an example.
- the characteristics of the time-series data T[t] can be analyzed by analyzing the geometric figures drawn on the LP.
- dt is greater than 1, that is, it is not limited to two points temporally adjacent to each other.
- the RRI data to be used is not limited to the original sequence RRI[t], and for example, a difference sequence ⁇ RRI[t] obtained by taking a difference between adjacent data points may be used.
- an LP matrix that expresses the frequency of appearance of data points on the LP as density (or brightness) is calculated.
- FIG. 4A which will be described later, when the frequency of appearance is high, the density is high, and when the frequency of appearance is low, the density is low.
- the range of the x-axis and the y-axis is set to 0 ms to 2560 ms, and a grid is defined so as to divide this range into 64 in each axis direction.
- each mesh indicates a region with a mesh width of 40 ms and an area of 40 ms ⁇ 40 ms. Then, the number of data points on the LP belonging to each mesh is counted, and the density of each mesh region is defined as the component of the LP matrix.
- the drawing range of the LP and the division amount of each axis, that is, the mesh width are not limited to this embodiment.
- the mesh area may be defined such that the drawing range is the same and the mesh width is set in units of 80 ms, that is, each axis is divided into 32 parts.
- a threshold is set for the density of the mesh area, and the density exceeding the threshold is set to the threshold.
- the threshold is set to 3 times
- the maximum value of the density will be 3 times even if the data appears 4 times or more in the mesh area.
- the period B unit LP generation unit 23 generates the period B unit LP data 43 as a 64 ⁇ 64 LP matrix with densities of four gradations from 0 to 3.
- the pretreatment for LP generation is not limited to saturation treatment.
- a known time domain analysis method is used for an RRI having a length of about n times as long as the assumed RRI due to n consecutive R wave detection errors.
- Abnormal value exclusion or replacement processing using a certain median filter, interpolation processing of detected miss beats using the Kalman filter, which is also a known technique, and the like may be performed together.
- the chronic onset discrimination model learning unit 22 generates the process of calculating the period B unit LP data 43 from the period B unit RRI data 41 divided by the period B unit for all the periods B included in the period A (S22 ). Note that even if the period B is included in the period A, the period B that does not exist as the period B unit RRI data 41 is treated as missing data.
- the chronic-onset discrimination model learning unit 22 aggregates the obtained period B unit LP data 43 in period A units to generate period A unit aggregate LP data 45, period A unit aggregate LP generation processing (S25). I do.
- the LP generated from the RRI data for this period B has been studied in the past, and there is a great deal of knowledge about what geometries are arrhythmia-like abnormalities.
- period A which has a window width equal to or greater than period B
- RRI data there are times when RRI cannot be continuously measured, times when measurement fails and RRI data does not exist, and when RRI data exists but is not transmitted to the server.
- RRI data actually measured during period A due to various factors, such as the time period when data was lost due to failure in Quantities vary.
- period A is simply divided by period B, there are 200 sections, but there may be only 190 sections of period B within period A that is actually measured.
- the generation of LP in the case where there is a missing RRI data in the section within the period A was not considered in the conventional example, and the geometrical features shown in this case were not clear.
- the chronic occurrence discrimination model learning unit 22 calls the period A unit aggregate LP processing unit 24, and the period A unit aggregate LP processing unit 24 exists within the period A to be evaluated. That is, the group of period B unit LP data 43 for period B that is not a data missing period is aggregated (S25), and the period A unit aggregated LP data 45 is an LP that quantifies the feature amount of the RRI data for period A. to generate
- the period A unit aggregated LP processing unit 24 generates the period A unit aggregated LP data 45 by, for example, averaging each component of the LP matrix for the period B unit LP data 43 group. In this embodiment, since the period A is one business day, the period A unit aggregate LP processing unit 24 generates the business day unit average LP and stores it as the period A unit aggregate LP data 45 .
- the aggregation processing method is not limited to averaging processing for each component of the LP matrix.
- aggregation processing may be performed by statistical processing such as calculating the N% quantile in period A for each component of the LP matrix.
- the aggregation processing is performed by the median value for each component.
- Aggregation processing can be performed in which the influence of data with different characteristics is reduced compared to averaging processing.
- the enhancement processing of the RRI abnormal value based on the concentration threshold does not perform the desired function. In some cases, geometric features that should appear above are no longer observed.
- the density is calculated as a continuous value from 0 to 3. is generated.
- the chronic occurrence discrimination model learning unit 22 calls the period A unit aggregate LP processing unit 24, and the period A unit aggregate LP processing unit 24, from the period A unit aggregate LP data 45, arrhythmia-like abnormality and measurement failure-like abnormality Aggregated LP feature quantity generation processing (S26) is performed using a feature extraction model 50, which is a model for extracting aggregated LP feature quantity data 47 representing the degree.
- the feature extraction model 50 is composed of analysis means for extracting features of a two-dimensional matrix. For example, a two-dimensional matrix of the period A unit aggregated LP data 45 is captured as an image, and is created based on prior knowledge of an arrhythmia-like abnormality or a measurement failure-like abnormality.
- a feature extraction method based on mask patterns that extracts data density and appearance frequency as feature quantities for masked regions, and a convolutional neural network base that learns information on arrhythmia-like abnormalities and measurement failure-like abnormalities as correct data based on the prior knowledge. Models, frequency domain analysis models based on Fourier transforms, etc. can be used. In this embodiment, the feature extraction method using mask patterns will be described in detail with reference to FIGS. 4A to 4F.
- the period B unit LP generation unit 23 reads the period B unit RRI data 41 in step S23, the history data 44 is also read, and only the data that matches the predetermined driver state is used.
- a state-appropriate feature extraction model 50 may be utilized. For example, since the measurement characteristics of the RRI data change due to the difference in driver behavior between when the driver is driving and when the vehicle is stopped, appropriate feature extraction can be expected by using the feature extraction model 50 properly.
- the chronic-onset discrimination model learning unit 22 reads the chronic-onset correct data 46 corresponding to the period A (S27).
- the chronic occurrence correct data 46 stores whether or not the arrhythmia-like abnormal value is chronically occurring in a certain period A using 0 and 1, for example. The above processing is performed for the 46 groups of chronic occurrence correct data to be learned.
- the chronic onset discrimination model learning unit 22 After the chronic onset correct data 46 and the corresponding aggregated LP feature amount data 47 are generated, the chronic onset discrimination model learning unit 22 generates a chronic onset discrimination model 52 from the aggregated LP feature amount data 47. A model learning process (S28) is performed.
- the chronic onset discrimination model 52 can be configured using a known discrimination algorithm. For example, for machine learning algorithms among discrimination algorithms, logistic regression models, decision trees, Random Forest, Support Vector Machine, neural networks, deep learning models, etc. can be used.
- the chronic occurrence discrimination model 52 can discriminate the presence or absence of chronic occurrence of arrhythmia from the aggregated LP feature amount data 47, and the discrimination probability of the presence or absence of chronic occurrence is calculated as the degree of anomaly indicating the degree of arrhythmia-like RRI abnormality in the period A. It can also be used as a calculation model.
- a plurality of chronic occurrence determination models 52 may be generated for each driver state stored in the history data 44 .
- the degree of occurrence of arrhythmia-like abnormal values can be evaluated based on the characteristics of the RRI data according to the measurement conditions.
- This learning process is executed at least once before the process of evaluating the degree of chronic occurrence of arrhythmia-like abnormal values shown in FIG. 5, which will be described later.
- this processing can be performed at regular intervals as the chronic-onset correct data 46 increases, and the chronic-onset discrimination model 52 can be re-learned. As described above, the chronic-onset discrimination model 52 with even higher discrimination accuracy can be generated.
- FIG. 4A to 4F are diagrams showing examples of mask region definitions in the feature extraction method using mask patterns used as the feature extraction model 50 used in the biometric data evaluation server 1.
