Health monitoring device, computer program and recording medium

Abstract

Provided is a health monitoring device which can also be used in the medical field by extracting, in particular, an apex beat component from a trunk acoustic vibration signal.　This health monitoring device (1000) analyzes a trunk acoustic vibration signal, and estimates the health condition of a person using a correlation between an indicator relating to a left ventricular pressure waveform which indicates the behavior of the heart, and the vibration frequency of a frequency component derived from the apex beat, the time of diastole in the cardiac cycle, or blood pressure. The former is input information and the latter is output information of the heart through which blood circulates, and thus by comparing these two pieces of information, health conditions relating to the function of the heart can be known to a higher degree of accuracy than was conventionally possible.

Description

[Technical field]

[0001] The present invention relates to a health monitoring device that processes trunk acoustic signals propagating through the human trunk to estimate a person's health condition, a computer program and a recording medium, and a biosignal measuring device suitable for capturing the above-mentioned trunk acoustic vibration signals. [Background technology]

[0002] In Patent Documents 1 to 4, etc., the inventors have proposed a technology for unconstrainedly capturing vibrations (trunk acoustic vibration signals) caused by biosignals propagating through the surface of the back of a person's trunk, and analyzing the vibrations to estimate the person's condition.

[0003] In Patent Document 1, a predetermined time width is applied to the time series waveform of the back surface pulse wave around 1 Hz extracted from the vibration (trunk acoustic vibration signal) due to the biological signal propagating through the body surface, and a sliding calculation is performed to obtain a time series waveform of the frequency gradient, and the biological condition is estimated from the tendency of the change, for example, whether the amplitude tends to increase or decrease. It also discloses that the trunk acoustic vibration signal is frequency analyzed to obtain the power spectrum of each frequency corresponding to the function adjustment signal, fatigue acceptance signal, and activity adjustment signal belonging to a predetermined ULF band (extremely low frequency band) to VLF band (very low frequency band), and the human condition is judged from the time series change of each power spectrum.

[0004] Patent documents 2 and 3 disclose a means for determining the homeostatic function level. The means for determining the homeostatic function level uses at least one of the following: the positive and negative of the differential waveform of the frequency slope time series waveform, the positive and negative of the integral waveform obtained by integrating the frequency slope time series waveform, and the absolute value of each frequency slope time series waveform obtained by absolute value processing of the frequency slope time series waveform using the zero crossing method and the frequency slope time series waveform using the peak detection method. By combining these, it is determined which level the homeostatic function corresponds to. In addition, Patent document 4 discloses a sensing system (sound and vibration information collecting mechanism) having a resonance layer equipped with a natural oscillator including a natural frequency corresponding to a trunk acoustic vibration signal. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2011-167362 A [Patent Document 2] JP 2014-117425 A [Patent Document 3] JP 2014-223271 A [Patent Document 4] JP 2016-26516 A Summary of the Invention [Problem to be solved by the invention]

[0006] The technologies for estimating biological conditions described in Patent Documents 1 to 4 identify signals predicting sleepiness, estimate fatigue, etc., and have been proposed primarily for use in estimating the state of a car driver, preventing drowsy driving, and encouraging the driver to transition to an alert state.

[0007] The sensing system using the above-mentioned natural oscillator can easily and stress-free measure trunk acoustic pulse wave (hereinafter referred to as APW), which is a measurement index for knowing the health of the cardiocirculatory system. The APW, which is a coupling of the cardiocirculatory system structures and blood, has frequency characteristics given by the pressure fluctuations during the cardiac cycle and the heart rate. The feedback force of the flow velocity and pressure generated during the process from blood filling to blood ejection becomes the apex beat and heart sound. From the frequency characteristics of the blood-vessel coupling of the APW and the physical properties obtained from the apex beat and heart sound, it may be possible to estimate information on the integrated function of the cardiocirculatory system, combining information on the cardiocirculatory system and blood. In particular, the apex beat is a palpable vibration generated by the collision of the apex of the heart with the chest wall, and is more suitable for capture by a sensing system using natural oscillators than heart sounds. Therefore, extracting the apex beat component from the trunk acoustic vibration signal obtained from such a sensing system and judging the state of the person can analyze the state of the person with higher accuracy, and not only can it estimate the state of the driver of a car, but it can also be used to judge the health state (for example, It is expected that this system will also be used in the medical field, such as to determine whether a patient is healthy, has mild abnormalities and requires follow-up observation, has lifestyle-related diseases, or has heart disease or circulatory disease, based on the assessment classification of the Japan Society of Human Health Checkups (revised on April 1, 2018 and partially updated on December 14, 2018).

[0008] The present invention has been made in consideration of the above, and is directed to a method for detecting a torso acoustic vibration signal, in particular Frequency components derived from the apex beat (apex beat components) The present invention aims to provide a health monitoring device, a computer program, and a recording medium that can be used in the medical field by extracting the above-mentioned components. Another aim is to provide a biosignal measuring device that is more suitable for capturing the apex pulsation component than conventional sensing systems and is suitable for collecting trunk acoustic vibration signals for use in the health monitoring device. [Means for solving the problem]

[0009] In order to solve the above problems, the health monitoring device of the present invention comprises: A health monitoring device that processes a trunk acoustic signal measured from a person's trunk and estimates a health condition of the person, a left ventricular pressure waveform determining means for determining an index relating to a left ventricular pressure waveform, which indicates cardiac behavior, from the trunk acoustic signal; an apex pulsation component extraction means for extracting a frequency component derived from the apex pulsation from the trunk acoustic signal; an estimation means for estimating a health condition of a person based on a correlation between an index relating to the left ventricular pressure waveform and a vibration frequency of a frequency component derived from the apex beat; The present invention is characterized by having the following. The estimation means can estimate the health condition from a correlation between an index relating to the left ventricular pressure waveform and a time of diastolic phase in a cardiac cycle which indicates the magnitude of the vibration frequency of the apex pulsation component.

[0010] A health monitoring device of the present invention is a health monitoring device that processes a trunk acoustic signal measured from a person's trunk and estimates a health condition of the person, comprising: a left ventricular pressure waveform determining means for determining an index relating to a left ventricular pressure waveform, which indicates cardiac behavior, from the trunk acoustic signal; an apex pulsation component extraction means for extracting a frequency component derived from the apex pulsation from the trunk acoustic signal; an estimation means for estimating a person's health condition based on a correlation between the index relating to the left ventricular pressure waveform and blood pressure; The present invention is characterized by having the following.

[0011] The apex pulsation component extracting means is preferably configured to extract a frequency component derived from the apex pulsation from the torso acoustic signal in comparison with a heart sound component. The apex beat component extracting means can further extract a frequency component derived from the apex beat in consideration of a relationship with a heart rate. It is preferable that the trunk acoustic signal is a signal captured by a biosignal measuring device placed on the rear of the person's chest.

