JP2021040753A - Heat syncope sign detection system, heat syncope sign detection device, heat syncope sign detection method and computer program - Google Patents

Abstract

To provide a heat syncope sign detection system that can catch a sign reaching the heat syncope, a heat syncope sign detection device, a heat syncope sign detection method and a computer program.SOLUTION: A heat syncope sign detection system comprises: an electrocardiographic sensor for detecting an electro-cardio wave of a human body; a pulse wave sensor for detecting a pulse wave of the human body; a blood pressure estimation unit for estimating a blood pressure on the basis of the electro-cardio wave detected by the electrocardiographic sensor and the pulse wave detected by the pulse wave sensor; a wearable device including a heart rate estimation unit for estimating a heart rate on the basis of the pulse wave detected by the pulse wave sensor, and a transmission unit for transmitting the blood pressure estimated by the blood pressure estimation unit and the heart rate estimated by the heart rate estimation unit; a reception unit for receiving the blood pressure and the heart rate from the wearable device; and an information processing apparatus including a detection unit for detecting the heat syncope sign on the basis of the blood pressure and the heart rate received by the reception unit and a notification unit for performing a predetermined notification on the basis of detection results of the detection unit.SELECTED DRAWING: Figure 1

Description

The present invention is a heat fainting sign detection system that detects and notifies a sign of heat fainting based on a blood pressure estimated based on cardiac radio waves and pulse waves and a heart rate estimated based on the pulse wave. It relates to a fainting sign detection device, a heat fainting sign detection method, and a computer program.

In recent years, due to the effects of global warming and the heat island phenomenon, the number of midsummer days and hot days is increasing in various places in summer. Not only the elderly and other vulnerable people, but also those who work outdoors during the day and young people who exercise in the scorching sun are often transported by emergency due to heat stroke. Heat stroke often develops without the person being aware of it, and it is important to deal with it as soon as possible if there is a sign of it. Among heat strokes, especially in the relatively mild symptom called heat syncope, it is said that blood pressure decreases and the pulse becomes faster or slower and weaker.

For example, Patent Document 1 discloses a measurement belt that measures relative blood pressure fluctuations based on a correlation between electrocardiographic data obtained by a potential sensor and pulse wave velocity measured based on pulse wave data obtained by an electrocardiographic sensor. ing. It is said that this measuring belt can diagnose heat stroke from the data of the heart rate by the potential sensor, the breathing curve by the impedance sensor of the abdominal cavity, and the core body temperature by the non-contact temperature sensor. Each sensor is located on the abdomen.

Further, Patent Document 2 discloses a heat stroke onset risk calculation device that calculates the heat stroke onset risk by the ratio of the heart rate based on the heart rate information by the pulse sensor and the systolic blood pressure based on the blood pressure information by the blood pressure sensor. Has been done. This device calculates the risk of developing heat stroke by the ratio of the heart rate based on the pulse wave information by the pulse wave sensor, which is a 3-axis pressure sensor, to the systolic blood pressure based on the blood pressure calculated based on the pulse wave information. It is said that it can also be done.

Japanese Patent No. 6521345 Japanese Unexamined Patent Publication No. 2016-163694

However, in the technique disclosed in Patent Document 1, it is not considered to diagnose the relative blood pressure fluctuation and the heart rate in association with each other. Further, in the technique disclosed in Patent Document 2, since the risk of onset is calculated based on whether or not the ratio of heart rate and systolic blood pressure exceeds the threshold value, it is necessary to change the threshold value according to the season. .. Moreover, the techniques described in any of the patent documents could only diagnose the comprehensive symptom of heat stroke.

The present invention has been made in view of such circumstances, and an object of the present invention is a heat fainting sign detection system capable of catching a sign leading to heat fainting, a heat fainting sign detection device, and a heat fainting. The purpose is to provide a sign detection method and a computer program.

The heat fainting sign detection system according to one aspect of the present disclosure includes an electrocardiographic sensor that detects an electrocardiographic radio wave of the human body, a pulse wave sensor that detects a pulse wave of the human body, an electrocardiographic radio wave detected by the electrocardiographic sensor, and the pulse. A blood pressure estimation unit that estimates blood pressure based on a pulse wave detected by a wave sensor, a heart rate estimation unit that estimates a heart rate based on a pulse wave detected by the pulse wave sensor, a blood pressure estimated by the blood pressure estimation unit, and the above. A wearable device including a transmitter for transmitting the heart rate estimated by the heart rate estimation unit, a receiving unit for receiving blood pressure and heart rate from the wearable device, and heat fainting based on the blood pressure and heart rate received by the receiving unit. It includes a detection unit that detects a sign and an information processing device including a notification unit that performs a predetermined notification based on the detection result of the detection unit.

The method for detecting a sign of heat fainting according to one aspect of the present disclosure is to acquire a heartbeat of a human body, acquire a pulse wave of the human body, estimate blood pressure based on the acquired heartbeat and pulse wave, and obtain a pulse. The heart rate is estimated based on the wave, the sign of heat fainting in the human body is detected based on the estimated blood pressure and the estimated heart rate, and a predetermined notification is performed based on the detection result of the sign.

In this embodiment, the wearable device detects the cardiac radio waves and pulse waves of the human body, estimates the blood pressure from the cardiac radio waves and pulse waves, estimates the heart rate from the pulse waves, and transmits the estimated blood pressure and heart rate. .. The information processing device detects a sign of heat fainting based on the received blood pressure and heart rate, and gives a predetermined notification based on the detection result. As a result, it is possible to catch a sign of heat fainting and give a warning such as alerting.

In the heat fainting sign detection system according to one aspect of the present disclosure, the wearable device has a first-order differential value of the pulse wave detected by the pulse wave sensor from the time when the electrocardiographic radio wave detected by the electrocardiographic sensor peaks. A time difference calculation unit for calculating the time difference up to the maximum time point is further provided, and the blood pressure estimation unit estimates the blood pressure based on the time difference calculated by the time difference calculation unit.

In this embodiment, the blood pressure is estimated based on the time difference from the time when the cardiac radio wave peaks to the inflection point where the first derivative value of the pulse wave becomes maximum. As a result, of the two reference points that define a specific time difference for estimating blood pressure, the point where the slope of the rise of the pulse wave is maximum becomes the second reference point, so that the detection error of the specific time difference Can be reduced.

In the heat fainting sign detection system according to one aspect of the present disclosure, the wearable device has a second blood pressure estimation unit and a blood pressure estimation unit that estimate a second blood pressure based on a pulse wave detected by the pulse wave sensor. Further includes a blood pressure difference calculation unit that calculates the difference between the second blood pressure estimated by the second blood pressure estimation unit and the blood pressure estimated by the blood pressure estimation unit when the blood pressure is estimated. The blood pressure estimation unit estimates the blood pressure based on the second blood pressure newly estimated by the estimation unit and the difference calculated by the blood pressure difference calculation unit.

In this embodiment, when the blood pressure is estimated from the cardiac radio wave and the pulse wave, the second blood pressure is estimated from the pulse wave and the difference between them is calculated. After that, when the second blood pressure is newly estimated from the pulse wave, the blood pressure is estimated based on the newly estimated second blood pressure and the calculated difference. As a result, after the blood pressure is estimated once from the cardiac radio wave and the pulse wave, it is relatively based on the second blood pressure estimated from the pulse wave and the fixed difference between the two estimated blood pressure values with different estimation methods. Blood pressure can be continuously estimated with high accuracy.

In the heat fainting sign detection system according to one aspect of the present disclosure, the detection unit detects a heat fainting sign based on the magnitudes of the blood pressure decreasing tendency and the heart rate change tendency received by the receiving unit. ..

In this embodiment, it is possible to capture a sign of heat fainting based on the estimated magnitude of the decrease tendency of blood pressure and the estimated magnitude of the change tendency of heart rate.

In the heat fainting sign detection system according to one aspect of the present disclosure, the wearable device further includes a pulse pressure estimation unit that estimates pulse pressure, which is the amplitude of the pulse wave detected by the pulse wave sensor, and the transmission unit , The pulse pressure estimated by the pulse pressure estimation unit is further transmitted, and in the information processing device, the receiving unit further receives the pulse pressure from the wearable device, and the blood pressure and the pulse pressure received by the receiving unit are each received. When the heart rate is lower than the first threshold and the second threshold and the heart rate received by the receiving unit is higher than or lower than the third threshold, the detecting unit detects a sign of heat fainting.

In this embodiment, the blood pressure, heart rate and pulse pressure are estimated, the blood pressure is lower than the first threshold, the pulse pressure is lower than the second threshold, and the heart rate is higher than the third threshold or the fourth. Detects signs of heat fainting when it is below the threshold. This makes it possible to more accurately capture the signs of heat fainting.

In the thermal fainting sign detection system according to one aspect of the present disclosure, the wearable device detects the respiratory rate estimation unit that estimates the respiratory rate based on the pulse wave detected by the pulse wave sensor and the epidermal body temperature of the human body. The skin temperature sensor is further provided, the transmitting unit further transmits the respiratory rate estimated by the respiratory rate estimation unit and the epidermis body temperature detected by the epidermis temperature sensor, and the information processing apparatus is such that the receiving unit is wearable. The receiving unit further includes a statistical value calculation unit that further receives the respiratory rate and the epidermal body temperature from the device and calculates the average value and the standard deviation for each of the blood pressure, heart rate, respiratory rate, and epidermal body temperature received by the receiving unit. Based on the comparison result between the deviation from the average value of each of the newly received blood pressure, heart rate, respiratory rate and epidermal body temperature and the standard deviation, the detection unit detects a sign of heat fainting.