- FIG. 4A to 4F are diagrams showing examples of mask region definitions in the feature extraction method using mask patterns used as the feature extraction model 50 used in the biometric data evaluation server 1.
- an example of a feature extraction method using a mask pattern composed of a total of five mask patterns shown in FIGS. 4B to 4F is shown, using the period A unit aggregated LP data 45 shown in FIG. 4A as an example.
- the period A unit aggregation LP processing unit 24 extracts features such as statistics such as the average or sum of densities inside and outside a defined mask region, and values such as the number of mesh regions whose densities are equal to or greater than a certain threshold. It is extracted as a quantity (S26 in FIG. 3).
- RRI[t] on the horizontal axis indicates the RRI at a given time
- RRI[t+1] on the vertical axis indicates the RRI from the given time to the next heartbeat.
- FIG. 4B is an example of a mask area (measurement failure-like mask Mask 1) that defines the range in which measurement failure-like abnormal values are distributed when measurement is unstable and is frequently observed due to factors other than beat detection failure. Since measurement failure due to unstable measurement occurs due to inability to capture the peak of the R wave, both R waves and noise are detected as R waves at intervals shorter than the average RRI that is originally measured. end up
- the mask area (Mask1) is defined by the following equation (1).
- FIG. 4C is an example of a mask area (Mask2) that defines the distribution range of RRI data in which the RRI is considered to fluctuate within the normal range.
- the mask area (Mask2) is defined by the following equation (2).
- the RRI data group of LP1 shown in FIG. 4A is included in the region of the normal range mask (Mask2).
- a mask region for separating an arrhythmia-like abnormal value and a measurement failure-like abnormal value due to omission of beat detection into a feature quantity is shown.
- arrhythmia-like abnormal values an example in which premature ventricular contraction (PVC) and premature atrial contraction (PAC) occur frequently will be referred to.
- FIG. 4D is an example of a mask region (Mask3) that defines the range in which arrhythmia-like abnormal values are distributed.
- the mask area is defined as follows.
- the RRI data groups of LP2-1 and LP2-2 shown in FIG. 4A are included in the area of the PVC-like outlier mask (Mask3).
- the RRI data groups of LP3-1 and LP3-2 shown in FIG. 4A are included in the region of the 1-beat detection omission mask (Mask4).
- the RRI data groups of LP4-1 and LP4-2 shown in FIG. 4A are included in the region of the continuous 2-beat detection omission mask (Mask5).
- the mask region is not limited to this embodiment.
- a mask region may additionally be defined and utilized to represent atrial fibrillation.
- FIG. 5 is a flow chart showing an example of processing for evaluating the degree of chronic occurrence of arrhythmia-like abnormal values performed by the biometric data evaluation server 1 . Note that steps S22 to S26 in the figure are the same processing as in the flow chart of FIG.
- the period B unit LP generator 23 reads the period B unit RRI data 41 obtained by dividing the period A RRI data by the period B unit (S23). Subsequently, the period B unit LP generator 23 uses the read period B unit RRI data 41 to generate the period B unit LP and stores it as the period B unit LP data 43 (S24).
- the period A unit aggregation LP processing unit 24 After the period B unit LP generation unit 23 generates the period B unit LP data 43 for all the periods B included in the period A (S22), the period A unit aggregation LP processing unit 24 generates the period B unit LP data 43 is aggregated in units of period A to generate aggregated LP data 45 in units of period A (S25).
- the period A unit aggregated LP processing unit 24 generates a feature extraction model 50 that is a model for extracting an aggregated LP feature quantity representing the degree of arrhythmia-like abnormality or measurement failure-like abnormality from the period A unit aggregated LP data 45. and perform the aggregated LP feature quantity generation process (S26).
- the abnormal value chronic occurrence evaluation unit 25 performs an abnormality degree calculation process (S31) for calculating the abnormality degree data 49 using the learned chronic occurrence discrimination model 52 for the obtained aggregated LP feature amount data 47. conduct.
- the degree of abnormality calculated in this process is obtained by inputting the aggregated LP feature amount data 47 to the chronic onset discrimination model 52, for example, as described above, and the probability of determining the presence or absence of chronic occurrence is the chronic arrhythmia-like abnormality 147 (Fig. 10E) and used as the abnormality degree data 49.
- the abnormal value chronic occurrence evaluation unit 25 performs chronic occurrence determination processing (S32) using the aggregated LP feature amount data 47, the abnormality degree data 49, and the chronic occurrence determination model 52. is chronically occurring, and stored in the period A unit abnormality determination data 51 .
- the abnormal value chronic occurrence evaluation unit 25 acquires the chronic arrhythmia-like abnormality degree 147 from the abnormality degree data 49 of the target driver, and if the chronic arrhythmia-like abnormality degree 147 exceeds a preset abnormality determination threshold value , the chronic arrhythmia-like abnormality determination result 157 (see FIG. 10F) is set to "1" in the period A unit abnormality determination data 51, and is set to "0" if it is equal to or less than the abnormality determination threshold.
- the history data 44 is additionally read in the period B data reading process (S23), and the period A unit aggregate LP is obtained only from the period B unit RRI data 41 measured in a predetermined driver state.
- the feature extraction model 50 used in the aggregated LP feature amount generation process (S26) and the chronic occurrence discrimination model 52 used in the abnormality degree calculation process S3 and the chronic occurrence discrimination process (S32) are for a predetermined driver state. model can be selected and used. In this case, it is possible to more accurately evaluate the occurrence of an arrhythmia-like abnormal value based on the RRI data measurement situation.
- the determination probability and determination result obtained by inputting the aggregated LP feature amount data 47 into the chronic occurrence determination model 52 may be stored as the abnormality degree data 49 and the period A unit abnormality determination data 51, respectively.
- period A unit abnormality discrimination data 51 it is possible to exclude inappropriate periods for period A even in an environment with data loss, and the effect of improving the reliability of autonomic nerve function evaluation. is obtained.
- FIG. 6 is a flow chart showing an example of processing for determining a driver unsuitable for autonomic nerve function evaluation performed by the biometric data evaluation server 1 .
- the period A is one working day, but the state of occurrence of the arrhythmia-like abnormal value of the driver may change according to the exercise load and health condition.
- the unsuitable driver determination unit 26 first determines, for the most recent period C, the period A unit abnormality determination data 51 existing within the period C. After reading, the determination result reading process (S41) is performed for the period C.
- the unsuitable driver determination unit 26 inputs the period A unit abnormality determination data 51 in the period C and uses the unsuitable driver determination model 54 to determine whether the driver is unsuitable for the autonomic nerve function evaluation based on the RRI data. (S42).
- the ANF evaluation unsuitability determination 167 of the unsuitability determination data 48 is set to "1" if unsuitable for autonomic nerve function evaluation, and set to "0" if usable for autonomic nerve function evaluation.
- period A is one business day
- period C may be five business days.
- a statistical model that statistically sets a determination threshold by comparing with the appearance distribution of the ANF evaluation unsuitable determination 167 may be used.
- the unsuitable driver determination unit 26 can detect a driver who is constantly considered unsuitable for the autonomic nerve function evaluation using the RRI data, regardless of changes in the driver's condition such as health condition and exercise load. An effect of improving the reliability of function evaluation can be obtained.
- FIG. 7 is a flowchart showing an example of processing for calculating an autonomic nerve function index performed by the biological data evaluation server 1.
- FIG. 7 an example of calculating an autonomic nerve function index using RRI data having a window width similar to that of period B used for LP generation will be described. That is, in this embodiment, an example of calculating an autonomic nerve function index in units of an analysis window of 2 minutes will be shown.
- the autonomic nerve function index calculator 27 performs a data reading process of reading the period B unit RRI data 41 (S51).
- the autonomic nerve function index calculation unit 27 determines the appropriateness of the analysis from the unsuitability determination data 48 ( S52).