[0012] The computer program of the present invention comprises: A computer program for processing a trunk acoustic signal measured from a human trunk and causing a computer to function as a health monitoring device for estimating a health condition of the human, comprising: A step of obtaining an index relating to a left ventricular pressure waveform, which indicates a behavior of the heart, from the trunk acoustic signal; extracting a frequency component originating from an apex beat from the trunk acoustic signal; a step of estimating a health condition of a person based on a correlation between an index relating to the left ventricular pressure waveform and a vibration frequency of a frequency component derived from the apex beat; The present invention is characterized in that the computer is caused to execute the above steps. The procedure for estimating the person's health condition may be a procedure for estimating the health condition from a correlation between an index related to the left ventricular pressure waveform and a time of the diastolic phase in the cardiac cycle, which indicates the magnitude of the vibration frequency of the apex beat component.

[0013] The computer program of the present invention further comprises: A computer program for processing a trunk acoustic signal measured from a human trunk and causing a computer to function as a health monitoring device for estimating a health condition of the human, comprising: A step of obtaining an index relating to a left ventricular pressure waveform, which indicates a behavior of the heart, from the trunk acoustic signal; extracting a frequency component originating from an apex beat from the trunk acoustic signal; The method is characterized in that the computer is caused to execute a procedure of estimating a person's health condition from the index relating to the left ventricular pressure waveform and the correlation with blood pressure. The step of extracting the frequency component originating from the apex beat of the heart preferably causes the computer to execute a step of extracting the frequency component originating from the apex beat of the heart from the trunk acoustic signal in comparison with a heart sound component. The step of extracting the frequency component derived from the apex beat may further extract the frequency component derived from the apex beat in consideration of a relationship with a heart rate.

[0014] The present invention also provides a recording medium on which the above computer program is recorded.

[0015] In addition, the biological signal measuring device of the present invention is A bead foam molded into a plate shape; A three-dimensional knitted fabric is loaded into the arrangement holes formed in the bead foam. A film is attached to both sides of the bead foam so as to cover both sides of the three-dimensional knitted fabric; A microphone sensor is disposed on the outer side of the film in correspondence with a location where the three-dimensional knitted fabric is disposed; A cover film for covering the microphone sensor; a disturbance intrusion suppression member that surrounds the microphone sensor from the outside within the cover film and suppresses the intrusion of disturbances; The present invention is characterized by having the following.

[0016] The disturbance-mixing suppression member is preferably a gel. Effect of the Invention

[0017] The present invention estimates a person's health condition by analyzing a trunk acoustic vibration signal and using a correlation between an index related to a left ventricular pressure waveform indicating the behavior of the heart and the vibration frequency of the frequency component derived from the apical pulsation, the diastolic time in the cardiac cycle, or blood pressure. The former is input information for the heart through which blood circulates, and the latter is output information, so by comparing these two pieces of information, it is possible to know the health condition related to the function of the heart with higher accuracy than before. In addition, since the left ventricular pressure waveform can also be composed of frequency components derived from the apical pulsation, according to the present invention, the health condition can be easily estimated by determining the frequency components derived from the apical pulsation. Therefore, the present invention is suitable for use in the medical field, such as health checkups.

[0018] In addition, in the biosignal measuring device of the present invention, the microphone sensor is disposed outside the three-dimensional knitted fabric that serves as the natural oscillator, and is surrounded by a disturbance suppression member. Conventional sensing systems in which a microphone sensor is disposed in a hole in which a natural oscillator made of a three-dimensional knitted fabric is disposed capture signals amplified by the three-dimensional knitted fabric (natural oscillator), but the biosignal measuring device of the present invention captures signals propagated through the three-dimensional knitted fabric (natural oscillator). Therefore, a weak input vibration signal is captured with specific frequency components amplified by stochastic resonance during the solid propagation process. As a result, not only heart sounds but also apex pulsation components can be captured more reliably, and the device is suitable as a device for collecting trunk acoustic vibration signals in the health condition estimation device of the present invention that uses information on the apex pulsation components. [Brief description of the drawings]