In this embodiment, the blood pressure, heart rate, and respiratory rate are estimated and the epidermal body temperature is detected, and the deviation from the average value for each of the blood pressure, heart rate, respiratory rate, and epidermal body temperature is compared with the standard deviation. Detect signs of heat faintness based on the results. This makes it possible to more accurately capture the signs of heat fainting.

In the thermal fainting sign detection system according to one aspect of the present disclosure, the wearable device further includes a pulse pressure estimation unit that estimates pulse pressure, which is the amplitude of the pulse wave detected by the pulse wave sensor, and the transmission unit , The pulse pressure estimation unit further transmits the pulse pressure estimated by the pulse pressure estimation unit, and the information processing device further receives the pulse pressure from the wearable device, and the blood pressure, heart rate and pulse pressure of the human body are input. In the case of a learning model that outputs information on the presence / absence of detection of a sign of heat fainting, the learning model is further provided with an acquisition unit that acquires the presence / absence information by inputting the blood pressure, heart rate, and pulse pressure received by the receiving unit. Based on the presence / absence information acquired by the unit, the detection unit detects a sign of the heat fainting.

In this embodiment, the blood pressure, heart rate, and pulse pressure are estimated, and the estimated blood pressure, heart rate, and pulse pressure are input to the learning model to acquire information on the presence or absence of detection of a sign of heat fainting. Detects signs of heat fainting based on the presence / absence information. As a result, AI (Artificial Intelligence) technology using a learning model can catch and notify signs of heat fainting.

In the thermal fainting sign detection system according to one aspect of the present disclosure, the wearable device detects the respiratory rate estimation unit that estimates the respiratory rate based on the pulse wave detected by the pulse wave sensor and the epidermal body temperature of the human body. The skin temperature sensor is further provided, the transmitting unit further transmits the respiratory rate estimated by the respiratory rate estimation unit and the epidermis body temperature detected by the epidermis temperature sensor, and the information processing apparatus is such that the receiving unit is wearable. The receiver is in a learning model that further receives the respiratory rate and epidermoid body temperature from the device and outputs information on the presence or absence of detection of a sign of heat fainting when the human body's blood pressure, heart rate, respiratory rate and epidermal body temperature are input. It further includes an acquisition unit that acquires the presence / absence information by inputting the received blood pressure, heart rate, respiratory rate, and epidermal body temperature, and the detection unit gives a sign of the heat fainting based on the presence / absence information acquired by the acquisition unit. Detect.

In this embodiment, the blood pressure, heart rate and respiratory rate are estimated and the epidermal body temperature is detected, and the estimated blood pressure, heart rate and respiratory rate and the detected epidermal body temperature are input to the learning model to cause heat fainting. The presence / absence information of the detection of the sign is acquired, and the sign of heat fainting is detected based on the acquired presence / absence information. As a result, it is possible to catch and notify a sign of heat fainting by AI technology using a learning model.

The heat fainting sign detection system according to one aspect of the present disclosure further includes a server device that distributes data, and the information processing device uses the communication unit to communicate with the server device and the learning model. It further includes a storage unit that is downloaded from the device and stored.

In this embodiment, the learning model downloaded from the server device is used. As a result, it is possible to catch and notify a sign of heat fainting by AI technology using the latest learning model that is updated in a timely manner.

The thermosyncope sign detection device according to one aspect of the present disclosure includes an electrocardiographic sensor that detects an electrocardiographic radio wave of the human body, a pulse wave sensor that detects a pulse wave of the human body, an electrocardiographic radio wave detected by the electrocardiographic sensor, and The blood pressure estimation unit estimates the blood pressure based on the pulse wave detected by the pulse wave sensor, the heart rate estimation unit estimates the heart rate based on the pulse wave detected by the pulse wave sensor, and the blood pressure estimation unit estimates. It is provided with a detection unit that detects a sign of heat fainting based on the blood pressure and the heart rate estimated by the heart rate estimation unit, and a notification unit that performs a predetermined notification based on the detection result of the detection unit.

In this embodiment, the human body's cardiac radio waves and pulse waves are detected, the blood pressure is estimated from the cardiac radio waves and pulse waves, the heart rate is estimated from the pulse waves, and the heat fainting is caused based on the estimated blood pressure and heart rate. Detects a sign and gives a predetermined notification based on the detection result. As a result, it is possible to catch a sign of heat fainting and give a warning such as alerting.

The computer program according to one aspect of the present disclosure estimates blood pressure based on the cardiac radio waves and pulse waves of the human body, communicates with a wearable device that estimates the heart rate based on the pulse wave, and uses the wearable device to detect the human body. The blood pressure and the heart rate are acquired, a sign of heat fainting in the human body is detected based on the acquired blood pressure and the obtained heart rate, and a computer is made to execute a process of performing a predetermined notification based on the detection result of the sign.

In this embodiment, the cardiac radio waves and pulse waves of the human body detected by the wearable device are acquired, and the signs of heat fainting are detected based on the acquired blood pressure and heart rate to perform a predetermined notification. As a result, it is possible to catch a sign of heat fainting and give a warning such as alerting.

The computer program according to one aspect of the present disclosure is a learning model that outputs information on the presence or absence of detection of a sign of heat fainting when the blood pressure and heart rate of a human body are input to the computer, and the blood pressure acquired from the wearable device. And the heart rate is input to acquire the presence / absence information, and the process of detecting the sign of the heat fainting is executed based on the acquired presence / absence information.

In this embodiment, the blood pressure, heart rate, and pulse pressure acquired from the wearable device are input to the learning model to acquire the presence / absence information of the detection of the sign of heat fainting, and the sign of heat fainting is obtained based on the acquired presence / absence information. Is detected. As a result, it is possible to catch and notify a sign of heat fainting by AI technology using a learning model.

According to the present invention, it is possible to capture a sign leading to heat fainting.

It is explanatory drawing which shows typically the configuration example of the sign detection system of heat fainting which concerns on Embodiment 1. FIG. It is a block diagram which shows the configuration example of a wearable device. It is a perspective view which shows the appearance of the wearable device. It is a block diagram which shows the configuration example of a mobile phone. It is explanatory drawing which shows the operation at the time of detecting a cardiac radio wave. It is a graph which shows the 1st example of the time difference for blood pressure estimation with the waveform of the detected cardiac radio wave and pulse wave. It is a graph which shows the 2nd example of the time difference for blood pressure estimation with the waveform of the detected cardiac radio wave and pulse wave. It is a chart which shows the measured value of PTT1. It is a chart which shows the measured value of PTT2. It is a chart which shows the mean square error and the fit coefficient of the BP-PTT approximation curve. It is a flowchart which shows the processing procedure of the control part which estimates the 2nd blood pressure, the heart rate, the pulse pressure and the respiratory rate based on a pulse wave. It is a flowchart which shows the processing procedure of the control part which transmits the estimated blood pressure, heart rate, pulse and respiration rate, and the detected epidermal body temperature. FIG. 5 is a flowchart showing a processing procedure of a control unit that acquires blood pressure and heart rate with a mobile phone according to the first embodiment and detects a sign of heat fainting. It is explanatory drawing which shows the 1st notification example with a sign of heat fainting. It is a flowchart which shows the processing procedure of the control part which detects the sign of heat fainting with the mobile phone which concerns on Embodiment 2. It is explanatory drawing which shows typically the configuration example of the sign detection system of heat fainting which concerns on Embodiment 3. It is a block diagram which shows the configuration example of a server apparatus. It is a flowchart which shows the processing procedure of the control part which relays the data received from the wearable device by the mobile phone which concerns on Embodiment 3 to a server device. FIG. 5 is a flowchart showing a processing procedure of a control unit that detects a sign of heat fainting in the server device according to the third embodiment. It is explanatory drawing which shows the 2nd notification example with a sign of heat fainting. It is explanatory drawing which shows the time-series display example of blood pressure. It is explanatory drawing which shows the time-series display example of the heart rate. It is explanatory drawing which shows the time-series display example of a respiratory rate. It is explanatory drawing which shows the time-series display example of the epidermal body temperature. It is a schematic diagram which shows the content example of the learning model which concerns on Embodiment 4. FIG. 5 is a flowchart showing a processing procedure of a control unit that acquires blood pressure, heart rate, and pulse pressure with a mobile phone according to the fourth embodiment and detects a sign of heat fainting. It is a flowchart which shows the processing procedure of the control part which stores the learning model distributed from the server apparatus which concerns on Embodiment 5. It is a schematic diagram which shows the content example of the learning model which concerns on Embodiment 6.

Hereinafter, the present invention will be described in detail with reference to the drawings showing the embodiments thereof.
(Embodiment 1)
In each of the following embodiments of the first embodiment, a system for notifying when a sign of heat fainting is detected based on the vital data of the user will be described. Heat syncope is a symptom included in so-called heat stroke. Heat stroke includes heat fainting (lightheadedness), heat cramps (leg cramps, etc.), heat exhaustion (headache, nausea, vomiting, diarrhea, malaise, collapse, decreased judgment and concentration, etc.), heat stroke (high) Symptoms become more severe in the order of body temperature). Here, a sign is detected at the stage of heat fainting, which has relatively mild symptoms, and a notification to that effect is given.

Heat syncope is said to be caused by dilation of cutaneous blood vessels, which lowers blood pressure and impairs blood flow to the brain. Other symptoms of heat syncope are known to increase or decrease heart rate and decrease pulse pressure. Therefore, in the first embodiment, the sign of heat fainting is detected based on the blood pressure and the heart rate.

FIG. 1 is an explanatory diagram schematically showing a configuration example of a heat fainting sign detection system according to the first embodiment. When the thermal fainting sign detection system 100a detects a thermal fainting sign based on the wearable device 1 which is attached to the user's arm to detect vital data and the data acquired from the wearable device 1, it indicates that fact. It is configured to include a mobile phone 2 (corresponding to an information processing device) for notifying. The wearable device 1 and the mobile phone 2 can transmit and receive information to and from each other via Bluetooth (registered trademark).