- the autonomic nerve function index calculation unit 27 extracts records in which the ANF evaluation unsuitable determination 167 of the unsuitable determination data 48 is "0", and the period from the analysis start 165 to the analysis end 166 is stepped with the B unit RRI data 41. Autonomic nerve function index data 53 in S53 to S55 are calculated. Thereby, the autonomic-nerve function index calculation unit 27 can exclude the records in which the ANF evaluation unsuitability determination 167 is "1".
- the autonomic nerve function index calculation unit 27 calculates the autonomic nerve function index data 53 by performing analysis as necessary from the RRI data. Examples of analysis as required include frequency domain analysis (S53), time domain analysis (S54), and RRI nonlinear domain analysis (S55).
- the autonomic nerve function index calculation unit 27 calculates the frequency domain index from the RRI time series via the power spectral density. Since the RRI time series is unevenly spaced time series data, it is resampled at even intervals by spline interpolation, etc., and then an autoregressive model or maximum entropy method is used, or Lomb-Scargle, which can use unevenly spaced data.
- a power spectral density PSD Power Spectral Density
- PSD Power Spectral Density
- the autonomic nerve function index calculation unit 27 calculates, for example, the integral value LF of the low frequency region of 0.05 Hz-0.15 Hz and the high frequency region of 0.15 Hz-0.40 Hz.
- the integrated value HF, TP that is the sum of LF and HF, LF/HF that is obtained by dividing LF by HF, and LFnu that is a percentage obtained by dividing LF by TP are calculated as frequency domain indexes of the autonomic nerve function index data 53. .
- the autonomic nerve function index calculation unit 27 calculates the time domain index by calculating the statistics of the RRI time series and the ⁇ RRI time series, which is the difference series of adjacent RRIs.
- the autonomic nerve function index calculation unit 27 calculates the average heart rate, which is the reciprocal of the average value of the RRI time series, and SDNN, which is the standard deviation of the RRI data. Further, from the ⁇ RRI time series, for example, NN50, which is the total number of data in which the absolute value of the difference value constituting the ⁇ RRI time series exceeds 50 ms, pNN50 obtained by dividing NN50 by the total number of data in the ⁇ RRI time series, and the standard deviation of ⁇ RRI SDSD is calculated as a time domain index of the autonomic nerve function index data 53 .
- the autonomic nerve function index calculation unit 27 calculates nonlinear feature amounts by various methods.
- the autonomic nerve function index calculator 27 calculates an elliptical area S by elliptical approximation of the area plotted as shown in FIG. 4A, for example, through LP analysis.
- the autonomic nerve function index calculator 27 calculates ⁇ 1 and ⁇ 2 by similar entropy and detrended fluctuation analysis, tone and entropy based on tone-entropy analysis, and calculates them as RRI nonlinear region indices of the autonomic nerve function index data 53.
- the autonomic nerve function index calculation unit 27 stores the calculated autonomic nerve function index group collectively in the autonomic nerve function index data 53 (S56).
- autonomic nerve function index calculation unit 27 may execute the processing of the above steps in parallel or sequentially.
- the autonomic-nerve function index calculator 27 stores the unsuitable flag in the autonomic-nerve function index data 53 without calculating the autonomic-nerve function index.
- the incompatibility flag may be stored in the autonomic nerve function index data 53 after calculating the autonomic nerve function index.
- FIG. 8 is a diagram showing an example of an autonomic nerve function evaluation result screen 1000 output by the result display unit 28 to the display of the input/output device 5.
- FIG. 8 When the biological data evaluation server 1 finishes measuring the RRI data for the period A and generates the period A unit abnormality determination data 51, the input/output device 5 displays the autonomic nerve function evaluation result screen 1000 for the period A unit. indicate.
- the driver ID 1011, the vehicle (ID) 1012 used, and the arrhythmia-like RRI abnormality level 1013 are displayed as summary information 1010 at the top. Further, at the lower part, transitions of the autonomic nerve function index are displayed in five steps in time series every 30 minutes as transition 1020 of the autonomic nerve function index.
- the autonomic nerve function index is an example showing the range from relaxation to stress.
- the degree of arrhythmia-like abnormal values can be evaluated even if there is a missing section in period A. Therefore, the upper summary information 1010 includes today's business hours (eg, eight and a half hours) and , the degree of the arrhythmia-like abnormal value during the working hours is displayed as an arrhythmia-like RRI abnormality level 1013 .
- the evaluation of the occurrence of the arrhythmia-like abnormal value is quantitatively displayed. good.
- the driver can ensure that the transition of the autonomic nerve function index displayed on the autonomic nerve function evaluation result screen 1000 is The effect of being able to quantitatively understand whether or not the reliability is sufficient can be obtained.
- FIG. 9 is a diagram showing an example of a detailed analysis screen 2000 of the reliability of the RRI data measured during the period A, output by the result display unit 28 to the display of the input/output device 5.
- FIG. 10 When the detail button 1014 displayed in the summary information 1010 of the autonomic nerve function evaluation result screen 1000 shown in FIG.
- a period A unit aggregation LP 2004 is displayed.
- Measured period B unit RRI data 2011 and period B unit LP data 2012 before aggregation to create the period A unit aggregate LP 2004 are displayed in the lower right. Time change of unit LP data can be confirmed.
- a low-reliability section 2014 in units of period B is shaded.
- a degree of arrhythmia-like RRI abnormality 2021 and degrees of measurement failure-like RRI abnormality 2022 and 2023 are displayed together with their grounds.
- the LP for each of the arrhythmia-like, measurement failure-like, and R-wave detection failure-like RRI abnormal values defined in the mask pattern-based feature extraction method used for the feature extraction model 50, the LP The above region definition is superimposed on the period A unit aggregation LP 2004 .
- Each abnormality level of the period A unit aggregation LP 2004 can be displayed as a percentage, for example, by normalizing the feature amount related to each mask by the maximum value of the feature amount.
- the grounds for determination may be visualized by displaying an attention map, which is means for visualizing the grounds for the determination.
- the user of the biological data evaluation server 1 can easily understand why the reliability of the autonomic nerve function evaluation due to the arrhythmia-like abnormal value is declining. can be presented.
- the driver can judge whether the RRI data is being measured normally. Furthermore, when it is found that the measurement condition is bad only in a part of the period B, the driver can consider the cause of the measurement failure by comparing it with his/her business knowledge.
- FIG. 10A is a diagram showing an example of the data structure of the period B unit RRI data 41 held in the storage device 4 in the biological data evaluation server 1.
- FIG. 10A is a diagram showing an example of the data structure of the period B unit RRI data 41 held in the storage device 4 in the biological data evaluation server 1.
- the period B unit RRI data 41 typically stores the driver ID 101, the vehicle ID 102, the RRI data measurement time 103, the RRI 104, and the acceleration norm 105 in one record.
- the driver ID 101 stores the identifier of the driver acquired by the driver ID reader 11 of the vehicle 7.
- the vehicle ID 102 stores the identifier of the vehicle 7 preset in the driving data collection device 10 .
- the measurement time 103 stores the date and time when the heartbeat sensor 14 measured the RRI data.
- RRI 104 stores the value (msec) detected by heart rate sensor 14 .
- the acceleration norm 105 stores the acceleration vector value detected by the acceleration sensor 15 .
- FIG. 10B is a diagram showing an example of the data structure of the period B unit LP data 43 held in the storage device 4 in the biological data evaluation server 1.
- the period B unit LP data 43 typically includes a driver ID 111, a vehicle ID 112, a period B window width 113, an analysis period 114, a period B unit LP 115, and an analysis source file name 116 in one record. Stored.
- the driver ID 111 is the same as in FIG. 10A and stores the identifier of the driver.
- the vehicle ID 112 stores the identifier of the vehicle 7 as in FIG. 10A.
- the period B 113 stores the length of the period B.
- the analysis period 114 stores the date and time when the period B starts.
- An array obtained by flattening the two-dimensional LP matrix elements in the period B may be stored in the period B unit LP115.