[0019] [Figure 1] Fig. 1(a) is a plan view showing a biosignal measuring device (4SR) according to one embodiment of the present invention, Fig. 1(b) is a cross-sectional view taken along line AA in Fig. 1(a), and Fig. 1(c) is a cross-sectional view showing a conventional biosignal measuring device (3SR). [Diagram 2] Fig. 2(a) is a diagram showing an image of blood flow accompanying pulsation with arrows. Fig. 2(b) and (c) are diagrams showing the arrangement positions of the upper left biosignal detector (sensor L), upper right biosignal detector (sensor R), and lower biosignal detector (sensor M) together with the state of arterial pulsation. [Diagram 3] FIG. 3(a) is a diagram for explaining a method of performance testing of the biosignal measuring device (4SR) of the above embodiment and a conventional biosignal measuring device (3SR) of the comparative example, and FIG. 3(b) is a diagram showing the characteristics of the input signal of the speaker used in the performance test. [Figure 4] Figure 4(a) shows the power spectrum of the inner microphone in a biosignal measuring device (3SR) relating to the comparative example, and Figure 4(b) shows the power spectrum of the outer microphone in the biosignal measuring device (4SR) relating to the above embodiment. [Diagram 5]FIG. 5(a) is a diagram showing the SNR of the biological signal measuring device (4SR) according to the above embodiment, and FIG. 5(b) is a diagram showing the frequency components of the apex beat and heart sounds in logarithmic scale. [Figure 6] FIG. 6(a) is a linear display of the biosignals captured by the biosignal measuring device (4SR) of the above embodiment, and FIG. 6(b) is a log-log enlarged display of the phonocardiogram and each of the biosignals obtained by the biosignal measuring device (4SR) of the above embodiment. [Figure 7] FIG. 7 shows waveforms at each processing stage until the apex beat component is extracted from the sensor waveform of the biosignal measuring device (4SR) according to the above embodiment for three subjects in their twenties with different heart rates and blood pressure values. [Figure 8] FIG. 8 is a diagram showing a medical index and a frequency analysis result at each processing stage of a signal waveform measured by the vital sign measuring device (4SR) according to the above embodiment. [Figure 9] FIG. 9 shows time-series data of the electrocardiogram, phonocardiogram, fingertip plethysmogram, and apex pulsation component. [Figure 10] FIG. 10 is a diagram showing the process of determining the frequency bands of the heart sounds and apex beats of subjects A, B, and C. [Figure 11] FIG. 11 is a model diagram of the volume and pressure changes of the cardiac structural system during the cardiac cycle. [Figure 12] FIG. 12 shows the pressure-volume loop of the human left ventricle along with the pressure fluctuations occurring within the left ventricle during the cardiac cycle. [Figure 13] FIG. 13(a) is a graph showing the relationship between heart rate and duration of a systole in a sitting position at rest, and FIG. 13(b) is a graph showing the relationship between heart rate and duration of a diastole. [Figure 14] FIG. 14 shows a particular solution of a hyperbolic function. [Figure 15] FIG. 15 is a diagram for explaining a method for determining the boundary value between the linearity and nonlinearity of the C value of a particular solution by a graphical solution method. [Figure 16] FIG. 16 is a bar graph showing the slope for each C value. [Figure 17]FIG. 17 is a diagram showing the apex pulsation component waveform and the differential waveform obtained by calculating the slope. [Figure 18] FIG. 18(a) is a graph showing the correlation between the C value and the heart rate at the time of measurement, and FIG. 18(b) is a graph showing the correlation between the frequency of the apex beat vibration component and the heart rate at the time of measurement on the horizontal axis. [Figure 19] FIG. 19 is a graph showing the correlation between the C value and the frequency of the apical pulsation component for healthy subjects, patients with mild abnormalities / requiring follow-up, patients with lifestyle-related diseases, and patients with heart disease / circulatory system diseases. [Figure 20] FIG. 20 is a diagram summarizing the data in FIG. [Figure 21] FIG. 21 is a diagram showing an example in which the apex pulsation component was extracted from the sensor waveform of the vital sign measuring device (4SR) according to the above embodiment for healthy subjects (subjects No. 1-6) in their 20s to 40s. [Figure 22] FIG. 22 is a diagram showing an example in which the apex pulsation component was extracted from the sensor waveform of the vital sign measuring device (4SR) according to the above embodiment for healthy subjects (subjects No. 7-12) in their 20s to 40s. [Figure 23] FIG. 23 is a diagram showing an example in which the apex pulsation component was extracted from the sensor waveform of the vital sign measuring device (4SR) according to the above embodiment for healthy subjects (subjects No. 13-18) in their 20s to 40s. [Figure 24] FIG. 24 is a diagram showing an example in which the apex pulsation component was extracted from the sensor waveform of the vital sign measuring device (4SR) according to the above embodiment for healthy subjects in their 20s to 40s (subjects No. 19-24). [Diagram 25] FIG. 25 is a diagram showing an example in which the apex pulsation component was extracted from the sensor waveform of the vital sign measuring device (4SR) according to the above embodiment for healthy subjects in their 20s to 40s (subjects No. 25-29). [Figure 26] Figure 26(a) is a graph showing the duration of the systole on the horizontal axis and the C value of the pressure waveform on the vertical axis, and Figure 26(b) is a graph showing the duration of the diastole on the horizontal axis and the C value of the pressure waveform on the vertical axis. [Figure 27]FIG. 27(a) is a graph showing the correlation between the apex beat component frequency and the time of the systole, and FIG. 27(b) is a graph showing the correlation between the apex beat component frequency and the time of the diastole. [Figure 28] Figure 28(a) is a graph showing the correlation between systolic blood pressure and C value, Figure 28(b) is a graph showing the correlation between pulse pressure and C value, Figure 28(c) is a graph showing the correlation between diastolic blood pressure and C value, Figure 28(d) is a graph showing the correlation between diastolic blood pressure and apex beat frequency, Figure 28(e) is a graph showing the correlation between pulse pressure and apex beat frequency, and Figure 28(f) is a graph showing the correlation between systolic blood pressure and apex beat frequency. [Figure 29] FIG. 29(a) is a graph showing the correlation between pulse pressure and systolic blood pressure, and FIG. 29(b) is a graph showing the correlation between diastolic blood pressure and systolic blood pressure. [Diagram 30] FIG. 30 is a block diagram showing the configuration of the health monitoring device according to the above embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] The present invention will be described in more detail below based on the embodiment of the present invention shown in the drawings. Figures 1(a) and (b) are diagrams for explaining a biosignal measuring device 1 used in this embodiment. As shown in Figure 1(a), this biosignal measuring device 1 has three biosignal detectors, namely, an upper left biosignal detector 11, an upper right biosignal detector 12, and a lower biosignal detector 13, provided on a base member 10.

[0021] The base member 10 is made of a plate-like body capable of providing three biosignal detection units 11-13 and having an area sufficient to cover the area from the chest to the abdomen of a person. The material is preferably a flexible synthetic resin that causes little discomfort when the person's back contacts it, but is more preferably made of a bead foam. The thin film of beads that make up the bead foam sensitively vibrates in response to micro-vibrations on the body surface based on biosignals, making it easy to transmit the biosignals to the biosignal detection units 11-13.

[0022] Specifically, when the base member 10 is placed along the back of a person, the two upper arrangement holes 10a, 10b are formed in a generally rectangular shape that is long in the vertical direction and are provided corresponding to the position of the fifth intercostal space 6 cm to the left and right of the midline of the chest (back) of an adult male (positions indicated by sensor L and sensor R in Figs. 2(b) and (c) respectively). The lower arrangement hole 10c is formed in a generally rectangular shape that is long in the horizontal direction and is provided corresponding to the position of the third and fourth lumbar vertebrae of an adult male (position indicated by sensor M in Figs. 2(b) and (c)).

[0023] The three-dimensional knitted fabric 100 is disposed in each of the three arrangement holes 10a, 10b, and 10c, and films 14 and 15 are laminated on both sides of the base member 10 so as to cover the front and back surfaces of the three-dimensional knitted fabric 100. A microphone sensor 110 is disposed on the outside of the film 15 located on the side opposite the human body, at a position corresponding to each of the arrangement holes 10a, 10b, and 10c. A gel (silicon gel) 111 is disposed as a disturbance intrusion suppression member so as to surround the outside of the microphone sensor 110, and further, a cover film 112 is provided which covers the outside of the gel 111 and has a periphery fixed to the film 15 (see FIG. 1(b)). As a result, the three-dimensional knitted fabric 100 placed in the placement hole 10a on the left side of the midline, and the microphone sensor 110, gel 111 and cover film 112 corresponding to it and provided on the outside of the film 15 form the biological signal detection unit 11 on the left side of the midline, the three-dimensional knitted fabric 100 placed in the placement hole 10b on the right side of the midline, and the microphone sensor 110, gel 111 and cover film 112 corresponding to it and provided on the outside of the film 15 form the biological signal detection unit 12 on the right side of the midline, and the three-dimensional knitted fabric 100 placed in the lower placement hole 10c, and the microphone sensor 110, gel 111 and cover film 112 corresponding to it and provided on the outside of the film 15 form the biological signal detection unit 13 located in the lower center. The biosignal detection unit 11 on the left side (hereinafter sometimes referred to as "microphone sensor L") mainly captures heart sounds, apical pulsation, and APW originating from the thoracic aorta, while the biosignal detection unit 13 on the lower center (hereinafter sometimes referred to as "microphone sensor M") mainly captures heart sounds, apical pulsation, and APW originating from the abdominal aorta. The biosignal detection unit 12 on the right side (hereinafter sometimes referred to as "microphone sensor R") is a compensation sensor.

