Biological information estimation method, biological information estimation program, biological information estimation device, and biological information estimation system

Abstract

To provide a biological information estimation method, a biological information estimation program, and a biological information estimation device, and a biological information estimation system capable of simply generating an accurate template.SOLUTION: The biological information estimation method for estimating the heart rate of a subject as biological information of the subject comprises acquiring a subject signal including information on the heart rate of the subject, generating a plurality of provisional templates by setting at least one parameter to a plurality of values, calculating the coincidence degree between each of the provisional templates and the subject signal, and estimating the heart rate of the subject from the subject signal by using a template when the coincidence degree is maximized.SELECTED DRAWING: Figure 8

Description

[Technical Field]

[0001] The present disclosure relates to a biological information estimation method, a biological information estimation program, a biological information estimation device, and a biological information estimation system. [Background technology]

[0002] As described in Patent Document 1, a method is known in which a template generated from a signal identified as a heartbeat component is used to identify the peak of the heartbeat. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2022-31000 Summary of the Invention [Problem to be solved by the invention]

[0004] A signal that detects cardiac motion may contain noise due to body movement, etc. To enable highly accurate detection of heartbeats from a signal that contains noise, it is necessary to easily generate a highly accurate template.

[0005] The present disclosure has been made in consideration of the above-mentioned points, and aims to provide a biometric information estimation method, a biometric information estimation program, a biometric information estimation device, and a biometric information estimation system that can easily generate highly accurate templates. [Means for solving the problem]

[0006] A biological information estimation method (1) according to one embodiment of the present disclosure is a method for estimating a heart rate of a subject as biological information of the subject, the biological information estimation method including the steps of acquiring a subject signal including information related to the heart rate of the subject, generating a plurality of temporary templates by setting at least one parameter to a plurality of values, calculating a degree of match between each of the plurality of temporary templates and the subject signal, and estimating the heart rate of the subject from the subject signal using the template with the highest degree of match.

[0007] (2) In the step of generating the plurality of temporary templates in the biometric information estimating method described in (1) above, each of the K value and the R value may be changed to a plurality of values.

[0008] (3) The method for estimating biometric information described in (1) or (2) above may further include the steps of estimating the pulsation of blood vessels in a part of the subject other than the heart as the biometric information of the subject, calculating the time difference between the heartbeat of the subject and the pulsation of the blood vessels as a pulse wave propagation time, and estimating the blood pressure of the subject as the biometric information of the subject based on the pulse wave propagation time.

[0009] (4) In the step of estimating the vascular pulsation of the biological information estimation method described in (3) above, the vascular pulsation of the subject's thigh may be estimated.

[0010] (5) A biological information estimation program according to an embodiment of the present disclosure is a program for causing a processor to estimate a heart rate of a subject as biological information of the subject. The biological information estimation program causes the processor to execute the steps of acquiring a subject signal including information related to the heart rate of the subject, generating a plurality of temporary templates by setting at least one parameter to a plurality of values, calculating a degree of match between each of the plurality of temporary templates and the subject signal, and estimating the heart rate of the subject from the subject signal using the template with the highest degree of match.

[0011] A biological information estimation device (6) according to an embodiment of the present disclosure includes a communication unit that acquires a subject signal including information about the subject's heart rate, and a control unit that estimates the subject's heart rate as biological information of the subject. The control unit generates a plurality of temporary templates by setting at least one parameter to a plurality of values, calculates a degree of match between each of the plurality of temporary templates and the subject signal, and estimates the subject's heart rate from the subject signal using the template that maximizes the degree of match.

[0012] (7) A biological information estimation system according to one embodiment of the present disclosure includes the biological information estimation device described in (6) above and a measurement device that measures a subject signal including information about the subject's heart rate. [Effects of the Invention]

[0013] According to the biometric information estimation method, biometric information estimation program, biometric information estimation device, and biometric information estimation system according to the present disclosure, highly accurate templates can be generated easily. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a block diagram illustrating a configuration example of a biological information estimation system according to the present disclosure. [Figure 2] FIG. 1 is a diagram illustrating an example of a model that approximates the heart with a sphere. [Figure 3] 10 is a graph showing an example of a template in which the change in the radius of the heart when the heart repeatedly expands and contracts is approximated by an expansion triangular wave. [Figure 4] 1 is a flowchart illustrating an example of a procedure of a biological information estimation method according to the present disclosure. [Figure 5] 10 is a graph showing an example of a subject signal. [Figure 6] 10 is a graph showing an example of a waveform of a signal of a temporary template approximated by an expanded triangular wave. [Figure 7] 10 is a graph showing an example of a cross-correlation function between a subject signal and an expanded triangular wave. [Figure 8]10 is a heat map showing an example of the distribution of the cross-correlation function when the K value and R value of the template approximated by the expanded triangular wave are changed. [Figure 9] 10 is an example of a heatmap generated for each of the eight subjects. [Figure 10] FIG. 10 is a diagram showing an example of a model that approximates a blood vessel with a cylinder. [Figure 11] 10 is a graph showing an example of measurement results of pulse wave transit time. DETAILED DESCRIPTION OF THE INVENTION

[0015] (overview) Monitoring vital signs is useful for detecting medical problems. When heart rate is monitored as a vital sign, for example, electrocardiograms, photoplethysmography, blood pressure monitors, ultrasound diagnosis, radio frequency (RF) imaging, computer tomography (CT) imaging, or magnetic resonance imaging (MRI) imaging is used. Heart rate monitoring using these techniques can be performed in intensive medical examinations conducted by adequate facilities and specialized staff at hospitals, etc. However, there are cases where the condition of a patient cannot be identified simply by monitoring heart rate during a medical examination at a specific date and time.