FIG. 2 is a block diagram showing a configuration example of the wearable device 1. FIG. 3 is a perspective view showing the appearance of the wearable device 1. The wearable device 1 is a wristwatch-type device having a belt 19 made of TPU (Thermoplastic Polyurethane), and includes a control unit 10, a storage unit 11, a display unit 12, an operation unit 13, and a Bluetooth communication unit 14. The wearable device 1 further includes a voltage sensor 15 (corresponding to an electrocardiographic sensor), an optical sensor 16 (corresponding to a pulse wave sensor), and a temperature sensor 17 (corresponding to an epidermal body temperature sensor) as sensors for detecting vital data. ..

The control unit 10 includes a processor such as a CPU (Central Processing Unit) and an MPU (Micro-Processing Unit), a timer, and an internal memory. The control unit 10 may be configured as one piece of hardware (SoC: System On a Chip) in which a processor, a timer, an internal memory, a storage unit 11, a Bluetooth communication unit 14, and the like are integrated. The control unit 10 controls the entire device by executing the control program stored in the storage unit 11.

The storage unit 11 includes a non-volatile memory such as a flash memory and a rewritable memory such as a DRAM (Dynamic Random Access Memory) and a SRAM (Static Random Access Memory). The non-volatile memory stores in advance a control program executed by the control unit 10 and various data. The rewritable memory stores data that is temporarily generated. The rewritable memory may be configured as a flash memory.

The display unit 12 is, for example, a display using an organic EL display, and is controlled by the control unit 10 to display various types of information. The operation unit 13 is an interface for receiving an operation by the user, and is composed of, for example, a touch key.

The Bluetooth communication unit 14 is an interface for connecting to an external device such as a mobile phone 2 by wireless communication conforming to the Bluetooth standard.

The voltage sensor 15 is a sensor for detecting the user's cardiac radio waves, and has a ground electrode (not shown) arranged on the back side of the display unit 12 and side electrodes 151 (not shown) arranged on both sides of the display unit 12. The potential difference between one of the two (not shown) is detected as a cardiac radio wave.

The optical sensor 16 detects a so-called photoelectric volume pulse wave (hereinafter, also simply referred to as a pulse wave) based on the reflected light obtained by irradiating the green light. The photoelectric volume pulse wave is circulatory system vital data obtained by measuring the amount of oxidized hemoglobin in capillaries from the amount of reflected light. A hole for irradiating and receiving green light is opened on the back side (not shown) of the display unit 12.

The temperature sensor 17 is a small thin film thermistor having good thermal responsiveness, and quickly detects the epidermal body temperature of the skin.

FIG. 4 is a block diagram showing a configuration example of the mobile phone 2. The mobile phone 2 is, for example, a smartphone, but may be a wearable device such as a tablet terminal, a general-purpose PC (Personal Computer), or a smart watch. The mobile phone 2 includes a control unit 20, a storage unit 21, a display unit 22, an operation unit 23, a Bluetooth communication unit 24, a public wireless communication unit 25, a Wi-Fi communication unit 26, and a voice input / output unit 27. The operation unit 23 is a touch panel integrated with the display unit 22, but is not limited thereto.

The control unit 20 includes one or a plurality of processors such as a CPU, an MPU, and a GPU (Graphics Processing Unit), a timer, and an internal memory. The control unit 20 may be configured as one SoC in which a processor, a timer, an internal memory, a storage unit 21, a Bluetooth communication unit 24, a public wireless communication unit 25, a Wi-Fi communication unit 26, and the like are integrated. The control unit 20 controls the entire device by executing the control program stored in the storage unit 21.

The storage unit 21 includes a flash memory, a non-volatile memory such as EPROM (Erasable Programmable Read Only Memory) and EEPROM (Electrically Erasable Programmable Read Only Memory) (registered trademark), and a rewritable memory such as DRAM and SRAM.

The non-volatile memory stores the control program executed by the control unit 20 and various data in advance, and also stores the application program 21a. The application program 21a may include a Web browser function, or a general-purpose Web browser program may be separately stored in the storage unit 21. In the application program 21a, the control unit 20 reads out what is recorded on the recording medium 29 via the Bluetooth communication unit 24, the public wireless communication unit 25, the Wi-Fi communication unit 26, or an input / output unit (not shown), and the storage unit 21. It may be a duplicate of. The rewritable memory stores data that is temporarily generated, and also stores the learning model X, which will be described later, in the storage area 21b.

The display unit 22 is a display device such as a liquid crystal display or an organic EL display, and is controlled by the control unit 20 to display various information. The operation unit 23 is an interface for receiving an operation by the user, and is composed of, for example, a touch panel and physical buttons.

The Bluetooth communication unit 24 is an interface for connecting to an external device such as a wearable device 1 by wireless communication conforming to the Bluetooth standard. The public wireless communication unit 25 is an interface for performing wireless communication via a mobile phone network by wireless communication conforming to the standard of a mobile communication system. The Wi-Fi communication unit 26 is an interface for connecting to a wireless LAN access point described later by wireless communication conforming to the Wi-Fi standard.

The voice input / output unit 27 includes a microphone and a speaker, and is used for inputting voice from the microphone to recognize the voice and expanding the voice to be notified to the outside from the speaker.

Next, the detection of the cardiac radio wave and the pulse wave and the estimation of the blood pressure in the wearable device 1 will be described. FIG. 5 is an explanatory diagram showing an operation when detecting a heart radio wave. FIG. 6 is a graph showing a first example of the waveforms of the detected cardiac radio waves and pulse waves and the time difference for estimating blood pressure. FIG. 7 is a graph showing a second example of the waveforms of the detected cardiac radio waves and pulse waves and the time difference for estimating blood pressure. As shown in FIG. 5, the wearable device 1 is mounted at a position separated from the root of the left ulnar styloid process of the user by 1.5 cm toward the upper arm side. This position is a position where the optical sensor 16 can reliably detect the pulse wave.

Cardiac radio waves are detected by the user continuing to touch the side electrodes 151 arranged on both sides of the display unit 12 inside the index finger and the middle finger of the right hand. The pulse wave continues to be detected as long as the wearable device 1 is properly attached. The first blood pressure (hereinafter, also simply referred to as blood pressure) is estimated based on the cardiac radio waves and pulse waves detected in this way. In addition, the second blood pressure is estimated based on the pulse wave. Details will be described later.

Moving to FIGS. 6 and 7, the waveform of the cardiac radio wave is shown in the upper part of each figure, and the waveform of the pulse wave is shown in the lower part of each figure. The vertical axis of each graph represents the relative amplitude, and the horizontal axis represents the relative position of the sampling points, that is, the relative time. The cardiac radio wave is said to consist of P wave, Q wave, R wave, S wave and T wave, and in FIGS. 6 and 7, the peak of the R wave is marked with a black circle.

Conventionally, blood pressure is estimated based on the pulse wave velocity (PTT = Pulse Transit Time) from the R wave to a specific location of the pulse wave. For example, in FIG. 6, the time difference from the peak of the R wave to the peak of the pulse wave (point marked with a black circle) is PTT1, and in FIG. 7, the time when the slope of the rise of the pulse wave from the peak of the R wave becomes maximum. The time difference to (the point marked with a black circle) is PTT2. The points marked with black circles in FIG. 7 are the points where the first-order differential value of the pulse wave becomes maximum, that is, the point where the second-order differential value of the pulse wave becomes 0. In FIGS. 6 and 7, since the cardiac radio waves and pulse waves for approximately 13 cycles are shown, 13 types of PTT1 and PTT2 are measured.

FIG. 8 is a chart showing the measured values of PTT1, and FIG. 9 is a chart showing the measured values of PTT2. In FIG. 8, the PTT1 from the first cycle to the thirteenth cycle varies from 326 ms to 342 ms, and the average value is calculated to be 333.5 ms and the standard deviation is calculated to be 4.4 ms. In FIG. 9, the PTT2 from the first cycle to the thirteenth cycle varies from 234 ms to 258 ms, and the average value is calculated to be 245.4 ms and the standard deviation is calculated to be 7.4 ms. Since the error is relatively small in each case, it may be a valid source data for calculating blood pressure.

Therefore, for 12 subjects (7 males and 5 females) between the ages of 20 and 30, blood pressure (SBP = systolic blood pressure, DBP = diastolic blood pressure, MBP = average blood pressure) and PTT1 and PTT2 were 9 The set was measured. However, MBP = (SBP-DBP) / 3 + SBP. The experimental methods for 9 sets are as follows.

(A) Measure 3 times for 2 minutes every 3 minutes at rest (b) Measure 3 times for 2 minutes every 3 minutes immediately after exercise (exercise 3 round trips on the stairs to the 2nd floor)
(C) Measured 3 times for 2 minutes every 3 minutes during the rest recovery process

After the measurement, 2 persons were randomly selected from 12 persons, and the relationship between SBP, DBP and MBP and PTT1 and PTT2 was fitted (linear approximation) by the least squares method. FIG. 10 is a chart showing the mean square error and the matching coefficient of the BP-PTT approximation curve. The upper and lower charts of the figure show the results of fitting the measured values for subject _1 and subject _2, respectively. In the cell at the intersection of SBP, DBP and MBP and PTT1 and PTT2 in each chart, the mean square error is shown on the upper side and the conformity coefficient is shown on the lower side. The suitability coefficient is also called the interphase coefficient, and the closer it is to 1, the better the approximation.