[0024] As disclosed in the above-mentioned Patent Document 1, the three-dimensional knitted fabric 100 is formed by connecting a pair of ground knitted fabrics arranged at a distance from each other with a connecting thread. Each ground knitted fabric can be formed, for example, from a yarn made of twisted fibers into a flat knitted fabric structure (fine stitches) that is continuous in both the wale direction and the course direction, or into a knitted fabric structure having a honeycomb-shaped (hexagonal) mesh. The connecting thread imparts a predetermined rigidity to the three-dimensional knitted fabric so that one ground knitted fabric and the other ground knitted fabric maintain a predetermined distance. Therefore, by applying tension in the surface direction, it is possible to string-vibrate the yarns of the opposing ground knitted fabrics that constitute the three-dimensional knitted fabric, or the connecting threads that connect the opposing ground knitted fabrics. As a result, string vibration is generated by the sound and vibration of the heart and blood vessel system, which are biological signals, and is propagated in the surface direction of the three-dimensional knitted fabric. The three-dimensional knitted fabric 100 of this embodiment that functions as a solid oscillator has a thickness of 13 mm.

[0025] The biosignal detection units 11-13 are disposed outside the three-dimensional knitted fabric 100 in which the microphone sensor 110 serves as a natural oscillator, and are surrounded by a disturbance suppression member made of gel 111. Therefore, the biosignal measurement device 1 of this embodiment can propagate weak trunk acoustic vibration signals transmitted from the body surface through a solid, and amplify and capture specific frequency components by stochastic resonance in the process.

[0026] Here, a performance test performed to confirm the performance of the biosignal measuring device 1 of this embodiment will be described. Also, a comparison will be made with the biosignal measuring device 2 described in the section on the prior art in which a three-dimensional knitted fabric 200 and a microphone sensor 210 are arranged together in the arrangement hole of a bead foam 20 shown in FIG. 1(c) and the outside of them is covered with films 24, 25. In the following, the biosignal measuring device 1 of this embodiment will be referred to as 4SR and its microphone sensor 110 will be referred to as an outer microphone because of its arrangement, and the conventional biosignal measuring device 2 will be referred to as 3SR and its microphone sensor 210 will be referred to as an inner microphone because of its arrangement. In both cases, only the output data of the biosignal detecting unit 11 (microphone sensor L) on the left side of the midline will be used. Figure 3(a) shows the experimental method for calculating the output signal-to-noise ratio (hereafter referred to as SNR). The external sound source is a speaker, and the mass of the speaker is 0.4 kg. The load applied to each microphone sensor 110, 210 is the load-deflection characteristic of the back trunk using a φ98 pressure plate, and the maximum load value is 19 N when the deflection is 13 mm. Experiments were conducted under two load conditions: when the speaker weight of 4 N is applied, and when the total value of the speaker weight and the weight is 19 N.

[0027] The input signal generated by the speaker is a log sweep waveform of 4 to 1000 Hz as shown in Fig. 3(b). The signal (noise) amplified by the 3D net using the speaker signal as the input source is captured by the inner microphone (microphone sensor 210) of the 3SR. The outer microphone (microphone sensor 110) of the 4SR captures the solid-borne sound generated by the trunk acoustic vibration signal being amplified by stochastic resonance. 4(a) shows the power spectrum of the inner microphone (microphone sensor 210) measured using the 3SR. The 3SR amplifies input signals from 15 to 100 Hz. Signals below 15 Hz and above 100 Hz are attenuated by the mechanical filter of the 3SR.

[0028] On the other hand, 4SR amplifies input signals from 5 to 200 Hz, with a maximum value at 50 Hz, as shown in Figure 4(b). The responsiveness SNR to the input noise intensity in the frequency band of 4SR is shown in Figure 5(a). The stochastic resonance phenomenon appeared at 10 to 20 Hz, 20 to 30 Hz, and 35 to 100 Hz.

[0029] The biosignals generated in the torso and anterior part of the body captured by a microphone for measuring cardiac sounds (hereafter referred to as PCG) are denoted by 1. PCG , 2 PCG , 3 PCG The results are shown in Fig. 5(a). There were three peaks between 10 and 20 Hz, and higher-order components appeared between 20 and 40 Hz.

[0030] Figure 5(b) shows the biosignals from the front of the chest captured by the PCG and the biosignals from the front and back of the chest captured by the 4SR. Biosignals from 1 to 300 Hz were obtained from the PCG on the front of the chest. There was a difference in the output of the biosignals measured using the 4SR on the front and back of the chest. The code 3 captured by the 4SR placed on the front of the chest 4SR:F is the code 3 captured by PCG. PCG Since the frequency was the same as each of the frequency groups and the output was large, it was assumed to be a heart sound.

[0031] Code 1 captured by PCG and 4SR PCG , 1 4SR:L , 1 4SR:F , 2 4SR:L , 2 4SR:F The biological signal in is a biological signal other than heart sounds, and is thought to be the apex beat component. The apex beat occurs when the apex of the heart strikes the chest wall, and is considered to be a low-frequency component because it can be felt. The heart sound component, which is a water hammer wave caused by turbulence, is a high-frequency component that can be auscultated. Heart sound components also include Nth and 1 / Nth order components, and Nth and 1 / Nth order components can be distinguished from fractal properties. However, when the heart rate is low, such as in the case of an athlete's heart, the apex beat, which has a long diastolic time, may be mixed in with the frequency of the heart sounds. Note that this figure corresponds to the former example, and is marked with symbol 2 4SR:L , 2 4SR:F is the apical pulsation component, and symbol 1 PCG , 1 4SR:L , 1 4SR:F was considered to be the 1 / Nth order component of the heart sound.

[0032] Figure 6(a) shows a linear display of the biosignal captured by the 4SR. 4SR:L , 1 4SR:F The output was the same despite the difference in sensor placement, and it is believed to have been amplified by stochastic resonance. 4SR:L , 1 4SR:F 3 is less than 18 to 30 Hz 4SR:L , 3 4SR:F The 1 / Nth order component of the heart sound is 4SR:L , 2 4SR:F was determined as the fundamental frequency component of the apex beat. Figure 6(b) shows the PCG and 4SR biosignals in double logarithmic enlarged format. PCG , 1 4SR:L , 1 4SR:F The number of frequency groups and the shape of the waveform indicate that the signal is fractal. PCG , 3 4SR:L , 3 4SR:F The 1 / Nth order component of 4SR:L , 2 4SR:F The power spectrum of the heart sounds between 14 and 17 Hz is larger when captured from the posterior part of the chest than from the anterior part, and is considered to be the apex pulsation component due to solid propagation. The anterior apex pulsation component, which is maximum at the left fifth intercostal space and about 10 cm from the midline of the chest, has a power spectrum value that is almost identical to the solid-propagated sound of the heart sounds captured from the front. PCG , 3 4SR:L , 3 4SR:F The power spectrum of 18 to 30 Hz shown in is derived from heart sounds, and the power spectrum of 8 to 14 Hz shown in is derived from heart sounds. PCG , 1 4SR:L , 1 4SR:F is the 1 / Nth order component of the heart sound, and is in the range of 14 to 17 Hz. PCG , 2 4SR:L , 2 4SR:F It was concluded that the spectrum was the apical pulsation component.