[0016] Therefore, there is a demand for a method that can monitor heart rate not only during medical checkups but also on a daily basis at home. To monitor heart rate at home, for example, home electrocardiographs, heart rate monitors, blood pressure monitors, etc. are used. Home electrocardiographs can easily measure heart rate using a single electrode. However, measurement accuracy can be reduced due to insufficient test information obtained through measurement or noise generated by slight body movement. Home heart rate monitors can measure heart rate using the photoplethysmography method, by transmitting red or infrared light through the skin. However, infrared light present in the surrounding environment can become noise, which can reduce measurement accuracy. Home blood pressure monitors can measure heart rate by wrapping a cuff around the body and measuring blood pressure. However, there is a problem in that heart rate measurement results vary depending on how the cuff is wrapped.

[0017] Furthermore, the above-mentioned home electrocardiographs, heart rate monitors, and blood pressure monitors are worn on the body. The need to wear a device on the body reduces the convenience of measuring heart rate. To improve the convenience of measuring heart rate, it is desirable to be able to measure heart rate without contact.

[0018] Non-contact heart rate measurement techniques include RF, laser, and radar. The RF method irradiates high-energy radio frequency waves onto the body and detects phase changes in the reflected waves due to the heartbeat to measure the heart rate. However, the RF method requires close proximity to the body to detect the reflected waves. This makes it difficult to use the RF method as a home heart rate measurement technique. The laser or radar method measures the heart rate by detecting the movement of the body surface caused by the heartbeat as a change in the distance from a sensor to the body surface. Specifically, the laser or radar method irradiates laser light or radar waves toward the chest of a subject whose heart rate is to be measured, and analyzes the reflected waves from the subject to detect the movement of the subject's chest and measure the heart rate. The laser or radar method allows the movement of the subject's chest to be detected from a distance from the subject. Furthermore, the radar method uses electromagnetic waves that penetrate non-metallic materials such as clothing and walls, allowing the movement of the subject's chest to be detected from a position where the subject's chest is not directly visible.

[0019] The chest movement of the subject that can be detected by the laser method or radar method includes the movement of the subject himself / herself whose heart rate is being measured, i.e., the subject's body movement. Therefore, in the laser method or radar method, it is necessary to remove the influence of the subject's body movement in order to improve the accuracy of heart rate measurement based on the chest movement of the subject.

[0020] A template matching method can be considered as a method for removing the influence of the subject's body movement. The template matching method detects a heartbeat when a template corresponding to a heartbeat overlaps with a signal that detects the movement of the subject's body surface. To improve the accuracy of heartbeat measurement using the template matching method, it is necessary to generate a template corresponding to the heartbeat with high accuracy. For example, a template may be generated from a signal that detects the movement of the subject's body surface using a statistical model or machine learning. When generating a template using a statistical model or machine learning, a large amount of data regarding the movement of the subject's body surface must be measured in advance. Therefore, it is difficult to apply this method to rapid analysis such as streaming analysis.

[0021] Furthermore, signals detecting the movement of the subject's body surface vary depending on the subject's characteristics or condition. This makes it difficult to select an appropriate template, which in turn makes it difficult to improve the accuracy of heartbeat detection using the template.

[0022] A biological information estimation system 1 (see FIG. 1) according to the present disclosure detects Doppler reflection waves using a Doppler sensor and analyzes them using a template method to measure the heartbeat of a subject in a non-contact manner. In the present disclosure, a template suitable for detecting the heartbeat from a signal obtained by detecting Doppler reflection waves is generated based on a simulation of Doppler reflection waves that is based on a mathematical model of heart movement. Generating the template based on the simulation eliminates the need to previously obtain data on Doppler reflection waves related to the movement of the subject's body surface.

[0023] Furthermore, by setting the parameters of the model used in the simulation to match the characteristics of the subject, it is possible to generate a template that matches the characteristics of the subject, and as a result, it is possible to obtain not only the heart rate but also information about the subject's characteristics as biological information of the subject.

[0024] The biological information estimation system 1 can also measure the subject's pulse wave transit time by measuring the subject's heart rate at multiple parts of the subject, and estimate the subject's blood pressure as biological information of the subject.

[0025] An example of the configuration and operation of the biological information estimation system 1 will be described in detail below.

[0026] (Configuration example of biological information estimation system 1) 1, a biological information estimation system 1 according to the present disclosure includes a biological information estimation device 10 and a measurement device 20. The biological information estimation system 1 acquires a signal including biological information of a subject 40 using the measurement device 20. The signal including the biological information of the subject 40 is also referred to as a subject signal. The biological information estimation device 10 estimates the biological information of the subject 40 based on the subject signal.

[0027] The measuring device 20 is a Doppler radar and includes a transmitter Tx and a receiver Rx for measuring electromagnetic waves. The transmitter Tx and receiver Rx may include antennas. The transmitter Tx transmits the electromagnetic waves for measurement toward the subject 40. The receiver Rx receives the electromagnetic waves reflected by the body surface of the subject 40 and returning. The measuring device 20 generates a subject signal based on the transmitted and received electromagnetic waves. The subject signal is a signal obtained by detecting Doppler reflected waves that reflect the movement of the body surface of the subject 40. The signal obtained by detecting Doppler reflected waves includes an I signal that is in phase with the transmitted electromagnetic waves and a Q signal that is quadrature-phase with the transmitted electromagnetic waves. The receiver Rx can separately detect the I signal and the Q signal by detecting the reflected waves using an IQ detector. In the present disclosure, it is assumed that the I signal is acquired as the subject signal. The subject signal may be a sampled signal output from a Doppler radar.

[0028] The biological information estimation device 10 includes a communication unit 11, a control unit 12, and an output unit 13.