From the results shown in FIG. 10, it can be seen that the mean square error of PTT2 is smaller and the matching coefficient is larger than that of PTT1 except for one place. From this result, in each of the first and subsequent embodiments, the blood pressure (first blood pressure) is estimated with PTT2 as PTT. Specifically, the relational expressions of the following equations (1) to (3) were derived. Hereinafter, SBP (systolic blood pressure) is referred to as blood pressure.

SBP = -1.68 * PTT + 299.8 ... (1)
DBP = -1.44 * PTT + 337.4 ... (2)
MBP = -1.67 * PTT + 332.9 ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ (3)

For the method of estimating blood pressure (second blood pressure) only from pulse waves, see "Development of blood pressure estimation system using smartphone" (Hayato Fukushima, Graduate School of Information Science, Aichi Prefectural University, 2013 Master's Thesis Summary), "Blood pressure estimation method by pulse wave velocity method that can be applied even during exercise and can consider individual difference correction of vascular condition" (Proceedings of the 2017 Spring Meeting of the Japan Society for Precision Engineering, p229-P230), and Japanese Patent Application Laid-Open No. 2018 Since the details are detailed in Japanese Patent Publication No. 047219 (feature extraction device and method for detecting biological information, biological information detection device and wearable device), the description thereof is omitted here.

The wearable device 1 can estimate the heart rate and the respiratory rate in addition to the estimation of the blood pressure. The heart rate is easily estimated from the cardiac radio wave or the pulse wave, but here, the pulse wave is used to estimate the heart rate at a fixed cycle. For the relationship between the pulse wave and the respiratory rate and the method of estimating the respiratory rate from the pulse wave, "On the effect of the respiratory rate on the heart rate variability" (Mutshiro Nakao et al., Psychiatric and Physical Doctor, August 1995, Vol. 35, No. 6 No.) and JP-A-2010-53547 (methods and devices for measuring respiratory rate) are detailed, and thus the description thereof is omitted here.

By the way, the first blood pressure is characterized in that the estimation accuracy is higher than that of the second blood pressure, although the complicated operation shown in FIG. 5 is required at the time of estimation. The second blood pressure has a characteristic that the error with respect to the actual blood pressure is almost constant while the estimation accuracy is relatively low. Therefore, in each of the first and subsequent embodiments, when the first blood pressure is estimated, the estimated blood pressure is transmitted via the Bluetooth communication unit 14 and the difference between the second blood pressure and the first blood pressure is calculated. I will do it. When the second blood pressure is estimated after the set time, the blood pressure estimated based on the difference between the estimated second blood pressure and the calculated blood pressure is transmitted via the Bluetooth communication unit 14. As a result, blood pressure with relatively high estimation accuracy is transmitted at set time intervals.

On the other hand, the mobile phone 2 acquires the blood pressure and the heart rate from the wearable device 1 via the Bluetooth communication unit 24, and is based on the magnitude of the acquired decrease tendency of the blood pressure and the magnitude of the acquired change tendency of the heart rate. A sign of heat fainting is detected, and a predetermined notification is given to the display unit 22 based on the detection result. In this case, the external device such as another mobile phone or administrator PC may be notified.

Hereinafter, the operation of the wearable device 1 and the mobile phone 2 described above will be described with reference to a flowchart showing the operation. FIG. 11 is a flowchart showing a processing procedure of the control unit 10 for estimating the second blood pressure, heart rate, pulse pressure and respiratory rate based on the pulse wave. FIG. 12 is a flowchart showing a processing procedure of the control unit 10 for transmitting the estimated blood pressure, heart rate, pulse and respiratory rate and the detected epidermal body temperature. FIG. 13 is a flowchart showing a processing procedure of the control unit 20 that acquires blood pressure and heart rate with the mobile phone 2 according to the first embodiment and detects a sign of heat fainting.

The process of FIG. 11 is started at a cycle (for example, 10 minutes) set by the wearable device 1. The process of FIG. 12 is performed, for example, when the user starts the detection of the cardiac radio wave with the touch key of the operation unit 13, or when the user's finger touches the side electrode 151 and the detection of the cardiac radio wave is started. It will be started. The process of FIG. 13 is timely (for example, every minute) activated by the mobile phone 2. It is assumed that the cardiac radio wave detected by the voltage sensor 15 and the pulse wave detected by the optical sensor 16 are stored as waveform data for a time longer than a certain time in the buffers secured in the storage unit 11, respectively. ..

When the process of FIG. 11 is activated, the control unit 10 takes in the pulse wave stored in the storage unit 11 as waveform data (S11). Next, the control unit 10 estimates the second blood pressure while performing an operation including, for example, a product-sum operation (S12: corresponding to the second blood pressure estimation unit). The estimated second blood pressure is stored in the storage unit 11 (hereinafter, the same applies to other estimation results). Next, the control unit 10 estimates the heart rate from the cycle of the pulse wave (S13: corresponding to the heart rate estimation unit), and further estimates the pulse pressure from the amplitude of the pulse wave (S14: corresponds to the pulse pressure estimation unit). Next, the control unit 10 estimates the respiratory rate (S15: corresponding to the respiratory rate estimation unit) while performing an operation including a filter process, and ends the process of FIG. 11.

According to the process shown in FIG. 11, the second blood pressure, heart rate, pulse pressure, and respiratory rate are estimated in the same cycle, but when the cycles for estimating and transmitting these are set separately, the second blood pressure, heart rate, pulse pressure, and respiratory rate are estimated in the same cycle. , Each may be estimated and transmitted individually in an independent process that is started separately.

When the process of FIG. 12 is activated, the control unit 10 takes in the cardiac radio waves stored in the storage unit 11 as waveform data (S20). Next, the control unit 10 determines whether or not the cardiac radio wave has reached the peak of the R wave (S21), and if it has not reached the peak (S21: NO), waits until the peak of the R wave is reached. When the cardiac radio wave reaches the peak of the R wave (S21: YES), the control unit 10 takes in the pulse wave stored in the storage unit 11 as waveform data after starting the time counting by the timer (S22) (S23). ..

After that, the control unit 10 differentiates the pulse wave once (S24), and determines whether or not the differentiated value is positive, that is, whether or not the pulse wave rises (S25). If the differential value is not positive (S25: NO), the control unit 10 shifts the process to step S23 in order to capture the pulse wave again. When the differential value is positive (S25: YES), the control unit 10 further differentiates the differential value once (S26), and whether or not the second differential value is 0, that is, the pulse wave is an inflection point. Is determined (S27).

When the second differential value is not 0 (S27: NO), the control unit 10 shifts the process to step S23 in order to capture the pulse wave again. When the second differential value is 0 (S27: YES), the control unit 10 ends the time counting by the timer (S28), and sets the time measured by the timer as PTT (S29: corresponding to the time difference calculation unit). Then, the first blood pressure is estimated based on the PTT (S30: corresponding to the blood pressure estimation unit).

After that, the control unit 10 reads out the second blood pressure stored in the storage unit 11 (S31), and sets the value obtained by subtracting the first blood pressure from the second blood pressure as the blood pressure difference (S32: blood pressure difference calculation unit). Equivalent to). Further, the control unit 10 reads out the heart rate, pulse pressure, and respiratory rate stored in the storage unit 11 and periodically updated (S33), and takes in the epidermal body temperature detected by the temperature sensor 17 (S34). Next, the control unit 10 transmits the estimated first blood pressure, the heart rate, pulse pressure, and respiratory rate read from the storage unit 11 and the epidermal body temperature captured from the temperature sensor 17 via the Bluetooth communication unit 14. (S35: Corresponds to the transmitter).

After that, the control unit 10 starts time counting by the timer (S36), and determines whether or not 10 minutes have passed since the time counting was started (S37). This 10 minutes is the time that changes depending on the setting. If 10 minutes have not passed (S37: NO), the control unit 10 determines whether or not the detection of the cardiac radio wave has been restarted (S38), and if it has not been restarted (S38: NO), the step S37 is performed. Move the process. On the other hand, when the detection of the cardiac radio wave is restarted (S38: YES), the control unit 10 temporarily ends the process of FIG.

When 10 minutes have passed in step S37 (S37: YES), the control unit 10 reads the updated second blood pressure from the storage unit 11 (S39) during the 10 minutes, and the blood pressure is read from the read second blood pressure. The difference is subtracted to formally obtain the first blood pressure (S40: corresponding to the blood pressure estimation unit). The control unit 10 then shifts the process to step S33 again in order to transmit the first blood pressure, heart rate, pulse pressure and epidermal body temperature.

According to the process shown in FIG. 12, the first blood pressure, heart rate, pulse pressure, respiratory rate and epidermal body temperature are transmitted in the same cycle (10 minutes), but the cycle of estimating and transmitting these is When set separately, only the first blood pressure is estimated and transmitted by the process shown in FIG. 12, and the heart rate, pulse pressure, respiratory rate and epidermal body temperature are estimated and transmitted individually by the process shown in FIG. do it.

Next, when the process of the mobile phone 2 is started and the process of FIG. 13 is activated, the control unit 20 determines whether or not data has been received from the wearable device 1 (S41: corresponding to the receiving unit). If it is not received (S41: NO), it waits until it is received. When the data is received (S41: YES), the control unit 20 acquires the blood pressure and the heart rate from the received data (S42), and first calculates the magnitude of the decreasing tendency of the blood pressure (S43). The magnitude of the decreasing tendency here is, for example, the rate of decrease in blood pressure or the amount of decrease in blood pressure.

After that, the control unit 20 determines whether or not the magnitude of the decrease tendency of blood pressure is larger than a predetermined threshold value (S44), and if it is not large (S44: NO), FIG. 13 without detecting a sign of heat fainting. Ends the processing of. When the magnitude of the decrease tendency of blood pressure is larger than a predetermined threshold value (S44: YES), the magnitude of the change tendency of the heart rate is calculated (S45). The magnitude of the change tendency here is, for example, the rate of increase or decrease in heart rate or the amount of increase or decrease in heart rate.