[0033] The sensing system using the 4SR to measure from the rear of the chest is a tonometry method in which the body weight is applied to the 4SR in each posture, such as sitting or lying down, and the posture and sensor placement at the time of measurement determine the measurement. It is a measurement system that causes little stress to the person. On the other hand, if the 4SR is placed on the front of the chest, pressure must be applied to the sensor body. In addition, the pressure applied to the sensor is also applied to the person. Therefore, although it is non-invasive, it creates a sense of restraint in the person. In addition, when using a PCG, it must be attached directly to the skin, so a wearable device is the basis. Therefore, biomeasurement using the 4SR from the rear of the chest is considered to be optimal for monitoring the subject's daily life without stress.

[0034] (Apex beat component extraction method using 4SR) Figure 7 shows the waveforms at each processing stage until the apex pulsation component is extracted from the 4SR sensor waveform of three subjects in their twenties with different heart rates and blood pressure values. Subject A has a heart rate of 75 / min, a blood pressure of 105 / 65mmHg, is 175cm tall and weighs 65kg. Subject B has a heart rate of 81 / min, a blood pressure of 109 / 70mmHg, is 160cm tall and weighs 53kg. Subject C has a heart rate of 57 / min, a blood pressure of 114 / 70mmHg, is 171cm tall and weighs 63kg. Subject C is an athlete who regularly plays sports and has excellent endurance and stamina. There are no significant differences in appearance between the three subjects, except for their height.

[0035] Figure 8 shows the results of frequency analysis at each processing stage of medical indicators and signal waveforms measured by 4SR. The results of frequency analysis of the electrocardiogram showed that RRI was extracted in a characteristic manner. The results of frequency analysis of heart sounds showed that heart sounds were a collection of signals from 10 to 100 Hz. The results of frequency analysis of fingertip plethysmograms showed that the amount of information in the low frequency band was greater than that of an electrocardiogram, and that the amount of information of the apex pulsation component was equivalent to that of fingertip plethysmograms. The results of frequency analysis at each processing stage of 4SR showed that the information was narrowed down.

[0036] Figure 9 shows the time-series data of the electrocardiogram, phonocardiogram, fingertip volume pulse wave, and apex pulsation component. The P wave in the electrocardiogram of subject B appears to be high and sharp. Subject B does not have symptoms such as shortness of breath or dyspnea when exercising, so is thought to be a normal subject with good autonomic nervous function and not right atrial overload. The differences in the waveforms of the fingertip volume pulse wave of subjects A, B, and C are due to differences in heart rate and pulse wave velocity, and all three are thought to be young people with normal blood pressure and arterial wall stiffness.

[0037] FIG. 10 shows the process of narrowing down the search according to the procedures shown in FIG. 5 and FIG. 6, and also shows the frequency bands of the heart sounds and apex beat for each subject. The heart sounds are composed of various frequency components, and the apex beat is composed of 1 to 2 peaks. For subject A, the apex beat component and the heart sound component are separated in frequency band, while for subjects B and C, the heart sound component and the apex beat component are mixed. This shows that a method is needed to mechanically know the frequency of the apex beat component in order to reduce the variability in judgment. Next, this point will be explained.

[0038] (Physical characteristics of waveforms extracted by 4SR) The apex beat and heart sounds are feedback forces of events that combine water hammer waves and the cardiac structure. The apex beat is a palpable vibration, and the heart sounds are sounds up to 100 Hz that can be heard with a stethoscope. Figure 11 shows a model diagram of the volume and pressure changes in the cardiac structure during the cardiac cycle. Figure 12 shows the pressure-volume loop of the human left ventricle, along with the pressure fluctuations that occur in the left ventricle during the cardiac cycle.

[0039] The pressure fluctuation waveforms in FIG. 11 and FIG. 12 are defined by a hyperbolic function, and the general solution of the wave equation is given by the following equation (1): TIFF0007607337000001.tif18136[In the formula, the C value is a constant given by the flow rate and pressure, and is an index of the left ventricular pressure waveform that indicates the behavior of the heart.] It becomes.

[0040] It is generally known that even if the heart rate increases, the time for ventricular filling during diastole is ensured, and even if the duration of contraction is shortened, the contraction velocity increases due to the regulation of contractility by the sympathetic nerves, and a sufficient amount of blood is ejected. Also, when cardiac output decreases, the filling time during diastole is shortened. The shortening of the ventricular filling time during diastole is compensated for by the increase in ventricular filling due to the increase in atrial contractility caused by the sympathetic nerves.

[0041] Figure 13(a) shows the relationship between the duration of systole and the heart rate in a sitting position at rest. Figure 13(b) shows the relationship between the duration of diastole and the heart rate. As shown in Figures 13(a) and (b), the duration of systole and diastole is maintained at 0.3 seconds in a sitting position at rest, even when the heart rate exceeds 100 / min. Figure 14 shows a special solution when X=0 and t=±0.35 seconds, such that the tangent of the first derivative is t=±0.175 seconds or less and ±0.15 seconds or more when C=10. Note that C is a constant given by the flow velocity and pressure, and this figure shows C=3 to 8.

[0042] FIG. 15 shows a pressure waveform group defined by a hyperbolic function normalized to obtain the boundary value between the linearity and nonlinearity of the C value of a particular solution by a graphical solution. From the top, a pressure waveform group, a normalized pressure waveform group, a first differential waveform group, a second differential waveform group, a third differential waveform group, and a fourth differential waveform group are shown. In the normalized pressure waveform group, the slope is obtained from the inflection point extracted from each differential waveform group, and the C value at which the characteristics of the pressure waveform change drastically is specified from the rate of change of each slope. FIG. 16 is a bar graph showing the slope for each C value. From these figures, in this case, C=5 was the boundary value. This is based on the verification result that strong nonlinearity is healthy, and in this embodiment, the C value of a healthy adult is 5 or more.

[0043] Figure 17 shows the apex beat component waveform and the differential waveform obtained by calculating the slope, as well as the specific solution and measured values ​​shown in Figure 14, and the C value derived from the cross-correlation function of the first differential waveform and the slope of each of the differential waveforms. The C value determination using the differential waveform has high determination accuracy and is easy to determine mechanically.