[0029] The communication unit 11 acquires a subject signal from the measurement device 20 and outputs it to the control unit 12. The communication unit 11 may acquire control information for the measurement device 20 from the control unit 12 and output it to the measurement device 20. The communication unit 11 may include a communication module configured to be able to communicate with the measurement device 20 via wired or wireless communication. The communication module may be compatible with mobile communication standards such as 4G (4th Generation) or 5G (5th Generation). The communication module may be compatible with communication standards such as LAN (Local Area Network). The communication module may be compatible with wired or wireless communication standards. The communication module is not limited to these and may be compatible with various communication standards. The communication unit 11 may be configured to be connectable to the communication module.

[0030] The control unit 12 acquires the subject signal from the communication unit 11 and estimates biological information such as the heart rate or blood pressure of the subject 40 based on the subject signal. The control unit 12 may cause the output unit 13 to output the estimated results of the biological information of the subject 40. The control unit 12 may output control information for the measurement device 20 from the communication unit 11 to the measurement device 20.

[0031] The control unit 12 may be configured to include one or more processors. The processor may be configured to include a CPU (central processing unit) or a GPU (graphics processing unit). The control unit 12 may be configured to include a programmable circuit such as an FPGA (field-programmable gate array) or a dedicated circuit such as an ASIC (application specific integrated circuit). The control unit 12 may be configured to include any combination of these.

[0032] The biometric information estimation device 10 may further include a memory unit. The memory unit may include, for example, a semiconductor memory, a magnetic memory, an optical memory, or the like, but is not limited to these. The memory unit may function as, for example, a main memory unit, an auxiliary memory unit, or a cache memory. The memory unit may include an electromagnetic storage medium such as a magnetic disk. The memory unit may include a non-transitory computer-readable medium. The memory unit stores any information or program used in the operation of the biometric information estimation device 10. The memory unit may store, for example, a system program or an application program. The memory unit may be included in the control unit 12, or may be configured separately from the control unit 12.

[0033] The output unit 13 outputs information, data, etc. to the user. The output unit 13 may include, for example, a display device that outputs visual information such as images, characters, or graphics. The display device may include, for example, an LCD (Liquid Crystal Display), an organic EL (Electro-Luminescence) display, an inorganic EL display, or a PDP (Plasma Display Panel). The display device is not limited to these displays and may include displays of various other types. The display device may include a light-emitting device such as an LED (Light Emitting Diode) or an LD (Laser Diode). The display device may include various other devices. The output unit 13 may include, for example, an audio output device such as a speaker that outputs auditory information such as sound. The output unit 13 is not limited to these examples and may include various other devices.

[0034] The bioinformation estimation device 10 may further include an input unit that accepts input of information, data, etc. from a user. The input unit may include, for example, a touch panel or a touch sensor, or a pointing device such as a mouse. The input unit may also include physical keys. The input unit may also include a voice input device such as a microphone.

[0035] The biological information estimation device 10 may be configured as a PC such as a notebook PC (Personal Computer) or a tablet PC. The biological information estimation device 10 may be configured as a mobile terminal such as a smartphone or a tablet. The biological information estimation device 10 may be configured to include one or more server devices that can communicate with each other. The biological information estimation device 10 is not limited to these examples and may be configured in various other forms.

[0036] The biological information estimation device 10 may be realized using a cloud service or in an on-premise environment. The biological information estimation device 10 may include a part that executes information processing and a part that provides a user interface as separate pieces of hardware.

[0037] (Example of operation of biological information estimation system 1) In the biological information estimation system 1, the measurement device 20 generates a subject signal by detecting chest movement corresponding to the heartbeat of the subject 40, and outputs the subject signal to the biological information estimation device 10. The biological information estimation device 10 acquires the subject signal with the communication unit 11, and estimates the heartbeat of the subject 40 based on the subject signal with the control unit 12. An example of the operation of the biological information estimation system 1 to generate a subject signal and estimate the heartbeat based on the subject signal will be described below.

[0038] <Generation of patient signals> In this operation example, in order to generate subject signals that detected chest movement for each of multiple subjects 40, the measuring device 20 was fixed on a tripod at a position 50 cm away from the subject 40 so as to be aimed at the underside of the sternum of the subject 40. The chest movement of the subject 40 measured by the measuring device 20 was measured for one minute while the subject 40 was sitting deeply in a chair and breathing.

[0039] The chest movement of the subject 40 may be measured with the subject 40 sitting shallowly in a chair, or may be measured with the subject 40 lying down. The posture of the subject 40 when measuring the chest movement of the subject 40 is not limited to these examples, and the subject 40 may be in various other postures that keep the subject 40 stable.

[0040] The chest movement of the subject 40 may be measured while the subject 40 is holding his / her breath, in which case the influence of respiratory movement is reduced.

[0041] <Estimation of heart rate of subject 40 based on subject signal> The communication unit 11 of the biological information estimation device 10 acquires the subject signal generated by performing the set of measurements described above from the measurement device 20. The control unit 12 of the biological information estimation device 10 calculates the movement of the chest of the subject 40 based on the subject signal.

[0042] <<Relationship between subject signal and movement of subject 40>> The measuring device 20 can calculate the change in distance from the measuring device 20 to the subject 40 based on the change in phase between the electromagnetic wave transmitted to the subject 40 and the wave reflected from the subject 40. When the measuring device 20 is a microwave Doppler sensor, the round-trip time Tr from transmitting the microwave to receiving it is expressed by the following equation (1): In equation (1), t is time, c is the speed of light, D(t) is the distance from the measuring device 20 to the subject 40 at time t, l0 is the distance from the measuring device 20 to the subject 40 when time t is 0, and v(t) is the velocity of the subject 40 at time t.

[0043]

number

[0044] The reflected wave from the subject 40 is received with a delay of the round-trip time Tr after the electromagnetic wave is transmitted. The delay causes a phase shift of φ(t) expressed by the following equation (2): In equation (2), f0 is the frequency of the electromagnetic wave transmitted from the measurement device 20.