- information obtained by binary-encoding an LP image obtained by imaging the LP matrix component as luminance may be stored.
- each component of the LP matrix component may be stored in a different column (or field) direction.
- the analysis source file name 118 stores the file name (or path) of the period B unit LP data 43 .
- FIG. 10C is a diagram showing an example of the data structure of the period A unit aggregated LP data 45 held in the storage device 4 in the biological data evaluation server 1.
- the period A unit aggregate LP data 45 typically includes a driver ID 121, a vehicle ID 122, a window width 123 of period A, a window width 124 of period B, and a period B in which data actually existed within period A.
- a number of windows 125 representing the number of sections of , an analysis period 126, a period A unit aggregation LP 127, and a method indicating an aggregation method 128 are stored in one record.
- the driver ID 121 and vehicle ID 122 store the identifier of the driver and the identifier of the vehicle 7 in the same manner as in FIG. 10A.
- the period A123 stores the length of the period A.
- the period B 124 stores the length of the period B.
- the analysis period 126 stores the date and time when the period A starts.
- the period A unit aggregation LP 127 may store an array obtained by flattening the two-dimensional LP matrix elements in period A. FIG. Alternatively, information obtained by binary-encoding an LP image obtained by imaging the LP matrix component as luminance may be stored. Alternatively, each LP matrix component may be stored in a different column direction.
- FIG. 10D is a diagram showing an example of the data structure of the period A unit aggregated LP feature data 47 held in the storage device 4 in the biometric data evaluation server 1.
- the period A unit aggregated LP feature data 47 typically includes a driver ID 131, a vehicle ID 132, a window width 133 of period A, a window width 134 of period B, and a period B that actually existed within period A.
- column name For example, the arrhythmia-like feature quantity 1137-1 to the measurement failure-like feature quantity N137-N) are stored in one record.
- the driver ID 131 to the analysis period 136 are the same as the driver I 121D to the analysis period 126 in FIG. 10C.
- the arrhythmia feature quantities 1 to N are, for example, the LP feature quantity included in the measurement failure-like mask Mask1 in FIG. 4B, the LP feature quantity included in the PVC-like abnormal value mask Mask3, , the feature amount of the LP included in the 1-beat detection omission mask Mask 4 and the continuous 2-beat detection omission mask Mask 5 are stored.
- FIG. 10E is a diagram showing an example of the data structure of the abnormality degree data 49 held in the storage device 4 in the biological data evaluation server 1.
- the abnormality degree data 49 typically includes a driver ID 141, a vehicle ID 142, a window width 143 of period A, a window width 144 of period B, and a window representing the number of sections of period B that actually existed within period A.
- a number 145, an analysis period 146, and a chronic arrhythmia-like abnormality degree 147 are stored in one record.
- the driver ID 141 to the analysis period 146 are the same as the driver I 131D to the analysis period 136 in FIG. 10D.
- the chronic arrhythmia-like abnormality degree 147 for example, the discrimination probability of the chronic occurrence of arrhythmia obtained by inputting the aggregated LP feature amount data 47 into the chronic occurrence discrimination model 52 is stored in one record.
- the period A unit abnormality determination data 51 typically includes a driver ID 151, a vehicle ID 152, a window width 153 of period A, a window width 154 of period B, and a section of period B that actually existed during period A.
- a window number 155 representing a number, an analysis period 156, and a chronic arrhythmia-like abnormality determination result 157 are stored in one record.
- the driver ID 151 to the analysis period 156 are the same as the driver I 141D to the analysis period 146 in FIG. 10E.
- the abnormal value chronic occurrence evaluation unit 25 determines whether the arrhythmia-like abnormal value is chronically occurring in the period A, and the result is a value of “0” or “1”. stored in . If the arrhythmia-like abnormality value is chronically occurring, "1" is stored in the chronic arrhythmia-like abnormality determination result 157, and otherwise "0" is stored.
- FIG. 10G is a diagram showing an example of the data structure of the unsuitability determination data 48 held in the storage device 4 in the biometric data evaluation server 1.
- the unsuitability determination data 48 typically includes a driver ID 161, a vehicle ID 162, a window width 163 in period A, a window width 164 in period C, an analysis start 165 in period A when analysis was started, and an analysis end 165.
- An analysis end 166 of period A and an autonomic nerve function (ANF in the figure) evaluation unsuitability determination 167 evaluated for period C are stored in one record.
- the driver ID161 to period A163 are the same as the driver ID151 to period A153 in FIG. 10F.
- the period C164 stores a period (window width) larger than the period A.
- the analysis start 165 and analysis end 166 store the start date and time of the period C and the end date and time.
- Autonomic nerve function (ANF) evaluation unsuitability determination 167 stores "1" if the RRI data of the driver ID 161 is not suitable for evaluating autonomic nerve function, otherwise "0" is stored. be.
- FIG. 10H is a diagram showing an example of the data structure of the autonomic nerve function index data 53 held in the storage device 4 in the biological data evaluation server 1.
- FIG. 10H In addition to the driver ID 171, the vehicle ID 172, and the date and time 173, the autonomic nerve function index data 53 typically stores various autonomic nerve function indices calculated by the autonomic nerve function index calculator 27 in one record.
- the driver ID 171 and vehicle ID 172 are the same as the driver ID 161 and vehicle ID 162 in FIG. 10J.
- the date and time 173 stores the date and time when the autonomic nerve function index was calculated.
- the autonomic function index is, for example, a frequency domain index LF/HF 174, a time domain index average heart rate 175, NN50 (176), or an RRI nonlinear region index ⁇ 1 (177). is mentioned.
- FIG. 10I is a diagram showing an example of the data structure of the business status data 42 held in the storage device 4 in the biometric data evaluation server 1.
- the business status data 42 stores a driver ID 181, a vehicle ID 182, a business day 183, a measurement start date and time 184, a measurement end date and time 185, a traveled distance 186, and the like in one record.
- the driver ID 181 and vehicle ID 182 are the same as the driver ID 171 and vehicle ID 172 in FIG. 10H.
- the date of work 183 stores the date when the driver performed the work.
- the measurement start date and time 184 and the measurement end date and time 185 store the date and time when the measurement of biological data was started and the date and time when the measurement was finished.
- the distance traveled 186 stores the distance the driver drove during the working day 183 .
- FIG. 10J is a diagram showing an example of the data structure of the history data 44 held in the storage device 4 in the biometric data evaluation server 1.
- the history data 44 typically includes a driver ID 191, a vehicle ID 192, a period B window width 193, a period B unit RRI data transmission time 194, a period B unit RRI data reception file name 195, and a representative state. 196 etc. are stored in one record.
- the driver ID 191 and vehicle ID 192 are the same as the driver ID 181 and vehicle ID 182 in FIG. 10I.
- the period B 193 stores the length of the period B.
- the file transmission time 194 stores the date and time when the driving data collection device 10 of the vehicle 7 transmitted the RRI data.
- File name 195 stores the file name (or path) of the RRI data.
- the representative state 196 stores the state of the driver during the period B in which the period B unit RRI data 41 was measured.
- FIG. 10K is a diagram showing an example of the data structure of the chronic occurrence correct data 46 held in the storage device 4 in the biological data evaluation server 1.
- the chronic occurrence correct data 46 typically includes a driver ID 201, a vehicle ID 202, a window width 203 of the period A, an analysis period 204, and a chronic arrhythmia-like abnormality label indicating whether or not the chronic arrhythmia-like abnormality is a correct label. 205 are stored in one record.
- the driver ID 201 and vehicle ID 202 are the same as the driver ID 191 and vehicle ID 192 in FIG. 10J.
- the period A 203 stores the length of the period A.
- the analysis period 204 stores the start date and time of the period A.
- the chronic arrhythmia-like abnormality label 205 stores "1" if it is a chronic arrhythmia-like abnormality, and "0" otherwise.
- FIG. 11 is a diagram showing an example related to the definition of period A and period B and the occurrence of data loss.
