Medical Instrument and Program

Abstract

To provide a blood pressure estimating device whereby information relating to the blood pressure of a subject can be acquired by a simple method. A blood pressure estimating device 10 is provided with a frequency analysis unit 11 for specifying the peak value of the heart sound frequency of a subject, and a blood pressure estimating unit 17 for determining information relating to the blood pressure of the subject on the basis of the peak value of the frequency. It is clear that there is a positive correlation between the peak value of the heart sound frequency and the blood pressure value; therefore, by analyzing the heart sound of a subject and specifying a frequency peak value, the blood pressure value of the subject can be easily and accurately estimated.

Claims

1. A medical device comprising: a frequency analysis unit that identifies a peak value of a frequency of a heart sound of a subject; and a blood pressure estimating unit that obtains the change or the estimated value of a blood pressure of the subject based on the peak value of the frequency.

2. The medical device according to claim 1, wherein the blood pressure estimating unit obtains the change in the blood pressure of the subject based on changes over time of the peak value of the frequency.

3. The medical device according to claim 1, wherein the blood pressure estimating unit obtains a current estimated value of the blood pressure of the subject based on a data set and a current peak value of the frequency of the heart sound of the subject, the data set including a peak value of the frequency of the heart sound and an actual measured value of the blood pressure of the subject measured in a past.

4. The medical device according to claim 1, wherein the blood pressure estimating unit obtains a current estimated value of the blood pressure of the subject based on a data set and a current peak value of the frequency of the heart sound and an actual measured current value of the heart rate of the subject, the data set including a peak value of the frequency of the heart sound, an actual measured value of the blood pressure, and an actual measured value of a heart rate of the subject measured in the past.

5. The medical device according to claim 1, wherein the blood pressure estimating unit obtains a current estimated value of the blood pressure of the subject based on a current peak value of the frequency of the heart sound of the subject using a learned model obtained by machine learning from a data set of peak values of heart sound frequencies and blood pressures of a plurality of subjects.

6. The medical device according to claim 3 or claim 4, further comprising a cardiac function diagnostic unit that identifies a change in a cardiac function of the subject based on an estimated value of the blood pressure of the subject and an actual measured current value of the blood pressure of the subject.

7. The medical device according to claim 1, further comprising a cardiac function diagnostic unit that identifies a change in a severity of a valvular heart disease of the subject from the difference between a peak value of a frequency of a systolic murmur or a diastolic murmur of a heart measured in a past and a current peak value of the frequency of the systolic murmur or the diastolic murmur of the heart of the subject.

8. A program for causing a computer to execute: a step of identifying a peak value of a heart sound frequency of a subject; and a step of obtaining a change or an estimated value of a blood pressure of the subject based on the peak value of the frequency.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0022] FIG. 1 is a function block diagram illustrating an example of a configuration of an entire system including a blood pressure estimating device and its peripherals.

[0023] FIG. 2 illustrates an example of spectrograms showing a heart sound frequency.

[0024] FIG. 3 illustrates an example of spectrograms showing the heart sound frequency.

[0025] FIG. 4 illustrates an example of spectrograms showing the heart sound frequency.

DESCRIPTION OF PREFERRED EMBODIMENTS

[0026] Hereinafter, configurations to embody the present invention will be described using the drawings. The present invention is not limited to the configurations described below but includes ones appropriately changed in an obvious range from the following configurations by those skilled in the art.

[0027] FIG. 1 illustrates the configuration of the entire system including a blood pressure estimating device 10 (medical device), a digital stethoscope 20, a sphygmomanometer 30, a display device 40, and a communication device 50 as its peripherals according to the present invention. The blood pressure estimating device 10 can be achieved by a computer storing specific programs. For example, the blood pressure estimating device 10 may be a mobile terminal, such as a laptop computer, a tablet computer, and a smart phone, or may be a stationary terminal, such as a desktop computer and a web server. When predetermined data is input from the digital stethoscope 20 and the sphygmomanometer 30, the blood pressure estimating device 10 performs an arithmetic operation in accordance with the program and outputs the result of the arithmetic operation to the display device 40 and the communication device 50.

[0028] FIG. 1 also illustrates a function block of the blood pressure estimating device 10 achieved by the program specific to the present invention. As illustrated in FIG. 1, the blood pressure estimating device 10 includes a frequency analysis unit 11, a blood pressure actual measurement unit 12, a heart rate measurement unit 13, a data accumulation unit 14, a database 15, a learned model 16, a blood pressure estimating unit 17, a cardiac function diagnostic unit 18, and an output unit 19. That is, this program is described such that the computer achieves these functions.