[0044] In Fig. 18(a), the gradient of the apex pulsation component, which is the experimental value obtained by the procedures in Figs. 7 to 10, is obtained from the gradient of the special solution in Fig. 14, and the gradient is confirmed by following the analysis procedures in Fig. 17. The graph plots the average C value of the actual measurement values ​​on the vertical axis with reference to the analysis value, and the heart rate at the time of measurement on the horizontal axis. Since there is a correlation between the time required for blood to fill and the heart rate, Fig. 18(a) plots the C value on the vertical axis and the heart rate on the horizontal axis. The coefficient of determination R 2 The R for patients with lifestyle-related diseases was 0.8536, for patients with heart disease and circulatory system diseases was 0.5948, and for healthy individuals it was 0.3751. 2 =0.2801, which shows that it is better to grasp the C value according to physical condition. The singular point of heart rate is 80~90 / min.

[0045] FIG. 18(b) is a graph showing the frequency of the apex pulsation vibration component that coincides with the time phase of ventricular systole obtained from the APW according to the examination procedure of FIG. 3 to FIG. 10 on the vertical axis and the heart rate at the time of measurement on the horizontal axis. This graph shows the coefficient of determination R 2 was 0.7636. Here again, the singular point of the heart rate was 80-90 / min. Figure 18(b) is intended to identify the frequency band of the apex beat from the heart rate when analyzing data from a subject in which heart sound components and apex beat components are mixed. Note that the classification of healthy subjects, patients with lifestyle-related diseases, and patients with heart disease / circulatory system disease in Figures 18(a) and (b) is based on self-reporting.

[0046] Combining Figs. 18(a) and (b) and eliminating the heart rate as in Figs. 19 and 20 to create a graph of C value and frequency components, it may be possible to distinguish between lifestyle-related diseases and other diseases. Fig. 19 shows data for healthy subjects, mild abnormalities and those requiring follow-up, lifestyle-related diseases, and patients with heart disease and circulatory disease, and Fig. 20 shows them all together. However, Fig. 20 shows data for healthy subjects and mild abnormalities and those requiring follow-up as healthy subjects in their 20s to 70s. Fig. 20 also shows data for a certain subject three months before and one month before death. Three months before death, the C value was about 7.5, a value classified as a healthy subject, while one month before death, the C value was about 4.5, a very low value, indicating a sudden change in physical condition.

[0047] In Figures 19 and 20, the vertical axis represents the behavior of the heart, such as the left ventricular pressure waveform (C value), and the horizontal axis represents the frequency components of the vibration transmitted to the chest wall. These C values ​​and frequency components of the vibration are due to the apex pulsation. The frequency of the impinging vibration on the horizontal axis is considered as a physical index of the stretching force of the myocardium during the ventricular ejection period. In the case of a subject at rest with a heart rate of 67 / min, if it is assumed that a half wave or one wave of APW occurs during 0.05 seconds of isovolumic contraction, the APW originating from the isovolumic contraction period will be 10 to 20 Hz. If this wave continues in the same manner for 0.3 seconds of the ventricular ejection period, three to six waves will occur as the APW originating from the ventricular ejection period. Next, taking into account individual differences, if a half wave or one wave of APW occurs during 0.035 seconds of isovolumic contraction, the APW originating from the ventricular contraction will be 14.5 to 29 Hz. If we consider the ventricular ejection period to be 0.175 seconds, the APW is 17.25 to 34.5 Hz. Therefore, the apical beat component derived from ventricular contraction is present in the APW range of 10 to 34.5 Hz.

[0048] The pressure waveform, which is the basis of cardiac behavior on the vertical axis, is considered to be one of the factors that determine the stroke volume experimentally obtained from APW. Following the contraction of the atrium, the ventricle begins to contract. Figures 11 and 12 show how the ventricular pressure rises due to the contraction of the ventricle, and the moment it becomes higher than the atrial pressure, the atrioventricular valve closes, and the tension caused by the contraction of the myocardium compresses the blood in the ventricle, causing the internal pressure to rise sharply. The moment the left ventricular pressure becomes higher than the arterial pressure, the aortic valve is pushed open and blood begins to be pumped out. The speed at which blood is pumped out of the left ventricle is greater than the speed at which blood flows out to the peripheral blood vessels, so most of the pumped blood expands the elastic blood vessels. To expand the elastic blood vessels, blood temporarily remains at the entrance of the left coronary artery. This causes the arterial pressure to rise, becoming the systolic blood pressure. Figure 11 shows the hypothetical process by which pressure waves are generated during the cardiac cycle. The baseline of the pressure fluctuation is assumed to be the pressure value of the remaining blood volume in the left ventricle, i.e., the end-systolic volume. The end-systolic volume is 1 / 3, and it is assumed that the internal pressure line of this 1 / 3 volume is used as the baseline, and that 3-4 phases of positive-negative-positive or negative pressure waves produce a steep hyperbolic positive pressure wave on the baseline just before the aortic valve opens when the ventricle contracts.

[0049] The envelope of the 1-6 or 8-phasic wave created by the APW is considered to approximate the volume fluctuation from the mitral valve closing to the mitral valve opening, and is assumed to contain information equivalent to the pressure-volume loop. Therefore, if the half wave is created during 0.3 seconds of the ventricular ejection period, it will be 1.7 Hz, if the half wave is created during 0.175 seconds of the ventricular ejection period, it will be 2.9 Hz, and if the half wave is created during 0.1 seconds of the ventricular ejection period, it will be 5 Hz. This is the basis for the low-pass filter of 5 Hz to create the envelope shown in Figure 7. The analysis procedure for extracting biosignals derived from ventricular contraction will be considered. As mentioned above, the analysis target frequency of the APW derived from ventricular contraction is 10 to 34.5 Hz, and the low-pass filter is 5 Hz. The waveform is captured by the microphone sensor L and is considered to be between the isovolumic contraction period and the isovolumic relaxation period. There are individual differences among healthy people in their 20s to 40s, and it is assumed that there are waves that fall somewhere between 10 and 34.5 Hz. All fundamental frequencies within the analysis range were analyzed, waveforms were extracted, and waveforms present in the time phase starting from the first sound of the phonocardiogram and the QRS wave of the electrocardiogram and ending at the second sound of the phonocardiogram were found. Figures 21 to 25 correspond to Figure 9, which follows the procedure of Figures 7 and 8 for healthy subjects (subjects No. 1 to 29) in their 20s to 40s. The simulation values ​​in the figures follow the analysis procedures of Figures 14 and 17. In addition, the graphs of subjects No. 1 to 29 shown in Figures 21 to 25 are, from the top, time series data of ECG (electrocardiogram), time series data of PCG (phonocardiogram), time series data of PPG (pigment plethysmogram), time series data of left ventricular pressure obtained from the phonocardiogram, and time series data of left ventricular pressure obtained from the biological signal detection unit 11 (microphone sensor L).

[0050] The APW measured from the 4SR has different time phases, heart murmurs, and slope of the onset compared to the waveform captured from the heart sounds in the front of the chest, and it can be seen that the time phase of the onset of the pressure wave and the timing of the disappearance of the first sound correspond to the 2 / 3 time phase of the pressure wave. This shows that the APW measured from the back of the chest provides different information from the heart sounds.