- the period A501 is set to 1 business day (that is, the period from the start to the end of work on a certain date) and the period B503 is set to 2 minutes
- the period A501 (501-1, 501 -2) and an example in which the period A501 is divided into periods B503 will be described.
- period A501 and period B503 when no data loss occurs will be described.
- period B unit RRI data 504 divided by the period B 503 exists for the number of windows n for the period defined by the period A (driver A) 501-1.
- the work start/end times are different from those of the driver A 502-1 on 12/1.
- the period B unit RRI data 504 is measured over the period of the driver B) 501-2.
- the period B unit RRI data 504 with the window number m which is different from the window number n that existed in the period A501-1 of the driver A on 12/1, becomes the analysis target.
- period A501 is defined as period A (driver A) 501-1 in driver A502-2 as well as driver A502-1.
- the period B unit RRI data 504 measured during period A (driver A) 501-1 has data loss due to various factors.
- the RRI data 505-1 for a certain period B unit has a significantly small amount of effective RRI data due to measurement failure, and is determined to have data loss.
- period B unit RRI data 505-3 does not exist on the biological data evaluation server 1 due to an external factor 73 such as communication failure, and it is determined that there is data loss.
- the period B unit RRI data 41 measured during the period A are n, n ⁇ 3, and m, respectively. Cases occur where the RRI data amount is different.
- the target period B unit RRI data 41 is examined to determine whether or not there is data loss. You may
- the biological data evaluation server 1 aggregates in period A units using the period B unit LP data 43 group calculated from the period B unit RRI data 41 group. calculating aggregated LP data 45 per period A, calculating aggregated LP feature amount data 47 including a feature amount representing the degree of arrhythmia-like abnormal value occurrence from the aggregated LP data 45 per period A using a feature extraction model 50; By inputting the aggregated LP feature amount data 47 to the chronic occurrence discrimination model 52, it is discriminated whether or not the arrhythmia-like abnormal value is chronically occurring in period A units.
- the biological data evaluation server 1 of the present embodiment can generate an aggregated LP that uses the degree of arrhythmia RRI abnormal value for evaluation even for RRI data with data loss.
- the biological data evaluation server 1 generates aggregated LP feature amount data 47 including a feature amount representing the degree of occurrence of arrhythmia-like abnormal values from the aggregated LP, thereby reducing RRI caused by factors other than arrhythmia, such as measurement failure-like abnormal values. It becomes possible to quantitatively evaluate the degree of occurrence of arrhythmia-like abnormal values separately from abnormal values.
- the biological data evaluation server 1 uses the aggregated LP feature amount data 47 to determine whether the occurrence of arrhythmia-like abnormal values is chronic in the period A for the RRI data in the period A with data loss. can be determined.
- the biometric data evaluation server 1 measures the RRI data obtained from the RRI data measured during the period A even when measuring the RRI data in a state that is not necessarily a resting state, such as during work involving driving the vehicle 7. It is possible to obtain the effect of being able to evaluate the reliability of the index.
- the present invention is not limited to this.
- the present invention can be applied to a moving object requiring a driver or operator, such as a railway vehicle, a ship, or an aircraft.
- LP is generated in real time from the RRI data received from vehicle 7, and LP generation may be terminated in units.
- Example 2 of the present invention will be described. Except for the differences described below, each part of the safe driving support system of the second embodiment has the same function as each part with the same reference numerals of the first embodiment, so description thereof will be omitted.
- FIG. 12 shows Example 2 of the present invention, in which the biometric data evaluation system also predicts the risk of a driver's traffic accident or incident (hereinafter referred to as accident risk) using the autonomic nerve function index data 53. It is a block diagram showing an example of main composition.
- the vehicle 7 collects a biometric sensor 12 for detecting biometric data of the driver, a driver ID reader 11 for identifying the driver, and collects the detected biometric data and the driver ID, and transmits the driving data to the biometric data evaluation server 1.
- a biometric sensor 12 for detecting biometric data of the driver
- a driver ID reader 11 for identifying the driver
- the vehicle 7 further includes an in-vehicle sensor 8 that detects the business state and driving state, and a prediction result notification device 9 that receives a warning according to the driver's accident risk from the biological data evaluation server 1 and presents it to the driver.
- the driving data collection device 10, the prediction result notification device 9, and the driver ID reading device 11 are independent devices, but they can be configured as one portable terminal.
- the mobile terminal functions as a driving data collection unit, a prediction result notification unit, and a driver ID reading unit.
- the in-vehicle sensors 8 include a GNSS (Global Navigation Satellite System) 15 that detects vehicle position information, an acceleration sensor 16 that detects the behavior and speed of the vehicle 7, a camera 17 that detects the driving environment as an image, and a driver. may include a business terminal 18 for presenting or recording business information.
- GNSS Global Navigation Satellite System
- the in-vehicle sensor 8 is not limited to the above, and includes a ranging sensor that detects objects and/or distances around the vehicle 7, a steering angle sensor that detects driving operation, and a turning operation of the vehicle 7.
- An angular velocity sensor or the like can be used.
- the acceleration sensor is preferably a triaxial acceleration sensor.
- the memory 3 of the biological data evaluation server 1 includes a data collection unit 21, a chronic occurrence discrimination model learning unit 22, a period B unit LP generation unit 23, a period A unit aggregation LP processing unit 24, and an abnormal value chronic occurrence evaluation unit.
- a data collection unit 21 a chronic occurrence discrimination model learning unit 22
- a period B unit LP generation unit 23 a period A unit aggregation LP processing unit 24
- an abnormal value chronic occurrence evaluation unit 25
- an unsuitable driver determination unit 26 an autonomic function index calculation unit 27
- a result display unit 28 a danger prediction unit 29, and a warning presentation unit 30 are loaded as programs.
- Each program is executed by processor 2 . Details of each functional unit will be described later.
- the storage device 4 of the biometric data evaluation server 1 stores the data used by the above functional units.
- the storage device 4 stores period B unit RRI data 41, period B unit LP data 43, period A unit aggregated LP data 45, aggregated LP feature amount data 47, anomaly degree data 49, and period A unit anomaly determination data.
- 51 autonomic nerve function index data 53, work status data 42, history data 44, chronic occurrence correct data 46, inappropriate determination data 48, feature extraction model 50, chronic occurrence determination model 52, inappropriate driver
- it further stores work/environment data 55, attribute information data 57, accident risk prediction data 56, and an accident risk prediction model 58.
- FIG. 13A is a flow chart showing an example of the process of predicting the accident risk during work performed by the biometric data evaluation server 1.
- FIG. This process can be executed when biometric data is received from the vehicle 7 .
- the risk prediction unit 29 first inputs the RRI data acquired from the vehicle 7 to the autonomic nerve function index calculation unit 27 to calculate the autonomic nerve function index data 53 (S61).
- the autonomic nerve function index calculation unit 27 calculates the frequency domain analysis and the power spectrum density PSD as described above, and calculates LF/HF, LFnu, etc. as frequency domain indices of the autonomic nerve function index data 53 .
- the danger prediction unit 29 selects and reads the accident risk prediction model 58 suitable for the driver of the vehicle 7 in motion (S62).
- the accident risk prediction model 58 is a well-known or known machine learning model that receives as input the autonomic function index data 53 of the driver of the vehicle 7 in motion, predicts and outputs the accident risk after a predetermined time, and is learned in advance. model.
- the accident risk prediction model 58 a plurality of models are prepared in advance according to the type of accident risk, the driving environment measured by the in-vehicle sensor 8 of the vehicle 7, and the work/environment data 55 storing predetermined work hours. It is good to generate it. By generating a plurality of models, it becomes possible to select and use a suitable model in the accident risk prediction process (S63).
- a plurality of models may be generated according to the attribute information data 57 .
- a suitable accident risk prediction model 58 is selected based on the work/environment data 55 and attribute information data 57 (S62).