After that, the control unit 20 determines whether or not the magnitude of the change tendency of the heart rate is larger than the second threshold value (S46), and if it is not large (S46: NO), the control unit 20 does not detect a sign of heat fainting. The process of FIG. 13 is terminated. When the magnitude of the change tendency of the heart rate is larger than the second threshold value (S46: YES), it is assumed that the sign of heat fainting is detected (corresponding to the detection unit), and the fact that there is a sign of heat fainting is notified (S47). : Corresponds to the notification unit), the process of FIG. 13 is terminated.

FIG. 14 is an explanatory diagram showing a first notification example with a sign of heat fainting. This notification is displayed on the display unit 22, but as described above, the same content may be notified to other external devices. In the example of notification in FIG. 14, signs or signs of heat fainting are emphasized and further encouraged to loosen clothing, cool the body and rehydrate. Not only the notification by display but also the notification by voice may be performed from the voice input / output unit 27.

As described above, according to the first embodiment, the wearable device 1 detects the user's cardiac radio wave and pulse wave, estimates the blood pressure from the cardiac radio wave and pulse wave, and estimates and estimates the heart rate from the pulse wave. Send blood pressure and heart rate. The mobile phone 2 detects a sign of heat fainting based on the received blood pressure and heart rate, and gives a predetermined notification based on the detection result. Therefore, it is possible to catch a sign of heat fainting and give a warning such as alerting.

Further, according to the first embodiment, the blood pressure is estimated based on the time difference from the time when the R wave of the cardiac radio wave peaks to the inflection point where the first derivative value of the pulse wave becomes maximum. Therefore, of the two reference points that define PTT for estimating blood pressure, the point where the slope of the rise of the pulse wave is maximum is the second reference point, so that the detection error of PTT can be reduced.

Further, according to the first embodiment, when the blood pressure (first blood pressure) is estimated from the cardiac radio wave and the pulse wave, the second blood pressure is estimated from the pulse wave, and the difference between them is calculated as the blood pressure difference. After that, when the second blood pressure is newly estimated from the pulse wave, the blood pressure is estimated based on the newly estimated second blood pressure and the calculated blood pressure difference. Therefore, once the blood pressure is estimated from the cardiac radio wave and the pulse wave, the blood pressure is relatively highly accurate based on the difference between the second blood pressure estimated from the pulse wave and the two estimated blood pressure values with different estimation methods. Can continue to be estimated.

Further, according to the first embodiment, it is possible to capture a sign of heat fainting based on the estimated rate of decrease or amount of decrease in blood pressure and the estimated rate of change or amount of change in heart rate.

(Embodiment 2)
The first embodiment detects a sign of heat fainting based on blood pressure and heart rate, whereas the second embodiment detects a sign of heat fainting based on blood pressure, heart rate and pulse pressure. is there. Since the configuration of the heat fainting sign detection system 100a according to the second embodiment is the same as that of the first embodiment, the corresponding parts are designated by the same reference numerals and the illustration and description thereof will be omitted.

In the second embodiment, the mobile phone 2 acquires the blood pressure, the heart rate and the pulse pressure from the wearable device 1, the acquired blood pressure is lower than the first threshold value, the pulse pressure is lower than the second threshold value, and the heart rate is the third. A sign of heat fainting is detected when it is above the threshold value or below the fourth threshold value. The operation of the wearable device 1 is the same as that of the first embodiment.

Hereinafter, the operation of the mobile phone 2 described above will be described with reference to a flowchart showing the operation. FIG. 15 is a flowchart showing a processing procedure of the control unit 20 for detecting a sign of heat fainting in the mobile phone 2 according to the second embodiment. The process of FIG. 15 is timely (for example, every minute) activated by the mobile phone 2.

When the process of FIG. 15 is activated, the control unit 20 determines whether or not data has been received from the wearable device 1 (S51: corresponding to the receiving unit), and if it does not receive (S51: NO), until it is received. stand by. When the data is received (S51: YES), the control unit 20 acquires the blood pressure, heart rate and pulse pressure from the received data (S52), and determines whether or not the acquired blood pressure is lower than the first threshold value (S51: YES). S53). When the blood pressure is not lower than the first threshold value (S53: NO), the control unit 20 ends the process of FIG. 15 without detecting a sign of heat fainting.

When the blood pressure is lower than the first threshold value (S53: YES), the control unit 20 determines whether or not the acquired pulse pressure is lower than the second threshold value (S54). When the pulse pressure is not lower than the second threshold value (S54: NO), the control unit 20 ends the process of FIG. 15 without detecting a sign of heat fainting. When the pulse pressure is lower than the second threshold value (S54: YES), the control unit 20 determines whether or not the acquired heart rate is higher than the third threshold value (S55).

When the heart rate is not higher than the third threshold value (S55: NO), the control unit 20 determines whether or not the heart rate is lower than the fourth threshold value (S56). When the heart rate is not less than the fourth threshold value (S56: NO), the control unit 20 ends the process of FIG. 15 without detecting a sign of heat fainting. When the heart rate is higher than the third threshold value (S55: YES) or lower than the fourth threshold value (S56: YES), the control unit 20 considers that it has detected a sign of heat fainting (corresponding to the detection unit) and heat. Notifies that there is a sign of fainting (S57: corresponding to the notification unit), and ends the process of FIG. The content of the notification in step S57 is the same as that shown in FIG.

As described above, according to the second embodiment, the wearable device 1 estimates the blood pressure, the heart rate and the pulse pressure, the blood pressure is lower than the first threshold value, the pulse pressure is lower than the second threshold value, and the heart rate is high. The mobile phone 2 detects a sign of heat fainting when it is greater than or less than the third threshold. Therefore, it is possible to more accurately grasp the signs leading to heat fainting.

(Embodiment 3)
The first embodiment is a form in which the mobile phone 2 detects a sign of heat fainting, whereas the third embodiment is a form in which a server device detects a sign of heat fainting. Further, in the first embodiment, the sign of heat fainting was detected based on the blood pressure and the heart rate, and in the second embodiment, the sign of the heat fainting was detected based on the blood pressure, the heart rate and the pulse pressure. Detects signs of heat syncope based on heart rate, respiration rate and epidermal body temperature.

FIG. 16 is an explanatory diagram schematically showing a configuration example of a heat fainting sign detection system according to the third embodiment. The thermal fainting sign detection system 100b further includes a server device 3 (corresponding to an information processing device) and an administrator PC 4 in addition to the wearable device 1 and the mobile phone 2 according to the first embodiment. The mobile phone 2, the server device 3, and the administrator PC 4 can send and receive information to and from each other via, for example, the Internet Ni. The thermal fainting sign detection system 100b may include a plurality of sets of the wearable device 1 and the mobile phone 2.

FIG. 17 is a block diagram showing a configuration example of the server device 3. The server device 3 includes a control unit 30, a storage unit 31, and a communication unit 34. The control unit 30 is a processor using a CPU, MPU, or GPU, and includes an internal memory and a clock such as a timer. The control unit 30 executes each process based on the server program 31P stored in the storage unit 31 to make the general-purpose server computer function as an information processing device.

The storage unit 31 uses a hard disk and a DRAM, and stores setting information referred to by the control unit 30 and temporarily generated data in addition to the server program 31P. The server program 31P may be a program recorded on the recording medium 39, read by the control unit 30 from an external arbitrary server device via the communication unit 34, and duplicated in the storage unit 31.

The communication unit 34 includes a network card. The control unit 30 can communicate via the Internet Ni by the communication unit 34.

The mobile phone 2 accesses the Internet Ni via the mobile phone network by the public wireless communication unit 25 shown in FIG. 4 of the first embodiment, or accesses the Internet Ni via the access point by the Wi-Fi communication unit 26. Can be accessed. When the mobile phone 2 acquires data such as blood pressure from the wearable device 1, the acquired data is transferred to the server device 3, and when notified from the server device 3, the notification content is displayed on the display unit 22 and is necessary. The voice is expanded from the voice input / output unit 27 accordingly.

The server device 3 detects a sign of heat fainting on behalf of the mobile phone 2 based on data such as blood pressure transferred from the mobile phone 2, and notifies the administrator PC 4 and the mobile phone 2 to that effect. When a plurality of sets of the wearable device 1 and the mobile phone 2 are included, each mobile phone 2 is identified by, for example, an ID.

As a method of detecting the sign of heat fainting, the method of discriminating by the discriminant function shown in the following equation (4) is
"Development of a system for determining core body temperature rise to prevent heat stroke" (Yasuyuki Watai et al., Biomedical Engineering, 2016, 54 Annual, Proc, p.3T5-3-6-1-3T5-3-6-2 ) Is introduced.

Z = 0.06Y1-0.02Y2 + 0.3Y3-17.13 ... (4)
However,
Y1: Heart rate Y2: Respiratory rate Y3: Body surface temperature (epidermal body temperature)
Z ≧ 0: Heat stroke reserve group Z <0: Healthy group

According to formula (4), 90% or more of the people with a core body temperature of 37.5 ° C or higher were correctly identified, but those with a core body temperature of less than 37.5 ° C were in the healthy group but in the heat stroke preliminary group. It is said that there were many cases in which it was mistakenly determined. Therefore, in the third embodiment, the accuracy of detecting the sign of heat stroke is improved by adding the blood pressure, which is predicted to decrease in the case of the symptom of heat fainting, to the analysis element.

Specifically, the average value and standard deviation of the blood pressure, heart rate, respiratory rate and epidermal body temperature of each user are calculated in chronological order, and the latest average values of blood pressure, heart rate, respiratory rate and epidermal body temperature are calculated respectively. When the deviation from and exceeds the standard deviation, a sign of heat fainting is detected and a notification to that effect is given.