[0029] The frequency analysis unit 11 analyzes the frequency of the heart sound data of the subject obtained from the digital stethoscope 20. The publicly known digital stethoscopes can be used as the digital stethoscope 20. The digital stethoscope 20 internally includes, for example, a microphone for biological sound and directly contacts a skin of the subject to acquire the biological sound (mainly the heart sound). Although a dynamic microphone and a condenser microphone can be used as the microphone for biological sound, to accurately collect a biological sound in a further lower frequency band, the use of a piezoelectric microphone is especially preferred. The piezoelectric microphone converts a sound vibration added to a piezoelectric element into a voltage and basically includes the piezoelectric element and a plurality of electrodes between which the piezoelectric element is sandwiched. The microphone for biological sound only needs to have performance that can collect a frequency (10 Hz to 500 Hz) of the heart sound.

[0030] The frequency analysis unit 11 analyzes the heart sound data obtained from the digital stethoscope 20 and identifies the peak value of the frequency [Hz]. The peak value of the frequency is a peak value of an envelope. The frequency analysis unit 11 preferably generates a spectrogram (three-dimensional graph) expressing changes over time of the sound volume of each frequency from the heart sound data (see FIG. 2 to FIG. 4). For example, the spectrogram plots the frequency on a vertical axis and plots the time on a horizontal axis, and expresses the sound volume by a color tone or brightness in the graph (the vertical axis and the horizontal axis can be switched). Note that FIG. 2 to FIG. 4 are illustrated in monochrome, but a frequency bandwidth in which the sound volume is high is expressed by red and the frequency bandwidth in which the sound volume is low is expressed by blue actually. The spectrogram allows easily identifying the peak value of the heart sound frequency.

[0031] The heart sound is a sound generated in association with a pulse of a heart and generates the first heart sound and the second heart sound. Among these sounds, a sound generated immediately after the start of the systole of the heart is the first heart sound and a sound generated at the border between the systole and the diastole is the second heart sound. The heart sound includes heart murmurs in some cases. The heart murmur is a sound generated in association with the heartbeat but is not generated in the normal heart. The frequency analysis unit 11 preferably identifies the frequency peak value of the sound of both or any one of the first heart sound component or a second component heart sound among these sounds included in the heart sound. As described later, the estimated value of a systolic blood pressure can be obtained from the frequency peak value of the first heart sound component and the estimated value of a diastolic blood pressure can be obtained from the frequency peak value of the second heart sound component. Additionally, when the heart murmur is included in the heart sound, the frequency analysis unit 11 may identify a frequency peak value of a heart murmur component.

[0032] The blood pressure actual measurement unit 12 measures the actual blood pressure of the subject from the data obtained from the sphygmomanometer 30. The publicly known sphygmomanometer 30 can be used. The sphygmomanometer 30, for example, includes a cuff wound around an arm of the subject, a pump that supplies an air to the inside of the cuff, a pressure sensor that converts an air pressure inside the cuff into an electrical signal, and so on. For example, the blood pressure measurement unit 12 measures the actual blood pressure of the subject based on the electrical signal obtained from this pressure sensor. The blood pressure value measured by the blood pressure measurement unit 12 may be a systolic blood pressure, a diastolic blood pressure, or the both. The blood pressure actual measurement unit 12 preferably measures the blood pressure in the heart systole and the blood pressure in the heart diastole. Although a method for measuring the blood pressure is not specifically limited, the known invasive arterial pressure measurement or non-invasive arterial pressure measurement, for example, may be employed. The blood pressure can be measured from the pulse pressure of the subject.

[0033] The heart rate measurement unit 13 measures the heart rate of the subject based on the heart sound data obtained from, for example, the digital stethoscope 20. For example, the heart rate can be obtained by counting periodicity of the strength and weakness of the sound components included in the heart sound data for a certain period. While the example illustrated in FIG. 1 obtains the heart rate from the heart sound data acquired by the digital stethoscope 20 to make the entire system compact, an electrocardiograph (such as electrodes) may be separately connected to the blood pressure estimating device 10 to measure the heart rate.

[0034] The data accumulation unit 14 causes the database 15 to store data of the peak value of the heart sound frequency identified in the frequency analysis unit 11, the blood pressure value measured in the blood pressure actual measurement unit 12, and the heart rate measured in the heart rate measurement unit 13. Especially, in the initial examination, the blood pressure and the heart sound of the subject are acquired simultaneously or under identical conditions. The actually measured blood pressure values obtained from the blood pressure and the heart sound, the heart sound frequency peak value, and the heart rate are preferably associated and stored in the database as one data set. A plurality of data sets may be generated for one subject. Even when the data set as described above is not generated, when the peak power of the heart sound frequency, the actual measured value of the blood pressure, and the heart rate are obtained, the data accumulation unit 14 preferably stores these values in the database 15 as needed. Further, as described later, the blood pressure estimating unit 17 calculates the estimated value of the blood pressure of the subject based on the peak value of the heart sound frequency. The data accumulation unit 14 preferably associates this estimated value of the blood pressure with the source peak value of the heart sound frequency and stores it in the database 15. Thus, the data accumulation unit 14 appropriately accumulates the various pieces of biological data obtained in the blood pressure estimating device 10 in the database 15.