[0051] This pressure wave is assumed to follow the general solution of equation (1). Therefore, the rise of the pressure wave generated in the aorta is also assumed to follow the general solution of equation (1). 3 / 4 of the stroke volume is pumped out in the first 0.05 + 0.15 = 0.2 seconds of the systole, and the remaining 1 / 4 is pumped out in the second half of the systole. Then, the peak of the pressure wave according to the solution of equation (1) is located at approximately half the time phase of the ventricular systole, starting from the first heart sound. In addition, since the aortic valve is released the moment the left ventricular pressure becomes higher than the aortic pressure, the half-power point of the waveform derived from the solution of equation (1) is assumed to be the starting point. In addition, since the half-power point is the aortic valve opening, it is the timing when the first heart sound disappears. From these, the aortic pressure wave rises from a pressure value 0.707 times the amplitude of the wave formation. Therefore, if the systolic blood pressure is 120mmHg, the diastolic blood pressure is 120×0.707=84.8mmHg. In the latter half of systole, the blood pumping speed decreases, and the blood flows out to the peripheral blood vessels at a higher speed, so the blood pressure starts to decrease. Then, the aortic valve closes due to a momentary backflow. With the aortic valve closing, a positive pressure wave is mixed into the apical pulse wave. At this time, 2 / 3 of the blood in the left ventricle at the end of diastole is pumped out. From the viewpoint of the stroke volume, the baseline of the aortic pressure wave can be said to be the baseline pressure of about 2 / 3 of the pressure amplitude created by the fluctuation of the left ventricular pressure. In other words, if the systolic blood pressure is 120mmHg, the diastolic blood pressure is 120×2 / 3=80mmHg, and the baseline of the aortic pressure wave, the diastolic blood pressure, is between about 80-85mmHg, and the hypothesis is that the timing of the disappearance of the first heart sound is the 2 / 3 point of the pressure wave. These hypotheses will be examined based on experimental data.

[0052] In the normal hearts of healthy people in their 20s to 40s, the larger the ventricular volume during diastole, the more myocardial fibers are stimulated to stretch, and the greater the subsequent contractile force. Therefore, as ventricular preload increases, the amount of blood pumped from the ventricle increases. Figures 26(a) and (b) show the duration of diastole on the horizontal axis and the C value of the pressure waveform on the vertical axis. The regression line for physical condition shows a positive correlation between the duration of diastole and the C value, which is indicative of the present description.

[0053] If the diastolic time is long, the apical beat component vibration frequency will be high, and conversely, if the diastolic time is short, the apical beat component vibration frequency will be low. Therefore, in a normal heart, there is a high correlation between the apical beat component frequency and preload. Figure 27 is a graph showing the correlation between the apical beat component frequency and the time of systole and diastole. It can be seen that the coefficient of determination is higher during diastole compared to systole. This is because when myocardial cells contract, they are subjected to a certain resistance called afterload, which is determined by the systolic intraventricular pressure (arterial pressure), and is not dependent on the preload on the myocardial fibers before contraction. Contractility is independent of preload and afterload, and is affected by chemical stimuli and hormones during contraction, which is the basis for the difference in the coefficient of determination.

[0054] Figure 28 shows the correlation between blood pressure and C value, and blood pressure and apical beat frequency. In subjects affected by afterload, C value decreases. Arterial pressure or pulse pressure becomes afterload, and when pulse pressure is high, C value decreases. Figure 29(a) shows the correlation between pulse pressure and systolic blood pressure, and Figure 29(b) shows the correlation between diastolic blood pressure and systolic blood pressure. Both show a positive correlation. Compared to normal hearts, subjects with lifestyle-related diseases who have a high afterload have a higher intercept value and a steeper slope. Furthermore, because it is strongly influenced by factors that determine cardiac output, the coefficient of determination R 2 The coefficient of determination R 2 In addition, because afterload is also controlled, the C value increases in contrast to patients with lifestyle-related diseases. The slope of the regression line shows the same tendency as patients with lifestyle-related diseases. Therefore, it is possible that when the vibration frequency of the apical pulsation component, which depends on arterial pressure, is high, the contractile force is large and the C value increases.

[0055] Here, in Figures 19 and 20, if we think of it in terms of the cardiac cycle, the vertical axis is the input and the horizontal axis is the output. The labels in the figures are based on the classification of judgments made by the Japan Society of Human Health Checkups, revised on April 1, 2018 and partially revised on December 14, 2018. Classification A: No abnormality for healthy people in their 20s to 40s, B: Mild abnormality / C: Needs follow-up for mild abnormality / needs follow-up, and D: Needs treatment / E: Under treatment (lifestyle-related disease). Patients with heart disease or circulatory system disease are patients whose basic daily life is to stay in a supine position and be treated by a doctor and certified as requiring nursing care level 3. People who have heart disease but are not able to live a normal daily life are classified as being under treatment (lifestyle-related disease). The C value of healthy people in their 20s to 40s was the median, and the C value of subjects with lifestyle-related disease was low, with the C value increasing in the order of mild abnormality / needs follow-up and patients with heart disease or circulatory system disease.

[0056] According to this embodiment, the apical pulsation component was extracted using the APW captured by the posterior chest. The pressure waveform during the ventricular ejection period was created from the apical pulsation component, and the pressure waveform during the ventricular ejection period was identified by a hyperbolic function, and the C value related to the pressure and flow rate of the measured pressure waveform, that is, an index related to the left ventricular pressure waveform, was obtained. In addition, it was found that the C value of the apical pulsation component and the diastolic duration have a correlation according to physical condition, and the frequency of the apical pulsation component and the diastolic duration have a high correlation regardless of physical condition. In addition, it was found that the C value, the frequency of the apical pulsation component, and blood pressure tend to have a high correlation with subjects with lifestyle-related diseases. From these findings, it was suggested that by using the C value and the frequency of the apical pulsation component as parameters, it is possible to classify subjects based on the judgment classification of the Japanese Society of Human Health Checkups revised on April 1, 2018 and partially revised and updated on December 14, 2018, and to determine whether they are healthy, have mild abnormalities and require follow-up, have lifestyle-related diseases, or have heart disease or circulatory disease.

[0057] Based on the above description, in this embodiment, a health monitoring device shown in Fig. 30 is configured. That is, the health monitoring device 1000 processes a trunk acoustic signal measured from a person's trunk and estimates the health condition of the person, and includes a left ventricular pressure waveform identification means 1100 that determines an index related to a left ventricular pressure waveform indicating cardiac behavior from the trunk acoustic signal obtained from the above-mentioned vital sign measuring device 1 (4SR), an apex pulsation component extraction means 1200 that extracts a frequency component derived from the apex pulsation from the trunk acoustic signal, and an estimation means 1300 that estimates the health condition of the person from the correlation between the index related to the left ventricular pressure waveform and the vibration frequency of the frequency component derived from the apex pulsation.