[0045]

number

[0046] The measuring device 20 detects the reflected wave by IQ detection and attenuates the high frequency components with a low pass filter, thereby generating an I signal represented as I(t) in the following equation (3a) and a Q signal represented as Q(t) in the following equation (3b). In equations (3a) and (3b), A s is the amplitude of the transmitted wave, and A r is the amplitude of the received wave, i.e., the reflected wave.

[0047]

number

[0048] According to the above equations (2), (3a), and (3b), it can be seen that the signal obtained by detecting the Doppler reflected wave depends on D(t). That is, D(t) is calculated based on the subject signal, which is the signal obtained by detecting the Doppler reflected wave. When D(t) is the distance from the measuring device 20 to the chest of the subject 40, D(t) changes depending on the heart rate of the subject 40.

[0049] <<Heart model for estimating heart rate>> In the present disclosure, in order to estimate the heart rate of the subject 40 from D(t), a model in which the heart is approximated by a sphere is used, as shown in Fig. 2. In the model in Fig. 2, R represents the radius of the heart, d represents the distance from the measurement device 20 to the center of the heart, and L xy (R, d) represents the distance from the measuring device 20 to each point on the surface of the heart.

[0050] Assuming that the distance from the measuring device 20 to the surface of the heart is equal to D(t), D(t) is expressed by the following equation (4) by surface integrating the distance between the point (x, y, 0) on the xy plane and the measuring device 20.

[0051]

number

[0052] Here, the relationship of the following formula (5) holds between the point (x, y) on the spherical surface and the radius R.

[0053]

number

[0054] By substituting equation (5) into equation (4) and expanding it, the following equation (6) is obtained.

[0055]

number

[0056] By substituting equation (6) into equation (2), φ(t) representing the phase change is calculated.

[0057] The amplitude of the Doppler reflection depends on the size of the object from which the transmitted electromagnetic wave is reflected. To express the dependence of the amplitude on the size of the object, the solid angle ω is introduced into the above equations (3a) and (3b) to determine whether the I signal is I * (t) is expressed by the following equation (7a), and the Q signal is Q * (t) is expressed by the following equation (7b).

[0058]

number

[0059] In a model in which the heart is approximated by a sphere, heartbeats are represented as changes in the radius of the heart. In the present disclosure, changes in the radius of the heart are represented as an diastolic triangular wave, as illustrated in the graph of FIG. 3. The horizontal axis of the graph in FIG. 3 represents time, and the vertical axis represents the radius of the heart. The diastolic triangular wave includes a triangular waveform corresponding to a period in which the radius of the heart changes and a flat waveform corresponding to a period in which the radius of the heart does not change. The triangular waveform represents the movement in which the radius of the heart increases as the heart transitions from systole to diastole, and the movement in which the radius of the heart decreases as the heart transitions from diastole to systole. The flat waveform represents the period in which the radius of the heart does not change until the next diastole. When comparing the diastolic triangular wave with an electrocardiogram waveform, the end of the T wave in the electrocardiogram waveform corresponds to the end of the triangular waveform in the diastolic triangular wave, i.e., the start of the flat waveform. Furthermore, the peak of the R wave in the electrocardiogram waveform corresponds to the peak of the triangular waveform in the diastolic triangular wave.

[0060] The difference between the radius of the heart during diastole and the radius during systole is represented by Δr. In the present disclosure, Δr is set to the same value of 1 cm for each of multiple subjects 40. On the other hand, the value R representing the radius of the heart during systole is set to a different value for each subject 40 as a parameter of the model.

[0061] The period of the dilated triangular wave is represented by T. The length of the triangular waveform of the dilated triangular wave, that is, the length of the period during which the radius of the heart changes, is T. ds It is expressed as T for T ds The ratio (T ds / T) is defined as the K value. The K value is a parameter that represents the movement of the heart in the model and is set to a different value for each subject 40. When the K value is 1, the change in the radius of the heart is represented as a triangular wave. When the K value is 0, the heart is not beating. Therefore, the K value is set to a value greater than 0 and less than or equal to 1.

[0062] <<Estimation of the heart rate of a subject 40 by analyzing the subject's signal using a model>> The control unit 12 of the biological information estimation device 10 analyzes the subject signal using the above-described heart model and estimates the heart rate of the subject 40. To estimate the heart rate of the subject 40, the control unit 12 may execute a biological information estimation method including, for example, the steps of the flowchart shown in FIG. 4. The biological information estimation method may be realized as a biological information estimation program executed by the control unit 12. The biological information estimation program may be stored in a non-transitory computer-readable medium.

[0063] The control unit 12 acquires a subject signal (step S1). Specifically, the control unit 12 acquires, from the communication unit 11, the subject signal that the communication unit 11 acquired from the measurement device 20. In the present disclosure, the subject signal is assumed to be an I signal. The subject signal may be a Q signal, or may be both an I signal and a Q signal.

[0064] The control unit 12 filters the subject signal (step S2). Specifically, the control unit 12 filters the subject signal using a band-pass filter. The band-pass filter is configured to pass frequency components in the range of 0.8 Hz or more and 5 Hz or less, and to attenuate frequency components in the range of less than 0.8 Hz or more than 5 Hz. By attenuating frequency components in the range of less than 0.8 Hz or more than 5 Hz using the band-pass filter, noise caused by the breathing and body movement of the subject 40 is reduced from the subject signal. Furthermore, by passing frequency components in the range of 0.8 Hz or more and 5 Hz or less using the band-pass filter, signal components corresponding to the subject's heart rate remain in the subject signal. Figure 5 shows an example of the subject signal after filtering using the band-pass filter. The horizontal axis of the graph in Figure 5 represents time, and the vertical axis represents the voltage of the subject signal.