[0035] The learned model 16 is model data in which parameters (what is called “weights”) are adjusted through machine learning on the biological data of multiple subjects. For example, performing the machine learning, such as deep learning, using the data sets of the peak values of the heart sound frequencies and the actually measured values of the blood pressures of the multiple subjects as teaching data generates the learned model 16. In this case, reference to this learned model 16 with the peak value of the heart sound frequency of a certain subject as an input value allows obtaining the estimated value of the blood pressure corresponding to the input value as an output value. The blood pressure estimating device 10 may preliminarily include the learned model 16. Note that this learned model 16 is not an indispensable element.

[0036] The blood pressure estimating unit 17 estimates the blood pressure fluctuation and the blood pressure value of the subject based on at least the peak value of the heart sound frequency obtained from the frequency analysis unit 11. The blood pressure estimating unit 17 can perform a predetermined arithmetic operation according to a situation, such as a mode that obtains only the blood pressure fluctuation of the subject and a mode that obtains the estimated value of the blood pressure of the subject. Details of the arithmetic operation mode by the blood pressure estimating unit 17 will be described later.

[0037] The cardiac function diagnostic unit 18, for example, compares the estimated value of the blood pressure obtained by the blood pressure estimating unit 17 with the actual measured value of the blood pressure measured in the blood pressure actual measurement unit 12, and determines whether the cardiac function of the subject is normal, and the severity of the cardiac dysfunction or its change based on the difference and the ratio of these values. For example, the actual measured value and the estimated value of the blood pressure of the subject are obtained simultaneously or under the identical conditions using the digital stethoscope 20 and the sphygmomanometer 30. When the difference and the ratio of these values are equal to or more than certain thresholds, the cardiac function diagnostic unit 18 determines that the cardiac function is abnormal. According to the difference between the actual measured value and the estimated value of the blood pressure, presence of a sign of the cardiac dysfunction, such as myocardial infarction, heart failure, and arrhythmia, may be identified.

[0038] Additionally, the cardiac function diagnostic unit 18 can identify the severity of the valvular heart disease from the difference between the peak value of the frequency of the systolic murmur or the diastolic murmur of the heart measured in the past and the current peak value of the frequency of the systolic murmur or the diastolic murmur of the heart of the subject. For example, a comparison between the peak value of a heart sound frequency of a person who has been already affected with the valvular heart disease and the peak value of the heart sound frequency of a certain subject allows determining whether the subject is affected with the valvular heart disease and its severity. Additionally, a comparison of the peak value of the heart sound frequency measured in the past with the current peak value of the heart sound frequency of a certain subject allows identifying the change (the exacerbation or improvement) in the severity of the valvular heart disease.

[0039] The output unit 19 outputs the results obtained from the blood pressure estimating unit 17 and the cardiac function diagnostic unit 18 to the display device 40 and the communication device 50. For example, the output unit 19 can display the blood pressure fluctuation, the estimated value of the blood pressure, and the presence/absence of the cardiac dysfunction in the display device 40. The output unit 19 may display an analysis result (frequency peak value) of the heart sound frequency, the actual measured value of the blood pressure, or information on the heart rate in the display device 40. Further, the output unit 19 can transmit various pieces of the information obtained by the blood pressure estimating device 10 to an external terminal via an information communication network, such as the Internet, through the communication device 50. For example, it is preferred that the subject himself/herself operates the blood pressure estimating device 10 and its peripherals 20, 30, 40, 50 to measure the estimated value of the blood pressure and transmits the measured value to a terminal of the doctor residing in a remote location. Thereby, the estimated value of the blood pressure can be used for telemedicine.

[0040] Next, an example of the arithmetic operation mode that can be performed by the blood pressure estimating unit 17 will be described.

[1. Estimation of Blood Pressure Fluctuation]

[0041] The blood pressure estimating unit 17 can estimate the blood pressure fluctuation of the subject based on the change over time of the peak value of the heart sound frequency. Since the peak value of the heart sound frequency and the blood pressure value are related to positive correlation, when the peak value of the heart sound frequency changes, it can be estimated that the blood pressure value similarly changes. When the blood pressure varies, the frequency peak value of a second heart sound component makes the biggest change among the heart sounds, and therefore the frequency peak value of the second heart sound component is preferably referred for estimation of the blood pressure fluctuation. Although it is medically known that the second heart sound increases at a high blood pressure in auscultatory findings of a high blood pressure patient (that is, power of the second heart sound increases), the inventor has found new knowledge that the blood pressure and the peak value of the heart sound frequency are related to positive correlation in real-time, and the blood pressure can be estimated from the change in the peak value of the frequency of the second heart sound. Further, for further accurate identification of blood pressure fluctuation, a data set as a reference in which an actually measured normal blood pressure value is associated with the peak value of the heart sound frequency may be preliminarily generated, and the blood pressure estimating unit 17 may compare a peak value of a heart sound frequency measured after that with this data set to obtain the extent of change in the peak value of the heart sound frequency. When the change in the peak value of the heart sound frequency is large, it can be determined that the blood pressure greatly changes by the amount.