- a plurality of accident risk prediction models 58 may be selected instead of selecting only a single model. In this case, it is desirable to give the accident risk prediction data 56 a code that can be identified by the accident risk prediction model 58 used for prediction.
- the risk prediction unit 29 uses the selected accident risk prediction model 58 to input the autonomic nerve function index data 53, predict the accident risk after a predetermined time, and store it in the accident risk prediction data 56.
- the accident risk prediction data 56 can be calculated as the probability of occurrence of incidents or accidents.
- FIG. 13B is a flowchart showing an example of the process of issuing an alert warning of an increase in accident risk, which is performed by the biometric data evaluation server 1.
- FIG. This process is the process performed in step 63 of FIG. 13A.
- the warning presenting unit 30 searches the accident risk prediction data 56 for data with a high accident risk for which an alert is issued (S71).
- the warning presentation unit 30 can determine data in which the probability of occurrence of an incident or accident exceeds a predetermined threshold as data with a high accident risk.
- the warning presenting unit 30 determines whether or not the accident risk prediction data 56 to be issued exists (S72). If the accident risk prediction data 56 to be issued exists, the process proceeds to step S73. , if it does not exist, terminate the process.
- the warning presenting unit 30 then performs evaluation appropriateness determination processing (S73) to determine whether there is any concern about the evaluation of the autonomic nerve function index that was used as the input for the accident risk prediction model 58.
- evaluation appropriateness determination processing S73
- the warning presenting unit 30 determines, for example, based on the degree of anomaly data 49, whether there is a possibility that the RRI data of the driver in question has been significantly affected by the measurement failure. Determination of measurement failure is performed, for example, by generating LP from the RRI data read by the warning presentation unit 30, and using measurement failure-like mask Mask1, 1-beat detection omission mask Mask4, and continuous 2-beat detection omission mask Mask5 in FIG. 4B. If the amount of data (concentration) included exceeds a predetermined threshold, it can be determined that the measurement is faulty.
- the warning presenting unit 30 After that, the warning presenting unit 30 generates warning content for warning of an increase in accident risk according to the situation based on the accident risk prediction data 56, work/environment data 55, and attribute information data 57 of the driver (S74 ) and issues this as an alert to the driver (S75).
- the warning presenting unit 30 acquires the vehicle ID to be transmitted from the driver ID based on the work/environment data 55 and identifies the transmission target. Note that if the transmission target is not the driver but the driver administrator, the warning presenting unit 30 may instead target the input/output device 5 of the biometric data evaluation server 1 as the transmission target.
- the prediction result notification device 9 notifies the driver of the warning. Further, in the biological data evaluation server 1 , the warning presenting unit 30 displays the vehicle 7 to which the warning has been sent on the display of the input/output device 5 .
- FIG. 14 is a diagram showing an example of a warning presentation screen 3000 output by the prediction result notification device 9 of the vehicle 7 and issued to the driver when an increase in accident risk is detected.
- the prediction result notification device 9 has a display (not shown), and displays a warning presentation screen 3000 upon receiving a warning from the biological data evaluation server 1 .
- the warning display screen 3000 includes an area 3001 that displays an accident risk alert, a comment area 3003 that displays countermeasures for eliminating the increased accident risk, and a possibility that the displayed alert content is incorrect. It includes an area 3002 that displays information related to sex.
- a warning message of increased accident risk can be displayed.
- the comment area 3003 of the warning presentation screen 3000 for example, by presenting a specific countermeasure for eliminating the increase in accident risk, the driver who receives the warning is not finished with the warning, Able to understand and act on the next action to be taken in order to resolve a dangerous situation.
- the notification area 3002 of the possibility of false alarm of the warning presentation screen 3000 information for notifying the possibility of false alarm based on the reliability of the autonomic nerve function index measured from the driver who issued the alarm is displayed.
- the driver can determine whether the alert notification is due to low reliability of the measurement data and autonomic nerve function evaluation, or whether the accident risk is truly increasing, and the reliability of the alert notification content. improves.
- the warning may be presented by other methods.
- the warning may be presented in the form of a voice that mechanically reads a text equivalent to the content displayed on the warning presentation screen 3000 .
- the biological data evaluation system of the present embodiment performs the processing described in Embodiment 1, and the autonomic nerve function index data 53 calculated from the RRI data is input to the accident risk prediction model 58, and after a predetermined period of time, Accident risk prediction data 56 is calculated, and when an increase in the risk of occurrence of an accident or incident is detected from the accident risk prediction data 56, the contents of the abnormality degree data 49, the unsuitability determination data 48, and the period A unit abnormality determination data 51 In consideration of this, an alert is issued to the driver to warn of an increase in accident risk.
- the biological data evaluation server 1 of the present embodiment when reporting an increase in accident risk based on the autonomic nerve function index data 53 calculated from the RRI data, issues an erroneous report such as arrhythmia or RRI data measurement failure. It is possible to issue an alert based on the characteristics of the driver who is notified about an event that may be a factor of.
- the present invention is not limited to the above-described embodiments, and includes various modifications.
- the above-described embodiments have been described in detail in order to explain the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the described configurations.
- it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment and it is also possible to add the configuration of another embodiment to the configuration of one embodiment.
- each of the above configurations, functions, and the like may be realized by software by the processor 2 interpreting and executing a program for realizing each function.
- Information such as programs, tables, files, etc. that realize each function is stored in storage devices such as memory 3, hard disk drives, SSD (Solid State Drives), or computer-readable non-temporary storage devices such as IC cards, SD cards, DVDs, etc. It can be stored on a data storage medium.
- the biological data evaluation system of the above embodiment can be configured as follows.
- a biometric data evaluation server that has a processor (2) and a memory (3) and evaluates biometric data, wherein data equivalent to heartbeat intervals (period B unit RRI data) is obtained from the biometric data of a subject (driver) 41)
- a data collection unit (21) that receives data (21) calculates a Lorenz plot (period B unit LP data 43) in a predetermined period from the heartbeat interval equivalent data (41), and calculates an aggregate Lorenz plot (period A unit aggregate LP data 45) and a Lorenz plot generation unit (period B unit LP generation unit 23, period A unit aggregation LP processing unit 24).
- an LP is generated for each period B in which RRI data loss does not occur, and by aggregating the LP for each period A, knowledge about period B can be applied to period A including RRI data loss. becomes.
- the predetermined period includes a first period (period A) and a second period shorter than the first period (period A).
- Period B the Lorenz plot generation unit (23, 24) converts the heartbeat interval equivalent data (41) into the Lorenz plot (Period B unit LP data 43) in units of the second period (Period B) is calculated, and the Lorentz plot (43) calculated in units of the second period (period B) is aggregated in units of the first period (period A), and the first period (period A) , and outputs the aggregated Lorenz plot (45) in the first period (period A).
- an LP is generated for each period B in which no RRI data loss occurs, and each LP is aggregated in units of period A, so that the knowledge about period B that has been researched in the past can be used to eliminate RRI data loss. applicable for period A inclusive.
- the Lorenz plot generator (23, 24) converts the Lorenz plot (43) in units of the second period (period B) to the heartbeat Determining whether or not the interval equivalent data (41) is missing, excluding the part where the heart beat interval equivalent data (41) is missing in units of the second period (period B), and then performing the aggregation process A biological data evaluation server characterized by calculating the aggregated Lorenz plot (45) in the first period (period A).
- the chronic occurrence discrimination model learning unit 22 calls the period A unit aggregate LP processing unit 24, and the period A unit aggregate LP processing unit 24 is the period to be evaluated.
- Aggregation processing (S25) is performed on the period B unit LP data 43 group of period B that exists in A, that is, is not a data missing period, and the period A unit is an LP that quantifies the feature amount of the RRI data of period A.
- Aggregated LP data 45 can be generated.
- the Lorenz plot generation unit (23, 24) performs a plurality of calculations calculated in units of the second period (period B) as the aggregation process A biological data evaluation server, wherein a statistical value is calculated for each matrix component of a Lorenz plot matrix indicating the Lorenz plot (43) in units of the first period (period A).