[0065] 4, the control unit 12 estimates the heartbeat period of the subject 40 based on the subject signal after filtering (step S3). Specifically, the control unit 12 may estimate, as the heartbeat period of the subject 40, the inverse of the frequency of the maximum component of the FFT spectrum obtained by performing an FFT (Fast Fourier Transform) analysis on the subject signal.

[0066] The control unit 12 generates a temporary template representing the cardiac motion (step S4). The control unit 12 generates, as the temporary template, a signal simulating the subject signal detected by the measurement device 20 during one cycle of the heartbeat estimated in step S3. Specifically, the control unit 12 generates multiple temporary templates corresponding to each of the multiple cardiac motion patterns by applying each of the multiple cardiac motion patterns to a model and performing a simulation. The cardiac motion is represented by an expanding triangular wave specified by a combination of an R value representing the radius of the heart and a K value representing the cardiac motion. In other words, the cardiac motion is represented by multiple patterns corresponding to multiple combinations of changed R and K values.

[0067] As shown in Figure 3, the expansion triangle wave is the length of one cycle of the heartbeat, T, and the length of the period during which the radius of the heart changes, T. ds and the change in the radius of the heart Δr. The length of one cycle of the heartbeat estimated in step S3 is used as the length T of one cycle of the heartbeat. The change in the radius of the heart Δr was set to 1 cm. The K value was set in increments of 0.01 in the range of 0.1 to 1.0. The length T of one cycle of the heartbeat is constant. Therefore, the length T of the period during which the radius of the heart changes is determined according to the setting of the K value. ds is calculated and applied to the model. The R value was set in increments of 0.2 cm in the range of 4 cm or more and 7 cm or less. The range of the K value or R value is not limited to these examples and may be set to other ranges as appropriate. Furthermore, the increments of the K value or R value are not limited to these examples and may be set to other values ​​as appropriate.

[0068] In the above example, multiple temporary templates are generated by running a simulation while varying two parameters of an expansion triangular wave representing cardiac motion to multiple values. The number of parameters varied to generate multiple temporary templates may be one or three or more. That is, the control unit 12 generates multiple temporary templates by setting at least one parameter to multiple values. Only one of the K value or the R value may be used as a parameter that is set to multiple values. In addition to the K value and the R value, the Δr value may be used as a parameter that is set to multiple values. An example of the waveform of a temporary template signal is shown in FIG. 6. The horizontal axis of the graph in FIG. 6 represents time, and the vertical axis represents voltage.

[0069] When estimating a heart rate from a subject signal using a template, it is necessary to generate a template that closely matches the subject signal. The control unit 12 calculates the degree of match between each of a plurality of temporary templates and the subject signal, and generates a template with the highest degree of match. In the present disclosure, the degree of match between the temporary template and the subject signal is measured using a cross-correlation coefficient between the temporary template and the subject signal. The degree of match between the temporary template and the subject signal is not limited to the cross-correlation coefficient, and may be other indices.

[0070] The control unit 12 calculates the cross-correlation coefficient between the temporary template and the subject signal (step S5). The subject signal is the subject signal after filtering. The control unit 12 sets a search window for one period of the expanded triangular wave representing the temporary template, moves the temporary template in the time direction at the sampling interval of the subject signal, and calculates the cross-correlation coefficient between the temporary template and the subject signal at each time point. In the time plot of the cross-correlation coefficient illustrated in FIG. 7, the value of the cross-correlation coefficient changes depending on the time at which the temporary template is placed. The horizontal axis of the graph in FIG. 7 represents the time corresponding to the position of the temporary template, and the vertical axis represents the cross-correlation coefficient between the temporary template and the subject signal.

[0071] The control unit 12 generates a template to be used for estimating the heartbeat of the subject 40 from the subject signal (step S6). As described above, the control unit 12 generates a template to maximize the degree of match with the subject signal. Specifically, the control unit 12 generates a template to maximize the median cross-correlation coefficient as the degree of match. The median cross-correlation coefficient is the median of the distribution of peak values ​​of the cross-correlation coefficient. The control unit 12 extracts the peak values ​​of the cross-correlation coefficients from the time plot of the cross-correlation coefficients of each temporary template calculated in step S5, and calculates the median of the extracted multiple peak values ​​as the median cross-correlation coefficient. A larger median cross-correlation coefficient indicates a closer match of the temporary template to the waveform corresponding to the heartbeat of the subject 40 contained in the subject signal. The control unit 12 may determine, among the median cross-correlation coefficients calculated for each temporary template, the temporary template for which the median cross-correlation coefficient is the largest, as the template to be used for estimating the heartbeat of the subject 40 from the subject signal.

[0072] The temporary template is identified by a combination of an R value and a K value. Therefore, the median cross-correlation coefficient calculated for the temporary template is associated with the combination of the R value and the K value. The control unit 12 may determine the combination of the R value and the K value that maximizes the median cross-correlation coefficient based on the correspondence between the median cross-correlation coefficient calculated for each temporary template and the combination of the R value and the K value. The control unit 12 may generate the template identified by the determined combination of the R value and the K value as a template to be used for estimating the heart rate of the subject 40 from the subject signal.

[0073] The control unit 12 may generate a heat map of median cross-correlation coefficients corresponding to each combination of R and K values, as illustrated in FIG. 8 . The heat map is a two-dimensional map in which the horizontal axis corresponds to the K value, the vertical axis corresponds to the R value, and the color of points corresponding to combinations of R and K values ​​is represented by a grayscale corresponding to the median cross-correlation coefficient. The control unit 12 may determine the point in the heat map where the median cross-correlation coefficient is the maximum, and generate a template identified by the combination of R and K values ​​corresponding to that point as a template to be used for estimating the heart rate of the subject 40 from the subject signal. In the heat map of FIG. 8 , the point where the median cross-correlation coefficient is the maximum is represented by a solid circle (●). The maximum value of the median cross-correlation coefficient (CCF_med) in the heat map of FIG. 8 was 0.868.