[0058] The left ventricular pressure waveform specification means 1100 is a means for specifying the C value indicating the left ventricular pressure waveform as described above. The C value is obtained by filtering the trunk acoustic vibration signal in a predetermined frequency band as described above, excluding components related to the heart sound component, determining the remaining frequency components as the apex pulsation component, forming a pressure waveform from this apex pulsation component, and determining an index related to the slope of the pressure waveform using a hyperbolic function. Depending on the C value obtained as described above, a different value is output depending on whether the person is healthy or not, and other health conditions. When the heart sound component and the apex pulsation component are mixed, the frequency of the apex pulsation component is determined using the correlation diagram with the heart rate shown in FIG. 18(b).

[0059] The frequency component derived from the apex pulsation identified by the apex pulsation component extraction means 1200 is the frequency of the apex pulsation, which is the feedback force when pumping blood from the heart, and if the C value indicating the state of the pressure waveform is the input information, the frequency component becomes the output information. Therefore, by plotting these on the vertical and horizontal axes as described above, it is possible to create correlation diagrams according to healthy subjects, mild abnormalities / requiring follow-up, patients with lifestyle-related diseases, and patients with heart disease / circulatory system diseases.

[0060] If the health monitoring device 1000, which is composed of a computer, stores this as a database, the estimation means 1300 can analyze the trunk acoustic vibration signal obtained from the above-mentioned vital sign measuring device 1 for the subject whose health is being monitored using the left ventricular pressure waveform identification means 1100 and the apex beat component extraction means 1200, and compare this with the above-mentioned database, thereby estimating the health condition of the subject whose health is being monitored.

[0061] Since there is a high correlation between the C value and the diastolic duration as shown in FIG. 26(b), the estimation means 1300 can use this to estimate whether a patient is a healthy subject or a patient with a lifestyle-related disease.

[0062] In addition, as shown in FIGS. 28(a) to (c), the C value also correlates with blood pressure, so the estimation means 1300 can also estimate blood pressure using the C value.

[0063] In the health monitoring device 1000, a computer program for executing procedures functioning as the above-mentioned left ventricular pressure waveform identifying means 1100, apical pulsation component extracting means 1200, and estimation means 1300 is stored in a storage unit (including storage media such as a built-in hard disk of the computer (health monitoring device 1000), as well as various removable storage media and storage media of other computers connected via communication means). Furthermore, the computer program causes a computer to execute each procedure, thereby functioning as the left ventricular pressure waveform identifying means 1100, apical pulsation component extracting means 1200, and estimation means 1300. Furthermore, the computer program can also be realized by an electronic circuit having one or more storage circuits incorporating computer programs for realizing the left ventricular pressure waveform identifying means 1100, apical pulsation component extracting means 1200, and estimation means 1300.

[0064] The computer program can be provided by storing it in a recording medium. The recording medium storing the computer program may be a non-transient recording medium. There is no particular limitation on the non-transient recording medium, and examples of the non-transient recording medium include a flexible disk, a hard disk, a CD-ROM, an MO (magneto-optical disk), a DVD-ROM, and a memory card. It is also possible to transmit the computer program to a computer via a communication line and install it therein. [Explanation of symbols]

[0065] 1 Biosignal measuring device (4SR) 10 Base material 11 Upper left vital sign detector (sensor L) 12 Upper right biosignal detector (sensor R) 13 Lower vital sign detection unit (sensor M) 14,15 Film 100 3D knitting 110 Microphone Sensor 1000 health monitoring equipment 1100 Left ventricular pressure waveform identification means 1200 Apex beat component extraction means 1300 Estimation means

Claims

1. A health monitoring device that processes a trunk acoustic signal measured from a person's trunk and estimates a health condition of the person, a left ventricular pressure waveform determining means for determining an index relating to a left ventricular pressure waveform, which indicates cardiac behavior, from the trunk acoustic signal; an apex pulsation component extraction means for extracting a frequency component derived from the apex pulsation from the trunk acoustic signal; an estimation means for estimating a health condition of a person based on a correlation between an index relating to the left ventricular pressure waveform and a vibration frequency of a frequency component derived from the apex beat; A health monitoring device comprising:

2. The estimation means includes:

2. The health monitoring device according to claim 1, wherein the health condition is estimated from a correlation between the index relating to the left ventricular pressure waveform and a time of diastolic phase in the cardiac cycle which indicates the magnitude of the vibration frequency of the frequency component derived from the apex beat.

3. 3. The health monitoring device according to claim 1, wherein the apex pulsation component extracting means extracts a frequency component derived from the apex pulsation from the trunk acoustic signal in comparison with a heart sound component.

4. 4. The health monitoring device according to claim 3, wherein the apex pulsation component extracting means further extracts the frequency component derived from the apex pulsation in consideration of a relationship with a heart rate.

5. 5. The health monitoring device according to claim 1, wherein the trunk acoustic signal is a signal captured by a vital sign measuring device placed on the rear of the person's chest.

6. The biosignal measuring device, A bead foam molded into a plate shape; A three-dimensional knitted fabric is loaded into the arrangement holes formed in the bead foam. A film is attached to both sides of the bead foam so as to cover both sides of the three-dimensional knitted fabric; A microphone sensor is disposed on the outer side of the film in correspondence with a location where the three-dimensional knitted fabric is disposed; A cover film for covering the microphone sensor; a disturbance intrusion suppression member that surrounds the microphone sensor from the outside within the cover film and suppresses the intrusion of disturbances; have 6. The health monitoring device of claim 5.

7. A computer program for processing a trunk acoustic signal measured from a human trunk and causing a computer to function as a health monitoring device for estimating a health condition of the human, comprising: A step of obtaining an index relating to a left ventricular pressure waveform, which indicates a behavior of the heart, from the trunk acoustic signal; extracting a frequency component originating from an apex beat from the trunk acoustic signal; a step of estimating a health condition of a person based on a correlation between an index relating to the left ventricular pressure waveform and a vibration frequency of a frequency component derived from the apex beat; A computer program that causes the computer to execute the above.

8. 8. The computer program of claim 7, wherein the step of estimating the health condition of the person comprises the step of causing the computer to execute a step of estimating the health condition from a correlation between an index related to the left ventricular pressure waveform and a time of diastolic phase in a cardiac cycle which indicates the magnitude of the vibration frequency of the frequency component derived from the apex beat.

9. 9. The computer program product according to claim 7, wherein the step of extracting the frequency component derived from the apex beat comprises the step of extracting the frequency component derived from the apex beat from the trunk acoustic signal in comparison with a heart sound component.

10. The computer program according to claim 9 , wherein the step of extracting the frequency component derived from the apex beat further extracts the frequency component derived from the apex beat taking into consideration a relationship with a heart rate.

11. A recording medium on which the computer program according to any one of claims 7 to 9 is recorded.

Images