- the period A unit aggregate LP processing unit 24 performs period A unit aggregate LP data 45 by performing statistical processing such as averaging processing for each component of the LP matrix for the period B unit LP data 43 group.
- statistical processing such as averaging processing for each component of the LP matrix for the period B unit LP data 43 group.
- the Lorenz plot generator (23, 24) replaces the second period (period B) with the first period (period A)
- the period A unit aggregated LP processing unit 24 can generate period A unit aggregated LP data 45 for executing heart rate variability analysis from period B unit LP data 43 of period B that is not a data missing period. .
- the Lorenz plot generation unit (23, 24) receives information on the time width of the loss of the heartbeat interval equivalent data (41), A biological data evaluation server characterized in that the second period (period B) is determined based on the information of the missing time width of the heartbeat interval equivalent data (41).
- the period B unit LP generation unit 23 determines the time width of the period B based on the information on the time width of loss or missing of the RRI data received from the input device or the like, thereby adjusting the biometric data measurement environment. It is possible to generate the corresponding period B unit LP data 43 .
- the abnormal value chronic occurrence evaluation unit 25 calculates the discrimination probability of the presence or absence of chronic occurrence obtained by inputting the aggregated LP feature amount data 47 to the chronic occurrence discrimination model 52 as the chronic arrhythmia-like abnormality degree 147. degree data 49 can be generated.
- the unsuitable driver determination unit 26 can detect a driver who is constantly considered unsuitable for autonomic nerve function evaluation using RRI data regardless of changes in the driver's own condition such as health condition and exercise load. It is possible to obtain the effect of improving the reliability of nerve function evaluation.
- the autonomic nerve function index calculation unit 27 can give an unsuitability flag to a driver who is unsuitable for autonomic nerve function evaluation using RRI data due to steady arrhythmia, and from the obtained autonomic nerve function index The effect of reducing the possibility of misinterpretation and improving the reliability of autonomic nerve function evaluation can be obtained.
- the warning presenting unit can prevent a decrease in the reliability of alert issuance due to frequent false alarms.
- the driver can determine whether the notified alert is due to the low reliability of the measurement data and autonomic nerve function evaluation, or whether the accident risk is truly increasing, and the reliability of the alert content. improve sexuality.
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Cardiology (AREA)
- Pathology (AREA)
- Animal Behavior & Ethology (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Biophysics (AREA)
- Veterinary Medicine (AREA)
- Physics & Mathematics (AREA)
- Physiology (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Psychiatry (AREA)
- Neurology (AREA)
- General Business, Economics & Management (AREA)
- Business, Economics & Management (AREA)
- Signal Processing (AREA)
- Developmental Disabilities (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Child & Adolescent Psychology (AREA)
- Databases & Information Systems (AREA)
- Educational Technology (AREA)
- Hospice & Palliative Care (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Artificial Intelligence (AREA)
- Neurosurgery (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
Description
図1は、本発明の実施例1を示し、生体データ評価システムの主要な構成の一例を示すブロック図である。本実施例の生体データ評価システムは、ネットワーク13を介して1以上の車両7から受信するデータを処理する生体データ評価サーバ1を含む。
図2は、車両の運転データ収集装置10で行われるRRIデータ送信処理の一例を示すフローチャートである。本実施例では、慢性不整脈様異常値発生の程度を評価する対象期間である第1の期間A(窓幅)を、例えば、1業務日、期間A以下の窓幅を有する第2の期間Bを、例えば、2分単位とした場合の例を示す。
次に、生体データ評価システムで使用する各データの特徴的な構造について示す。
以上のように、上記実施例の生体データ評価システムは、以下のような構成とすることができる。
Claims (15)
- プロセッサとメモリを有して、生体データを評価する生体データ評価サーバであって、
対象者の前記生体データから心拍間隔相当データを受け付けるデータ収集部と、
前記心拍間隔相当データから所定の期間でローレンツプロットを算出して、集約ローレンツプロットとして出力するローレンツプロット生成部と、
を有することを特徴とする生体データ評価サーバ。 - 請求項1に記載の生体データ評価サーバであって、
前記所定の期間は、第1の期間と、前記第1の期間よりも短い第2の期間を含み、
前記ローレンツプロット生成部は、
前記心拍間隔相当データを前記第2の期間単位で前記ローレンツプロットを算出し、前記第2の期間単位で算出した前記ローレンツプロットを前記第1の期間単位で集約処理を実施し、前記第1の期間における前記集約ローレンツプロットを算出し、前記第1の期間における前記集約ローレンツプロットを出力することを特徴とする生体データ評価サーバ。 - 請求項2に記載の生体データ評価サーバであって、
前記ローレンツプロット生成部は、
前記第2の期間単位の前記ローレンツプロットから前記心拍間隔相当データの欠損の有無を判定し、前記心拍間隔相当データの欠損が生じた部分を前記第2の期間単位で除外してから前記集約処理を行って前記第1の期間における前記集約ローレンツプロットを算出することを特徴とする生体データ評価サーバ。 - 請求項2に記載の生体データ評価サーバであって、
前記ローレンツプロット生成部は、
前記集約処理として、前記第2の期間単位で算出した複数の前記ローレンツプロットを示すローレンツプロット行列の各行列成分について、前記第1の期間単位で統計値を算出することを特徴とする生体データ評価サーバ。 - 請求項2に記載の生体データ評価サーバであって、
前記ローレンツプロット生成部は、
前記第2の期間を、前記第1の期間未満かつ心拍変動解析に要求される時間幅以上とすることを特徴とする生体データ評価サーバ。 - 請求項2に記載の生体データ評価サーバであって、
前記ローレンツプロット生成部は、
前記心拍間隔相当データの欠損の時間幅の情報を受け付けて、前記第2の期間を前記心拍間隔相当データの欠損の時間幅の情報に基づいて決定することを特徴とする生体データ評価サーバ。 - 請求項2に記載の生体データ評価サーバであって、
前記集約ローレンツプロットにおいて、予め定義した領域における特徴量を算出して、前記特徴量に基づいて、不整脈様異常値の発生の程度を異常度データとして算出して出力する異常値慢性発生評価部を、さらに有することを特徴とする生体データ評価サーバ。 - 請求項7に記載の生体データ評価サーバであって、
前記第1の期間以上の第3の期間において、前記集約ローレンツプロットから前記不整脈様異常値の発生頻度を計算し、自律神経機能評価に適する前記対象者か否かを判定する不適者判定部を、さらに有することを特徴とする生体データ評価サーバ。 - 請求項8に記載の生体データ評価サーバであって、
前記不適者判定部の判定結果に基づいて、前記自律神経機能評価に適する対象者の前記心拍間隔相当データに基づいて自律神経機能指標を算出して出力する自律神経機能指標算出部を、さらに有することを特徴とする生体データ評価サーバ。 - 請求項9に記載の生体データ評価サーバであって、
前記不整脈様異常値に基づき、前記自律神経機能指標の信頼性を判定して、出力装置に前記自律神経機能指標の前記信頼性を表示する警告提示部を、さらに有することを特徴とする生体データ評価サーバ。 - 請求項10に記載の生体データ評価サーバであって、
前記自律神経機能指標に基づいて、前記対象者に事故又はインシデントが所定時間後に発生する確率を事故リスクとして算出する危険予測部を、さらに有し、
前記警告提示部は、
前記事故リスクが所定の閾値を超えた対象者に対して事故リスクの増大を警告するアラートを発報することを特徴とする生体データ評価サーバ。 - プロセッサとメモリを有する生体データ評価サーバと、
生体センサを有する移動体と、を有して生体データを評価する生体データ評価システムであって、
前記移動体は、