[0074] In a heat map, the larger the K value, the larger the median cross-correlation coefficient. Furthermore, the median cross-correlation coefficient tends to change periodically with changes in the R value. The reason the median cross-correlation coefficient changes periodically with changes in the R value is because the intensity of the Doppler reflected wave is affected by both the distance from the measurement device 20 to the subject 40 and the area from which the measurement electromagnetic wave is reflected. For example, when the R value is large, the round trip time decreases and the phase change of the received wave decreases. The decrease in the phase change of the received wave decreases the intensity of the Doppler reflected wave. On the other hand, when the R value is large, the area from which the measurement electromagnetic wave is reflected increases. An increase in the reflection area increases the intensity of the Doppler reflected wave. The model used in the present disclosure can appropriately determine a template using an evaluation criterion that reflects the change in the intensity of the Doppler reflected wave due to changes in the R value.

[0075] The control unit 12 may cause the output unit 13 to display the combination of the R value and the K value when the median cross-correlation coefficient is at its maximum value. The control unit 12 may cause the output unit 13 to display a heat map. The control unit 12 may cause the output unit 13 to display a time plot of the cross-correlation coefficient.

[0076] The control unit 12 calculates a cross-correlation coefficient between the template generated in step S6 and the subject signal (step S7). The cross-correlation coefficient between the template and the subject signal is represented as a time plot of the cross-correlation coefficient, for example, as shown in FIG.

[0077] The control unit 12 calculates the heart rate of the subject 40 based on the cross-correlation coefficient between the template and the subject signal (step S8). Specifically, the control unit 12 calculates the time interval at which the cross-correlation coefficient reaches its peak value in the time plot of the cross-correlation coefficient shown in FIG. 7, for example. The control unit 12 calculates the heart rate of the subject 40 based on the time interval at which the cross-correlation coefficient reaches its peak value. The control unit 12 may calculate the reciprocal of the interval between the peak times of the cross-correlation coefficient as the time change in the heart rate of the subject 40. The control unit 12 may calculate the number of peaks of the cross-correlation coefficient included in the one-minute plot of the cross-correlation coefficient as the one-minute heart rate of the subject 40.

[0078] The control unit 12 may cause the output unit 13 to display the calculation result of the heart rate of the subject 40 in the form of a numerical value, an image, or the like. The control unit 12 may cause the output unit 13 to output the calculation result of the heart rate of the subject 40 in various forms, such as sound. After executing the procedure of step S8, the control unit 12 ends execution of the procedure of the flowchart in FIG. 4.

[0079] <<Example of heart rate estimation>> The heart rates of eight subjects 40 were estimated based on the measurement results of Doppler reflected waves. Table 1 shows the maximum median cross-correlation coefficient calculated for each subject 40, the K value and R value at which the median cross-correlation coefficient reached its maximum value, and the age, sex, height, weight, and HRA (Heart Rate Average) of each subject 40. In this example, the measurement device 20 measured the subject signals twice, with an interval of at least one minute. The maximum median cross-correlation coefficient and the K value and R value at which the median cross-correlation coefficient reached its maximum value listed in Table 1 include results calculated based on each of the two measurement results. FIG. 9 also shows a heat map of the median cross-correlation coefficient calculated for each subject 40 based on the first measurement result. In the heat map of FIG. 9, the point at which the median cross-correlation coefficient reached its maximum value is plotted as a solid circle (●). The median cross-correlation coefficient (CCF_med) at that point, i.e., a numerical value representing the maximum median cross-correlation coefficient, is also listed. In Table 1 and FIG. 9, the eight subjects 40 are distinguished by the symbols A to H.

[0080] [Table 1]

[0081] The maximum median cross-correlation coefficient was 0.8 or higher for all eight subjects 40. This indicates a high correlation between the template generated by the method of the present disclosure and the subject signal. The K values ​​ranged from 0.44 to 1.00 for the eight subjects 40. The R values ​​ranged from 0.42 cm to 0.70 cm for the eight subjects 40. The large distribution of R values, which represent the radius of the heart, is likely due to the fixed Δr value, which represents cardiac motion, set to 1.0 cm. The eight subjects 40 were Japanese men and women. Given that the radius of the heart for Japanese men and women is approximately 5 cm, the distribution of R values ​​is considered to be reasonable.

[0082] Furthermore, when the heart rates estimated using the method disclosed herein were compared with the pulse rates measured using a blood pressure monitor, it was confirmed that the distribution of heart rates estimated using the method disclosed herein was in good agreement with the pulse rates measured using a blood pressure monitor.

[0083] <<Summary of Heart Rate Estimation>> As described above, in the biometric information estimation system 1 according to the present disclosure, the biometric information estimation device 10 generates a temporary template based on a simulation using a model, calculates the degree of match between the subject signal and the temporary template, and generates a template with the highest degree of match.

[0084] As a comparative example, it is possible to generate a template by analyzing the subject signals in advance. According to the comparative example, it is necessary to acquire a large number of subject signals in advance to generate a template. Furthermore, it is necessary to prepare a template for each subject 40 in advance.

[0085] On the other hand, according to the method of the present disclosure, a template for estimating the heart rate of the subject 40 from the subject signal is generated by simulation in accordance with the subject signal. In other words, there is no need to prepare a template in advance. As a result, a template for estimating the heart rate of the subject 40 can be generated easily.

[0086] Furthermore, according to the technique of the present disclosure, by generating a template that maximizes the degree of match with the subject signal, the accuracy of estimating the heart rate of the subject 40 is improved. In other words, a highly accurate template can be generated easily.

[0087] In the above-described embodiment, a signal obtained by detecting a Doppler reflected wave is used as the subject signal. However, the signal used as the subject signal is not limited to a signal obtained by detecting a Doppler reflected wave, and may be any of various signals that include information about the cardiac movement of the subject 40.