[2. Estimation of Blood Pressure Value Based on Heart Sound Frequency]

[0042] The blood pressure estimating unit 17 can obtain the estimated value of the blood pressure (the systolic blood pressure or the diastolic blood pressure) based on the following relational equation using the peak value of the heart sound frequency as a parameter.

[00001]$\begin{array}{cc}\mathrm{Estimated}\mathrm{blood}\mathrm{pressure}=\left(\alpha \times \frac{D-C}{C}+1\right)\times A& \left[\mathrm{Formula}1\right]\end{array}$

[0043] α denotes a coefficient related to the peak value of the heart sound frequency used to estimate the blood pressure. To distinguish between the systolic blood pressure and the diastolic blood pressure, different values between the systolic blood pressure and the diastolic blood pressure may be used for α. A denotes a blood pressure value as a reference measured in the past. C denotes a frequency peak value (a value measured simultaneously with or under a condition identical to that of A) of the second heart sound component as a reference measured in the past. D denotes a frequency peak value of the second heart sound component measured this time. Values measured when the subject is in a normal health condition are preferably used for A and C. Any given value can be employed for the coefficient α.

[0044] The equation 1 is one example. When the fact that the peak value of the heart sound frequency and the blood pressure value show the positive correlation is applied, by identifying the peak value of the heart sound frequency, the blood pressure value corresponding to it can be estimated.

[3. Estimation of Blood Pressure Value Based on Heart Sound Frequency and Heart Rate]

[0045] The blood pressure estimating unit 17 can obtain the estimated values of the systolic blood pressure and the diastolic blood pressure based on the following relational equation using the peak value of the heart sound frequency and the heart rate as parameters.

[00002]$\begin{array}{cc}\mathrm{Estimated}\mathrm{blood}\mathrm{pressure}=\left(\beta \times \frac{F-E}{E}+1\right)\times \left(\gamma \times \frac{G-H}{G}+1\right)\times B& \left[\mathrm{Formula}2\right]\end{array}$

[0046] B denotes a blood pressure value as a reference measured in the past. β denotes a coefficient related to the peak value of the heart sound frequency used to estimate the systolic blood pressure. E denotes a frequency peak value (measured simultaneously with or under a condition identical to that of B) of the second heart sound component as a reference measured in the past. F denotes a frequency peak value of the second heart sound component measured this time. γ denotes a coefficient related to a heart rate used to estimate the systolic blood pressure. G denotes a heart rate (measured simultaneously with or under a condition identical to that of B) as a reference measured in the past. H is a heart rate measured this time. Values measured when the subject is in a normal health condition are preferably used for B, E, and G. The coefficients β, γ can employ any values. To distinguish between the systolic blood pressure and the diastolic blood pressure, different values between the systolic blood pressure and the diastolic blood pressure may be used for β and γ.

[0047] The equation 2 is one example. When the fact that the peak value of the heart sound frequency and the heart rate show the positive correlation to the blood pressure value is applied, by actually measuring the peak value of the heart sound frequency and the heart rate, the blood pressure value corresponding to it can be estimated. Especially, since it is known that the blood pressure value changes due to the heart rate, the use of both of the peak value of the heart sound frequency and the heart rate as the parameters allows further accurately estimating the blood pressure value.

[0048] In the present specification, the embodiments of the present invention have been described above by referring to the drawings to express the contents of the present invention. However, the present invention is not limited to the above-described embodiments and encompasses changed configurations and improved configurations obvious for those skilled in the art based on the matters described in the present specification.

DESCRIPTION OF REFERENCE SIGNS

[0049] 10 . . . blood pressure estimating device (medical device) [0050] 11 . . . frequency analysis unit [0051] 12 . . . blood pressure actual measurement unit [0052] 13 . . . heart rate measurement unit [0053] 14 . . . data accumulation unit [0054] 15 . . . database [0055] 16 . . . learned model [0056] 17 . . . blood pressure estimating unit [0057] 18 . . . cardiac function diagnostic unit [0058] 19 . . . output unit [0059] 20 . . . digital stethoscope [0060] 30 . . . sphygmomanometer [0061] 40 . . . display device [0062] 50 . . . communication device