Hereinafter, the operation of the mobile phone 2 and the server device 3 described above will be described with reference to a flowchart showing the operation. FIG. 18 is a flowchart showing a processing procedure of the control unit 20 that relays the data received from the wearable device 1 by the mobile phone 2 according to the third embodiment to the server device 3. FIG. 19 is a flowchart showing a processing procedure of the control unit 30 that detects a sign of heat fainting in the server device 3 according to the third embodiment. The processes of FIGS. 18 and 19 are started in a timely manner (for example, every minute). The processing of FIG. 19 is started in parallel by the number of mobile phones 2 identified by the ID.

When the process of FIG. 18 is activated by the mobile phone 2, the control unit 20 determines whether or not data has been received from the wearable device 1 (S61: corresponding to the reception unit). When the data is not received (S61: NO), the control unit 20 further determines whether or not the notification has been received from the server device 3 (S62), and when it does not receive the data (S62: NO), the wearable device 1 again determines. The process is moved to step S61 in order to wait for the data.

When data is received in step S61 (S61: YES), the control unit 20 acquires blood pressure, heart rate, respiratory rate and epidermal body temperature from the received data (S63), and transmits the acquired data to the server device 3. (S64) The process of FIG. 18 is completed.

When the notification is received in step S62 (S62: YES), the control unit 20 displays the notification content on the display unit 22 (S65), and in some cases, the notification content is loudened from the voice input / output unit 27 (S66). The processing of 18 is completed.

When the process of FIG. 19 is activated by the server device 3, the control unit 30 determines whether or not data has been received from the mobile phone 2 (S71: corresponding to the receiving unit), and does not receive the data (S71: NO). , Wait until it is received. When the data is received (S71: YES), the control unit 30 acquires the blood pressure, heart rate, respiratory rate and epidermal body temperature from the received data (S72), and obtains the mean value and standard deviation for each of the acquired data. Calculate (S73: Corresponds to the statistical value calculation unit). It is assumed that each data received from the mobile phone 2 is stored in a buffer reserved in the storage unit 31 in chronological order.

Next, the control unit 30 calculates a deviation from the average value of the acquired blood pressure (S74), and determines whether or not the absolute value of the deviation is larger than the standard deviation (S75). When the absolute value of the deviation is larger than the standard deviation (S75: YES), the control unit 30 stores in the storage unit 31 that the blood pressure exceeds the threshold value in association with the acquired blood pressure data (S76).

When the absolute value of the blood pressure deviation is not larger than the standard deviation (S75: NO) or when the process of step S76 is completed, the control unit 30 calculates the deviation from the average value of the acquired heart rate (S77). It is determined whether or not the absolute value of the deviation is larger than the standard deviation (S78). When the absolute value of the deviation is larger than the standard deviation (S78: YES), the control unit 30 stores in the storage unit 31 that the heart rate exceeds the threshold value in association with the acquired heart rate data (S79). ..

When the absolute value of the deviation of the heart rate is not larger than the standard deviation (S78: NO) or when the process of step S79 is completed, the control unit 30 calculates the deviation from the average value of the acquired breathing rate (S80). , It is determined whether or not the absolute value of the deviation is larger than the standard deviation (S81). When the absolute value of the deviation is larger than the standard deviation (S81: YES), the control unit 30 stores in the storage unit 31 that the respiratory rate exceeds the threshold value in association with the acquired respiratory rate data (S82). ..

When the absolute value of the deviation of the respiratory rate is not larger than the standard deviation (S81: NO) or when the process of step S82 is completed, the control unit 30 calculates the deviation from the average value of the acquired epidermal body temperature (S83). , It is determined whether or not the absolute value of the deviation is larger than the standard deviation (S84). When the absolute value of the deviation is larger than the standard deviation (S84: YES), the control unit 30 stores in the storage unit 31 that the epidermis temperature exceeds the threshold value in association with the acquired epidermis temperature data (S85). ..

When the absolute value of the deviation of the epidermal body temperature is not larger than the standard deviation (S84: NO) or when the process of step S85 is completed, the control unit 30 determines the absolute value of all the deviations of the blood pressure, the heart rate, the respiratory rate and the epidermal body temperature. It is determined whether the value is larger than the standard deviation (S86). When the absolute values of all deviations are larger than the standard deviation (S86: YES), the control unit 30 assumes that the sign of heat fainting has been detected (corresponding to the detection unit), and the administrator PC4 indicates that there is a sign of heat fainting. And the mobile phone 2 is notified (S87: corresponding to the notification unit), and the process of FIG. 19 is terminated. When the absolute values of all the deviations are not larger than the standard deviations (S86: NO), the control unit 30 ends the process of FIG. 19 without detecting a sign of thermal fainting. The content of the notification in step S87 is the same as that shown in FIG.

FIG. 20 is an explanatory diagram showing a second notification example with a sign of heat fainting. This notification is displayed on the administrator PC 4, and includes a warning display similar to the display example shown in FIG. 14 of the first embodiment, and for each wearable device 1 (wearable watch) identified by an ID. The presence or absence of an abnormality is displayed in. The mobile phone 2 that has transmitted data showing a sign of heat fainting is notified as shown in FIG. 14 of the first embodiment. This notification is displayed on the display unit 22 as in the first embodiment.

Next, the display of vital signs on the mobile phone 2 will be described. The user of the mobile phone 2 can receive the vital sign time-series display service by activating the browser from the screen of the display unit 22 and accessing the website provided by the server device 3. FIG. 21 is an explanatory diagram showing a time-series display example of blood pressure. FIG. 22 is an explanatory diagram showing a time-series display example of the heart rate. FIG. 23 is an explanatory diagram showing an example of displaying the respiratory rate in time series. FIG. 24 is an explanatory diagram showing an example of time-series display of epidermal body temperature.

In each of FIGS. 21 to 24, the magnitudes of vital signs (blood pressure, heart rate, respiratory rate and epidermal body temperature) for one day on a specified day are plotted on a graph and time series as numerical data. It is displayed. Some past displays are omitted on the graph. Vital signs that exceed the threshold are displayed in red circles (black circles in the examples of FIGS. 21 to 24) on the graph, and numerical data are displayed in red (bold in the examples of FIGS. 21 to 24). To.

The server device 3 acquires data from the mobile phone 2 in time series, and stores the acquired data in the storage unit 31 together with information on whether or not the data exceeds a threshold value. The server device 3 can generate the screens shown in FIGS. 21 to 24 and transmit the screens to the mobile phone 2 in response to the request from the mobile phone 2. The mobile phone 2 normally displays the vital sign screen received from the server device 3 on the browser.

As described above, according to the third embodiment, the wearable device 1 estimates the blood pressure, heart rate and respiratory rate and detects the epidermal body temperature, and the server device 3 detects the blood pressure, heart rate, respiratory rate and epidermal body temperature, respectively. The sign of heat fainting is detected based on the comparison result between the deviation from the mean value and the standard deviation. Therefore, it is possible to more accurately grasp the signs leading to heat fainting.

In the third embodiment, the server device 3 detects a sign of heat fainting, but the mobile phone 2 detects a sign of heat fainting based on blood pressure, heart rate, respiratory rate, and epidermal body temperature. May be good. In this case, the mobile phone 2 corresponds to the information processing device.

(Embodiment 4)
The second embodiment is a mode in which a sign of heat fainting is detected based on the comparison result between the blood pressure, the heart rate and the pulse pressure and the respective threshold values, whereas the fourth embodiment is an AI using the learning model X1. It is a form that detects signs of heat fainting by technology. Since the configuration of the heat fainting sign detection system 100a according to the fourth embodiment is the same as that of the first and second embodiments, the corresponding parts are designated by the same reference numerals and the illustration and description thereof will be omitted. The learning model X1 is stored in the storage area 21b of the storage unit 21 as a part of the learning model X shown in FIG. 4 of the first embodiment.

FIG. 25 is a schematic diagram showing a content example of the learning model X1 according to the fourth embodiment. The learning model X1 inputs numerical data of blood pressure, heart rate, and pulse pressure at each of the time points t1, t2, t3, and so on, and the probability that a monitoring target exists (that is, the monitoring target is detected) during the input and any of them. The output is the probability that there is no monitoring target (that is, no monitoring target is detected). The probability of output by each output node of the output layer is a value of 0 to 1.0, and the total of the probabilities of output by all output nodes is 1.0. The object to be monitored here is a sign of heat fainting.

The learning model X1 obtains information on the presence / absence of detection of a sign of heat fainting when teacher data including numerical data of a person's blood pressure, heart rate, and pulse pressure and information for identifying the presence / absence of a sign of heat fainting is input. A model trained to output. Specifically, the numerical data of blood pressure, heart rate, and pulse pressure of a person who has a sign of heat fainting is labeled with a sign of heat fainting and collected in large quantities, and the collected numerical data is used as a learning model X1. Input in sequence and let them learn. For those who do not have a sign of heat fainting, the same numerical data is labeled with no sign, collected in large quantities, and trained by the learning model X1.

For the learning model X1, for example, a multi-layer recurrent neural network (RNN: Recurrent Neural Network) learned by deep learning can be used. Instead of RNN, those learned by other machine learning such as a convolutional neural network (CNN) may be used. The RNN includes an intermediate layer between the input layer and the output layer. The intermediate layer has a plurality of fully bonded layers, and the number of fully bonded layers can be appropriately determined.