[0088] <Estimation of blood pressure of subject 40 based on subject signal> Human blood pressure may be correlated with pulse wave transit time. The biological information estimation device 10 may measure the pulse wave transit time of the subject 40 and estimate the blood pressure of the subject 40 based on the measurement results of the pulse wave transit time. The pulse wave transit time of the subject 40 is calculated from pulsations at multiple locations on the subject 40. In the blood pressure estimation embodiment described below, the biological information estimation device 10 measures the pulse wave transit time of the subject 40 based on the measurement results of the pulsations of the heart, i.e., the heart rate, of the subject 40, and the measurement results of the pulsations of the blood vessels in the femoral region, i.e., the pulse rate. The biological information estimation device 10 also estimates the blood pressure of the subject 40 based on the measurement results of the pulse wave transit time of the subject 40.

[0089] The measurement device 20 may include a sensor that generates a subject signal that detects the heartbeat of the subject 40, and a sensor that generates a subject signal that detects the pulsation of the blood vessels in the thighs of the subject 40, i.e., the thigh pulse. Hereinafter, the subject signal that detects the heartbeat will also be referred to as a heartbeat signal. The subject signal that detects the thigh pulse will also be referred to as a pulse signal. The measurement device 20 may be configured to measure both the heartbeat signal and the pulse signal using one sensor.

[0090] The communication unit 11 of the biological information estimation device 10 acquires a heartbeat signal and a pulse signal from the measurement device 20. The control unit 12 of the biological information estimation device 10 estimates the heartbeat of the subject 40 based on the heartbeat signal. As described above, the heartbeat of the subject 40 may be estimated by a method using a template based on a model that approximates the heart as a sphere.

[0091] The control unit 12 estimates the femoral pulse of the subject 40 based on the pulse signal. In the present disclosure, a model in which the blood vessels in the femoral region are approximated by a cylinder is used to estimate the femoral pulse, as shown in FIG. 10. In the model in FIG. 10, R represents the radius of the blood vessel, h represents the length of the blood vessel included in the measurement range, d represents the distance from the measurement device 20 to the central axis of the blood vessel, and L(φ, y) represents the distance from the measurement device 20 to each point on the surface of the blood vessel. Ix represents the distance from the measurement device 20 to the intersection of the circumferential tangent of the blood vessel and the x-axis at each point on the surface of the blood vessel.

[0092] Assuming that the distance from the measuring device 20 to the surface of the blood vessel is equal to D(t), which is the distance calculated from the pulse signal, D(t) can be expressed by the following equation (8) by surface integrating the distance between each point on the surface of the blood vessel and the measuring device 20. In equation (8), A=d 2 +R 2 and B=-2dR.

[0093]

number

[0094] The control unit 12 calculates φ(t), which represents the phase change, by substituting equation (8) into the above-described equation (2), and calculates the change in the radius of the blood vessel in the thigh of the subject 40 based on the calculation result of φ(t). In a model in which blood vessels are approximated by cylinders, the pulse is expressed as a change in the radius of the blood vessel. In the present disclosure, the change in the radius of the blood vessel is expressed as an expanding triangular wave, similar to the change in the radius of the heart. As in estimating the heartbeat, the control unit 12 generates a provisional template by combining various values ​​of the R value corresponding to the radius of the blood vessel and the K value representing the movement of the blood vessel, and generates a template based on the median cross-correlation coefficient. The control unit 12 calculates the cross-correlation coefficient between the generated template and the pulse signal, and estimates the pulse in the thigh of the subject 40.

[0095] The control unit 12 calculates the pulse wave transit time based on the estimated heart rate and the estimated thigh pulse of the subject 40. Specifically, the control unit 12 may calculate, as the pulse wave transit time, the difference between the time when the cross-correlation coefficient between the heart rate signal of the subject 40 and the template reaches its peak and the time when the cross-correlation coefficient between the pulse signal of the subject 40 and the template reaches its peak.

[0096] The control unit 12 estimates the blood pressure of the subject 40 from the calculation result of the pulse wave transit time based on the correlation between the pulse wave transit time and the blood pressure of the subject 40. For example, it is generally known that there is an inversely proportional or negative correlation between the pulse wave transit time and the blood pressure. The control unit 12 may estimate the blood pressure of the subject 40 based on the general correlation between the pulse wave transit time and the blood pressure.

[0097] The control unit 12 may acquire in advance the correlation between the pulse wave transit time and blood pressure of the subject 40. For example, the correlation between the pulse wave transit time and blood pressure of the subject 40 may be acquired by simultaneously measuring the pulse wave transit time and blood pressure of the subject 40 and associating the pulse wave transit time with the blood pressure.

[0098] The control unit 12 may continuously measure the change over time in the pulse wave transit time of the subject 40, as exemplified in Fig. 11. In the graph of Fig. 11, the horizontal axis represents time, and the vertical axis represents the pulse wave transit time of the subject 40. The control unit 12 may estimate the change over time in the blood pressure of the subject 40 based on the change over time in the pulse wave transit time of the subject 40.

[0099] The control unit 12 may cause the output unit 13 to display the estimated result of the blood pressure of the subject 40 in the form of a numerical value, an image, or the like. The control unit 12 may cause the output unit 13 to output the estimated result of the blood pressure of the subject 40 in various forms, such as sound.

[0100] <<Summary of Blood Pressure Estimation>> As described above, the biological information estimation system 1 according to this embodiment can calculate the pulse wave transit time of the subject 40 from the pulsations at multiple locations on the subject 40, and estimate the blood pressure of the subject 40. The pulse wave transit time of the subject 40 may be calculated based on the heart rate and thigh pulse of the subject 40. The thighs of the subject 40 hardly move when the subject 40 is seated in a chair, a car seat, or the like. Therefore, the calculation result of the thigh pulse is less affected by the body movement of the subject 40. By calculating the pulse wave transit time based on the calculation result of the thigh pulse, the calculation accuracy of the pulse wave transit time is improved. The improvement in the calculation accuracy of the pulse wave transit time improves the estimation accuracy of the blood pressure based on the pulse wave transit time.