前記生体センサが対象者から心拍間隔相当データを含む生体データを検出して前記生体データ評価サーバに送信するデータ収集装置を有し、
前記生体データ評価サーバは、
前記生体データを受信して前記心拍間隔相当データを受け付けるデータ収集部と、
前記心拍間隔相当データから所定の期間でローレンツプロットを算出して、集約ローレンツプロットとして出力するローレンツプロット生成部と、
を有することを特徴とする生体データ評価システム。 - 請求項12に記載の生体データ評価システムであって、
前記所定の期間は、第1の期間と、前記第1の期間よりも短い第2の期間を含み、
前記ローレンツプロット生成部は、
前記心拍間隔相当データを前記第2の期間単位で前記ローレンツプロットを算出し、前記第2の期間単位で算出した前記ローレンツプロットを前記第1の期間単位で集約し、前記第1の期間における前記集約ローレンツプロットを算出し、前記第1の期間における前記集約ローレンツプロットを出力することを特徴とする生体データ評価システム。 - プロセッサとメモリを有する計算機が生体データを評価する生体データ評価方法であって、
前記計算機が、対象者の前記生体データから心拍間隔相当データを受け付けるデータ収集ステップと、
前記計算機が、前記心拍間隔相当データから所定の期間でローレンツプロットを算出して、集約ローレンツプロットとして出力するローレンツプロット生成ステップと、
を含むことを特徴とする生体データ評価方法。 - 請求項14に記載の生体データ評価方法であって、
前記所定の期間は、第1の期間と、前記第1の期間よりも短い第2の期間を含み、
前記ローレンツプロット生成ステップは、
前記心拍間隔相当データを前記第2の期間単位で前記ローレンツプロットを算出し、前記第2の期間単位で算出した前記ローレンツプロットを前記第1の期間単位で集約し、前記第1の期間における前記集約ローレンツプロットを算出し、前記第1の期間における前記集約ローレンツプロットを出力することを特徴とする生体データ評価方法。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US18/282,308 US20240153625A1 (en) | 2021-03-25 | 2021-11-25 | Biometric data evaluation server, biometric data evaluation system, and biometric data evaluation method |
CN202180095148.6A CN116916826A (zh) | 2021-03-25 | 2021-11-25 | 生物体数据评价服务器、生物体数据评价系统以及生物体数据评价方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2021052246A JP2022149902A (ja) | 2021-03-25 | 2021-03-25 | 生体データ評価サーバ、生体データ評価システム及び生体データ評価方法 |
JP2021-052246 | 2021-03-25 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022201639A1 true WO2022201639A1 (ja) | 2022-09-29 |
Family
ID=83395259
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2021/043239 WO2022201639A1 (ja) | 2021-03-25 | 2021-11-25 | 生体データ評価サーバ、生体データ評価システム及び生体データ評価方法 |
Country Status (4)
Country | Link |
---|---|
US (1) | US20240153625A1 (ja) |
JP (1) | JP2022149902A (ja) |
CN (1) | CN116916826A (ja) |
WO (1) | WO2022201639A1 (ja) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH10137228A (ja) * | 1996-11-07 | 1998-05-26 | Nissan Motor Co Ltd | メンタルストレス判定装置 |
US20150223760A1 (en) * | 2012-09-12 | 2015-08-13 | To Health Limited | Screening Procedure for Identifying Risk of Arrhythmia |
WO2019003549A1 (ja) * | 2017-06-28 | 2019-01-03 | ソニー株式会社 | 情報処理装置、情報処理方法及びプログラム |
WO2019193998A1 (ja) * | 2018-04-04 | 2019-10-10 | シャープ株式会社 | 状態取得装置、ネットワークシステム、および状態処理装置 |
JP2020140623A (ja) * | 2019-03-01 | 2020-09-03 | ヤマトロジスティクス株式会社 | 事故予測システム、事故予測方法及び事故予測プログラム |
WO2021049573A1 (ja) * | 2019-09-10 | 2021-03-18 | 学校法人産業医科大学 | 深部体温推定装置、深部体温推定方法、深部体温推定プログラム |
-
2021
- 2021-03-25 JP JP2021052246A patent/JP2022149902A/ja active Pending
- 2021-11-25 CN CN202180095148.6A patent/CN116916826A/zh active Pending
- 2021-11-25 WO PCT/JP2021/043239 patent/WO2022201639A1/ja active Application Filing
- 2021-11-25 US US18/282,308 patent/US20240153625A1/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH10137228A (ja) * | 1996-11-07 | 1998-05-26 | Nissan Motor Co Ltd | メンタルストレス判定装置 |
US20150223760A1 (en) * | 2012-09-12 | 2015-08-13 | To Health Limited | Screening Procedure for Identifying Risk of Arrhythmia |
WO2019003549A1 (ja) * | 2017-06-28 | 2019-01-03 | ソニー株式会社 | 情報処理装置、情報処理方法及びプログラム |
WO2019193998A1 (ja) * | 2018-04-04 | 2019-10-10 | シャープ株式会社 | 状態取得装置、ネットワークシステム、および状態処理装置 |
JP2020140623A (ja) * | 2019-03-01 | 2020-09-03 | ヤマトロジスティクス株式会社 | 事故予測システム、事故予測方法及び事故予測プログラム |
WO2021049573A1 (ja) * | 2019-09-10 | 2021-03-18 | 学校法人産業医科大学 | 深部体温推定装置、深部体温推定方法、深部体温推定プログラム |
Also Published As
Publication number | Publication date |
---|---|
CN116916826A (zh) | 2023-10-20 |
US20240153625A1 (en) | 2024-05-09 |
JP2022149902A (ja) | 2022-10-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10602942B2 (en) | Method of detecting abnormalities in ECG signals | |
EP2944251B1 (en) | Method, apparatus and computer-readable medium for confidence level determination of ambulatory hr algorithm based on a three-way rhythm classifier | |
CN113613559A (zh) | 用于描绘和分类的心电图处理系统 | |
US9936923B2 (en) | System and methods for generating predictive combinations of hospital monitor alarms | |
US10485489B1 (en) | System and method for assessing and monitoring the hemodynamic condition of a patient | |
JP2015226696A (ja) | 眠気検出方法及び眠気検出装置 | |
US20230211780A1 (en) | Operation support method, operation support system, and operation support server | |
US11432778B2 (en) | Methods and systems for patient monitoring | |
US8949072B2 (en) | Novelty detection with missing parameters | |
US8805484B2 (en) | System, apparatus and method for diagnosing seizures | |
CN114469131A (zh) | 自适应实时心电信号质量评估方法 | |
US11564613B2 (en) | Non-invasive continuous heart rhythm monitoring based on wearable sensors | |
WO2022201639A1 (ja) | 生体データ評価サーバ、生体データ評価システム及び生体データ評価方法 | |
JP2022051216A (ja) | 安全運行支援方法、安全運行支援システム及び安全運行支援サーバ | |
Aswathi et al. | Comparison Of Machine Learning Algorithms For Heart Rate Variability Based Driver Drowsiness Detection | |
CN115349824A (zh) | 健康预警方法、装置、计算机设备及存储介质 | |
JP2016120061A (ja) | 眠気推定装置及び眠気推定プログラム | |
JP7433428B2 (ja) | 生体信号処理装置および見守りシステム、見守り方法 | |
Sannino et al. | Effective supervised knowledge extraction for an mhealth system for fall detection | |
WO2018234427A1 (en) | DETECTION OF HYPOXIA | |
US11964664B2 (en) | Driving assistance method, driving assistance system, and server | |
CN113243919A (zh) | 一种列车司机疲劳状态识别及监控系统 | |
Verkoeijen | Identifying Risk Factors and Readmission to the Hospital Using Joint Modeling in Telemonitored Data of Patients with Heart Failure | |
JP2022149902A5 (ja) | ||
KR20230085526A (ko) | 생체신호를 이용한 수면 호흡 상태 분류 장치 및 방법 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21933237 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2301005304 Country of ref document: TH |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202180095148.6 Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 18282308 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 21933237 Country of ref document: EP Kind code of ref document: A1 |