There are a plurality of nodes in each of the input layer, the intermediate layer, and the output layer. The nodes of each layer are unidirectionally connected to the nodes existing in the previous and next layers with desired weights and biases. When the data input to the input layer is input to the intermediate layer, the output of one layer is calculated using the activation function including the weight and the bias, and the calculated output is input to the subsequent layer. In this case, in order to consider the influence between time points, there is a path for transmitting the output from the middle layer at one time point to the middle layer at the next time point. Thereby, for example, the intermediate layer at the time point t2 receives the input from the intermediate layer at the previous time point t1 in addition to the input from the input layer at the same time point t2. In the same manner below, the output of the intermediate layer is transmitted to other layers one after another until the output of the output layer is obtained. In this way, the learning model X1 optimizes data such as coefficients and thresholds of functions that define predetermined operations performed on input values.

When CNN is used for the learning model X1, the intermediate layer has a convolutional layer and a pooling layer composed of a plurality of stages, and a fully connected layer in the final stage. In this case, data for a predetermined time including time points t1, t2, t3, ... Is input to the input layer of FIG. 25 while shifting the time. The RNN and CNN may be input to the data on the time axis converted into the data on the frequency axis by an operation such as FFT (First Fourier Transform).

The mobile phone 2 acquires blood pressure, heart rate, and pulse pressure from the wearable device 1 in chronological order, inputs them to the learning model X1, and detects a sign of heat fainting based on the presence / absence information output from the learning model X1. .. There is a one-to-one correspondence between the time series in which the mobile phone 2 acquires blood pressure, heart rate, and pulse pressure, and the time points t1, t2, t3 ... When each numerical data is input to the learning model X1.

Hereinafter, the operation of the mobile phone 2 described above will be described with reference to a flowchart showing the operation. FIG. 26 is a flowchart showing a processing procedure of the control unit 20 that acquires blood pressure, heart rate, and pulse pressure with the mobile phone 2 according to the fourth embodiment and detects a sign of heat fainting. The process of FIG. 26 is timely (for example, every minute) activated by the mobile phone 2.

When the process of FIG. 26 is activated, the control unit 20 determines whether or not data has been received from the wearable device 1 (S91: corresponding to the receiving unit), and if it does not receive (S91: NO), until it is received. stand by. When the data is received (S91: YES), the control unit 20 acquires blood pressure, heart rate and pulse pressure from the received data (S92), inputs each acquired data into the learning model X1 (S93), and learns. Obtain information on the presence or absence of detection of a sign of heat fainting from model X1 (S94: corresponding to the acquisition unit).

The control unit 20 determines whether or not the acquired presence / absence information indicates that there is no sign of heat fainting (S95), and if it indicates that there is no detection (S95: YES), the process of FIG. 26 is performed without notifying. finish. When the presence / absence information does not indicate no sign of heat fainting (S95: NO), the control unit 20 further determines whether or not the presence / absence information of the sign of heat fainting indicates that the sign is detected (S96). ). Whether or not to indicate the presence or absence of detection is determined, for example, whether or not the probability of the presence or absence of detection is greater than 0.6. The judgment threshold is not limited to 0.6.

When the presence / absence information indicates that a sign is detected (S96: YES), the control unit 20 displays on the display unit 22 that there is a sign of heat fainting, assuming that the sign of heat fainting has been detected (corresponding to the detection unit). (S97: Corresponding to the notification unit), and the process of FIG. 26 is terminated. The content of the notification in step S97 is the same as that shown in FIG.

As described above, according to the fourth embodiment, the wearable device 1 estimates the blood pressure, the heart rate, and the pulse pressure, and the mobile phone 2 inputs the blood pressure, the heart rate, and the pulse pressure into the learning model X1 to cause heat fainting. The presence / absence information of the detection of the sign of the blood pressure is acquired, and the sign of heat fainting is detected based on the acquired presence / absence information. Therefore, AI (Artificial Intelligence) technology using a learning model can catch and notify a sign of heat fainting. It is also possible to detect a sign of heat fainting based on the presence / absence information output when the blood pressure and the heart rate are input without inputting the pulse pressure into the learning model X1.

In the fourth embodiment, the sign of heat fainting was detected by the mobile phone 2, but the sign of heat fainting was detected by inputting the blood pressure, heart rate and pulse pressure into the learning model X1 by the server device 3. You may. In this case, the server device 3 corresponds to the information processing device. Further, the learning model X1 may input blood pressure and heart rate.

(Embodiment 5)
In the fourth embodiment, the learning model X1 is stored in the storage unit 21 in advance, but the learning model X1 is downloaded and stored in the storage unit 21 after being learned by a learning device composed of a personal computer, a server computer, or the like. It may be read from a portable recording medium that stores the learning model X1 and stored in the storage unit 21. The fifth embodiment is a mode in which the learning model X1 is downloaded from the server device 3.

Since the configuration of the heat fainting sign detection system 100b according to the fifth embodiment is the same as that shown in FIG. 16 of the third embodiment, the corresponding parts are designated by the same reference numerals and the illustration and description thereof are omitted. .. In the server device 3, the learning model X1 is learned and updated at any time using the teacher data. The mobile phone 2 downloads the updated learning model X1 from the server device 3 and stores it in the storage unit 21.

FIG. 27 is a flowchart showing a processing procedure of the control unit 20 that stores the learning model X1 delivered from the server device 3 according to the fifth embodiment. The process of FIG. 27 is started at a fixed cycle (for example, every hour), but the start cycle is not limited to this.

When the process of FIG. 27 is activated by the mobile phone 2, the control unit 20 determines whether or not there is a delivery notification from the server device 3 (S1), and when there is no delivery notification (S1: NO), The process of FIG. 27 is terminated without performing any special process.

When there is a delivery notification from the server device 3 (S1: YES), the control unit 20 downloads the learning model X1 from the server device 3 (S2) and stores it in the storage area 21b of the storage unit 21 (S3). The process of FIG. 27 ends. As a result, the contents of the learning model X1 are updated.

As described above, according to the fifth embodiment, the learning model X1 downloaded from the server device 3 is used. Therefore, it is possible to catch and notify a sign of heat fainting by AI technology using the latest learning model X1 that is updated in a timely manner.

(Embodiment 6)
In the fourth embodiment, when blood pressure, heart rate, and pulse pressure are input to the learning model X1, each probability of no sign of heat fainting and with detection of a sign of heat fainting is output. On the other hand, in the sixth embodiment, when blood pressure, heart rate, respiratory rate and epidermal body temperature are input to the learning model X2, the probability that no sign of heat fainting is detected and the sign of heat fainting are severe or moderate. And each probability of being mild is output.

Since the configuration of the heat fainting sign detection system 100a according to the sixth embodiment is the same as that of the first and second embodiments, the corresponding parts are designated by the same reference numerals and the illustration and description thereof will be omitted. The learning model X2 is stored in the storage area 21b of the storage unit 21 as a part of the learning model X shown in FIG. 4 of the first embodiment.

FIG. 28 is a schematic diagram showing a content example of the learning model X2 according to the sixth embodiment. The learning model X2 inputs numerical data of blood pressure, heart rate, respiratory rate, and epidermal body temperature at each of the time points t1, t2, t3, and so on, and the probability that a monitoring target exists (that is, the monitoring target is detected) during the input. And the probability that none of the monitoring targets exist (that is, no monitoring target is detected) is output. The probability of output by each output node of the output layer is a value of 0 to 1.0, and the total of the probabilities of output by all output nodes is 1.0. The object to be monitored here is a sign of heat fainting.

The learning model X2 detects a sign of heat fainting when teacher data including numerical data of a person's blood pressure, heart rate, respiratory rate, and epidermal body temperature and information for identifying the presence or absence of a sign of heat fainting is input. It is a model trained to output presence / absence information. Specifically, a large amount of numerical data of blood pressure, heart rate, respiratory rate, and epidermal body temperature of a person who has a sign of heat fainting is labeled with the degree of sign of heat fainting (severe / moderate / mild). Collected and the collected numerical data are sequentially input to the learning model X2 for training. For those who do not have a sign of heat fainting, the same numerical data is labeled with no sign, collected in large quantities, and trained by the learning model X2.

Similar to the learning model X1, the learning model X2 can use a multi-layer RNN learned by deep learning. Instead of RNN, those learned by other machine learning such as CNN may be used.

The flowchart showing the processing procedure of the control unit 20 in which the mobile phone 2 acquires the blood pressure, the heart rate, the respiratory rate and the epidermal body temperature to detect the sign of heat fainting in the sixth embodiment is shown in FIG. 26 of the fourth embodiment. Since it is similar to, the description here is omitted. However, in step S92 of FIG. 26, it is read so as to acquire blood pressure, heart rate, respiratory rate and epidermal body temperature. Further, the learning model X1 is read as the learning model X2. Further, in step S96, whether the sign is severe, moderate, or mild is determined by the magnitude of the probability (the one with the highest probability is adopted), and notification according to the degree of the determined sign is sent to step S97. To do.

Specifically, the degree of detected heat fainting (severe / moderate / mild) is displayed on the screen of the notification example shown in FIGS. 14 and 20. Such a notification is transferred from the mobile phone 2 to the wearable device 1, and for example, a part of the display unit 12 is changed to red / orange / yellow or the like so that the degree of heat fainting (severe / moderate / mild) can be identified. It may be displayed in different colors.

As described above, according to the sixth embodiment, the wearable device 1 estimates the blood pressure, the heart rate and the respiratory rate and detects the epidermoid body temperature, and the mobile phone 2 determines the blood pressure, the heart rate and the respiratory rate and the epidermoid body temperature. Is input to the learning model X2 to acquire the presence / absence information of the detection of the sign of heat fainting, and the sign of heat fainting is detected based on the acquired presence / absence information. Therefore, the AI technique using the learning model X2 can catch and notify the sign of heat fainting.