[0101] The signal used as the subject signal for calculating the pulse wave transit time is not limited to a signal obtained by detecting a Doppler reflected wave, but may be any of a variety of signals containing information about the movement of the heart or blood vessels of the subject 40.

[0102] (summary) As described above, the biological information estimation system 1, the biological information estimation device 10, and the biological information estimation method according to the present embodiment enable easy generation of a highly accurate template for estimating the heart rate of the subject 40 from the subject signal. Furthermore, the accuracy of estimating the heart rate of the subject 40 using the generated template is improved.

[0103] Furthermore, according to the biological information estimation system 1, the biological information estimation device 10, and the biological information estimation method of the present embodiment, the pulse wave transit time of the subject 40 is easily calculated by measuring the pulsations of multiple parts of the subject 40. Then, the blood pressure of the subject 40 is easily estimated based on the pulse wave transit time of the subject 40.

[0104] (Other Examples) In the above-described embodiment, the point at which the degree of match is maximized is determined using a heat map representing the distribution of degrees of match for each subject 40. The distribution of degrees of match represented by the heat map itself may be considered to represent the characteristics of the subject 40. In other words, the heat map may be used to identify the subject 40. For example, the biometric information estimation device 10 may prepare a heat map corresponding to each registered individual who is to be identified as a reference map in advance, and identify the subject 40 as a registered individual based on a comparison between the heat map corresponding to the subject 40 and the reference map corresponding to each registered individual. Specifically, the biometric information estimation device 10 may calculate, for each combination of K value and R value, the difference between the degree of match in the heat map corresponding to the subject 40 and the degree of match in the reference map corresponding to each registered individual, and calculate the sum of the absolute values ​​of the differences in degree of match calculated for each combination. The biometric information estimation device 10 may determine the reference map when the calculated sum is minimum and identify the subject 40 as a registered individual corresponding to the determined reference map.

[0105] Furthermore, the combination of parameter values ​​such as the K value and the R value when the degree of match is maximized may be used to identify the subject 40. In this case, the biological information estimation device 10 may determine a reference map when the combination of parameter values ​​such as the K value and the R value when the degree of match is maximized in the reference map corresponding to each registrant is closest to the combination of parameter values ​​such as the K value and the R value when the degree of match is maximized in the heat map corresponding to the subject 40, and identify the subject 40 as a registrant corresponding to the determined reference map.

[0106] In the above-described embodiment, an example has been described in which a change in the radius of the heart is approximated by an expansion triangular wave and used as a template. The waveform used as a template is not limited to an expansion triangular wave. When the movement of the heart is viewed as a change in the radius of the heart over time, the period is divided into a period in which the radius of the heart changes due to the expansion and contraction of the heart and a period in which the radius of the heart does not change during that period. Therefore, the waveform used as a template may be a waveform that includes a period in which the radius of the heart changes and a period in which the radius of the heart does not change during that period.

[0107] Although the embodiments of the present disclosure have been described based on the drawings and examples, it should be noted that those skilled in the art can make various modifications or alterations based on the present disclosure. Therefore, it should be noted that these modifications or alterations are included in the scope of the present disclosure. For example, the functions included in each component can be rearranged so as not to cause logical inconsistencies, and multiple components can be combined or divided into one. [Explanation of symbols]

[0108] 1. Biometric information estimation system 10 Biometric information estimation device (11: communication unit, 12: control unit, 13: output unit) 20 Measuring Equipment 40 Subjects

Claims

1. A biological information estimation method for estimating a heart rate of a subject as biological information of the subject, comprising: acquiring a subject signal comprising information regarding the subject's heart rate; generating a plurality of provisional templates by setting at least one parameter to a plurality of values; calculating a degree of match between each of the plurality of temporary templates and the subject signal, and estimating the heart rate of the subject from the subject signal using the template with the highest degree of match; A method for estimating biological information,

2. The biometric information estimating method according to claim 1 , wherein in the step of generating a plurality of temporary templates, each of the K value and the R value is changed to a plurality of values.

3. estimating pulsation of blood vessels other than the heart of the subject as biological information of the subject; calculating a time difference between the subject's heartbeat and the pulsation of the blood vessel as a pulse wave transit time; estimating a blood pressure of the subject as biological information of the subject based on the pulse wave transit time; The biological information estimation method according to claim 1 or 2, further comprising:

4. The biological information estimation method according to claim 3 , wherein the step of estimating blood vessel pulsation estimates blood vessel pulsation in a thigh of the subject.

5. A biological information estimation program that causes a processor to estimate a heart rate of a subject as biological information of the subject, acquiring a subject signal comprising information regarding the subject's heart rate; generating a plurality of provisional templates by setting at least one parameter to a plurality of values; calculating a degree of match between each of the plurality of temporary templates and the subject signal, and estimating the heart rate of the subject from the subject signal using the template with the highest degree of match; A biometric information estimation program that causes the processor to execute the above.

6. a communication unit for acquiring a subject signal including information about the subject's heart rate; a control unit that estimates a heart rate of the subject as biological information of the subject; Equipped with The control unit generating a plurality of provisional templates by setting at least one parameter to a plurality of values; calculating a degree of match between each of the plurality of temporary templates and the subject signal, and estimating the heart rate of the subject from the subject signal using the template with the highest degree of match; Biometric information estimation device.

7. A biological information estimation system comprising: the biological information estimation device according to claim 6; and a measurement device that measures a subject signal containing information related to the heartbeat of the subject.

Images