In the sixth embodiment, the sign of heat fainting was detected by the mobile phone 2, but the blood pressure, heart rate, respiratory rate and epidermal body temperature were input to the learning model X2 by the server device 3 to cause heat fainting. A sign may be detected. In this case, the server device 3 corresponds to the information processing device.

(Modification example)
In the first to sixth embodiments, a system including the wearable device 1 and the mobile phone 2 or the server device 3 detects a sign of heat fainting and notifies the fact. On the other hand, in the modified example, the wearable device 1 detects a sign of heat fainting and notifies the fact. The wearable device 1 according to the modified example has the same configuration as the wearable device 1 according to the first to sixth embodiments.

In this modification, for example, the control unit 10 of the wearable device 1 executes the processing of FIG. 14 of the first embodiment, FIG. 15 of the second embodiment, FIG. 19 of the third embodiment, or FIG. 26 of the fourth embodiment. However, the estimated blood pressure and other data are passed between the processing modules without being transmitted or received within the own device. When a sign of heat fainting is detected, an alarm may be notified by an icon, characters, or the like suitable for the display unit 12 having a small shape. The learning model X1 or X2 is stored in the storage unit 11.

As described above, according to the present modification, the wearable device 1 detects the user's cardiac radio wave and pulse wave, estimates the blood pressure from the cardiac radio wave and pulse wave, estimates the heart rate from the pulse wave, and estimates the blood pressure. And, the sign of heat fainting is detected based on the heart rate, and a predetermined notification is given based on the detection result. Therefore, only the wearable device 1 can catch a sign of heat fainting and give a warning such as alerting.

The embodiments disclosed this time should be considered as exemplary in all respects and not restrictive. The scope of the present invention is indicated by the scope of claims, not the above-mentioned meaning, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims. Also, the technical features described in each embodiment can be combined with each other.

100a, 100b Predictive detection system for thermal fainting 1 Wearable device 10 Control unit 14 Bluetooth communication unit 15 Voltage sensor 151 Side electrode 16 Optical sensor 17 Temperature sensor 2 Mobile phone 20 Control unit 21 Storage unit 21a App program 21b Storage area 22 Display unit 23 Operation unit 24 Bluetooth communication unit 29 Recording medium 3 Server device 30 Control unit 31 Storage unit 31p Server program 34 Communication unit 39 Recording medium 4 Administrator PC
Ni Internet X1, X2 learning model

Claims (13)

Electrocardiographic sensor that detects the electromagnetic waves of the human body,
A pulse wave sensor that detects the pulse wave of the human body,
A blood pressure estimation unit that estimates blood pressure based on the electrocardiographic radio wave detected by the electrocardiographic sensor and the pulse wave detected by the pulse wave sensor.
A wearable device including a heart rate estimation unit that estimates a heart rate based on a pulse wave detected by the pulse wave sensor, and a transmission unit that transmits the blood pressure estimated by the blood pressure estimation unit and the heart rate estimated by the heart rate estimation unit. When,
A receiver that receives blood pressure and heart rate from the wearable device,
A heat fainting device including a detection unit that detects a sign of heat fainting based on the blood pressure and heart rate received by the receiving unit and an information processing device including a notification unit that gives a predetermined notification based on the detection result of the detection unit. Predictive detection system. The wearable device is
A time difference calculation unit for calculating the time difference from the time when the electrocardiographic radio wave detected by the electrocardiographic sensor peaks to the time when the first-order differential value of the pulse wave detected by the pulse wave sensor becomes maximum is further provided.
The sign detection system for heat fainting according to claim 1, wherein the blood pressure estimation unit estimates the blood pressure based on the time difference calculated by the time difference calculation unit. The wearable device is
A second blood pressure estimation unit that estimates a second blood pressure based on the pulse wave detected by the pulse wave sensor, and a second blood pressure estimation unit that estimates the blood pressure when the blood pressure estimation unit estimates the blood pressure. It is further equipped with a blood pressure difference calculation unit that calculates the difference between the blood pressure and the blood pressure estimated by the blood pressure estimation unit.
The heat according to claim 1 or 2, wherein the blood pressure estimation unit estimates the blood pressure based on the second blood pressure newly estimated by the second blood pressure estimation unit and the difference calculated by the blood pressure difference calculation unit. Fainting sign detection system. The one according to any one of claims 1 to 3, wherein the detection unit detects a sign of heat fainting based on the magnitude of each of the decreasing tendency of blood pressure and the changing tendency of heart rate received by the receiving unit. Predictive detection system for heat fainting. The wearable device is
A pulse pressure estimation unit for estimating a pulse pressure, which is the amplitude of the pulse wave detected by the pulse wave sensor, is further provided.
The transmitting unit further transmits the pulse pressure estimated by the pulse pressure estimating unit, and the pulse pressure is further transmitted.
The information processing device
The receiving unit further receives the pulse pressure from the wearable device, and the receiving unit further receives the pulse pressure.
When the blood pressure and pulse pressure received by the receiving unit are lower than the first threshold value and the second threshold value, and the heart rate received by the receiving unit is higher than the third threshold value or lower than the fourth threshold value, the detecting unit is used. The system for detecting a sign of heat fainting according to any one of claims 1 to 4, wherein the system detects a sign of heat fainting. The wearable device is
Further equipped with a respiratory rate estimation unit that estimates the respiratory rate based on the pulse wave detected by the pulse wave sensor and an epidermal body temperature sensor that detects the epidermal body temperature of the human body.
The transmitting unit further transmits the respiratory rate estimated by the respiratory rate estimation unit and the epidermal body temperature detected by the epidermal body temperature sensor.
The information processing device
The receiver further receives respiratory rate and epidermal body temperature from the wearable device.
A statistical value calculation unit for calculating the average value and standard deviation for each of the blood pressure, heart rate, respiratory rate, and epidermal body temperature received by the receiving unit is further provided.
Claim 1 in which the detection unit detects a sign of heat fainting based on the comparison result between the deviation from the average value of each of the blood pressure, heart rate, respiratory rate and epidermal body temperature newly received by the reception unit and the standard deviation. The predictive detection system for heat fainting according to any one of claims 3. The wearable device is
A pulse pressure estimation unit for estimating a pulse pressure, which is the amplitude of the pulse wave detected by the pulse wave sensor, is further provided.
The transmitting unit further transmits the pulse pressure estimated by the pulse pressure estimating unit, and the pulse pressure is further transmitted.
The information processing device
The receiving unit further receives the pulse pressure from the wearable device, and the receiving unit further receives the pulse pressure.
The presence / absence of the blood pressure, heart rate, and pulse pressure received by the receiver is input to the learning model that outputs information on the presence / absence of detection of a sign of heat fainting when the blood pressure, heart rate, and pulse pressure of the human body are input. It also has an acquisition unit to acquire information.
The sign detection system for heat fainting according to any one of claims 1 to 3, wherein the detection unit detects the sign of heat fainting based on the presence / absence information acquired by the acquisition unit. The wearable device is
Further equipped with a respiratory rate estimation unit that estimates the respiratory rate based on the pulse wave detected by the pulse wave sensor and an epidermal body temperature sensor that detects the epidermal body temperature of the human body.
The transmitting unit further transmits the respiratory rate estimated by the respiratory rate estimation unit and the epidermal body temperature detected by the epidermal body temperature sensor.
The information processing device
The receiver further receives respiratory rate and epidermal body temperature from the wearable device.
Blood pressure, heart rate, respiratory rate and epidermal body temperature received by the receiver in a learning model that outputs information on the presence or absence of detection of signs of heat fainting when the human body's blood pressure, heart rate, respiratory rate and epidermal body temperature are input. Further provided with an acquisition unit for acquiring the presence / absence information by inputting
The sign detection system for heat fainting according to any one of claims 1 to 3, wherein the detection unit detects the sign of heat fainting based on the presence / absence information acquired by the acquisition unit. Including a server device that distributes data
The information processing device
The sign detection system for heat fainting according to claim 7 or 8, further comprising a communication unit that communicates with the server device and a storage unit that downloads and stores the learning model from the server device by the communication unit. An electrocardiographic sensor that detects the electromagnetic waves of the human body,
A pulse wave sensor that detects the pulse wave of the human body and
A blood pressure estimation unit that estimates blood pressure based on the electrocardiographic radio wave detected by the electrocardiographic sensor and the pulse wave detected by the pulse wave sensor, and
A heart rate estimation unit that estimates the heart rate based on the pulse wave detected by the pulse wave sensor, and
A detection unit that detects a sign of heat fainting based on the blood pressure estimated by the blood pressure estimation unit and the heart rate estimated by the heart rate estimation unit.
A predictive detection device for heat fainting including a notification unit that performs a predetermined notification based on the detection result of the detection unit. Acquires the heart radio waves of the human body,
Acquire the pulse wave of the human body and
Estimate blood pressure based on the acquired cardiac radio waves and pulse waves,
Estimate the heart rate based on the acquired pulse wave and
Detecting signs of heat fainting in the human body based on the estimated blood pressure and estimated heart rate,
A method for detecting a sign of heat fainting, which gives a predetermined notification based on the detection result of the sign. It estimates blood pressure based on the heart radio waves and pulse waves of the human body, and communicates with a wearable device that estimates the heart rate based on the pulse waves.
Obtaining the blood pressure and heart rate of the human body from the wearable device,
Based on the acquired blood pressure and the acquired heart rate, the sign of heat fainting in the human body is detected.
A computer program that causes a computer to execute a process of performing a predetermined notification based on the detection result of the sign. On the computer
When the blood pressure and heart rate of the human body are input, the presence / absence information is acquired by inputting the blood pressure and heart rate acquired from the wearable device into the learning model that outputs the presence / absence information of the detection of the sign of heat fainting.
The computer program according to claim 12, wherein a process for detecting a sign of heat fainting is executed based on the acquired presence / absence information.

Images