JP6668063B2 - Data analysis device, vein wave monitoring device, method and program - Google Patents

Description

  The present invention relates to a data analysis apparatus and method, and more particularly, to a monitoring apparatus and method for generating a venous wave signal from a microwave signal reflected from a vein of a subject.

  Conventionally, as a method of monitoring a pulse wave of a patient having a heart disease, a method using a diaphragm-type pressure transducer is known. Also, in recent years, biological signals such as a patient's heartbeat waveform and blood pressure have been measured using microwaves.

  For example, in Patent Literature 1 below, measurement of vibration such as heartbeat of a heart is performed based on the Doppler effect using microwaves. Further, Non-Patent Document 1 below discloses that a signal correlated with an arterial wave was obtained from the neck using a microwave Doppler sensor (Figs. 3-4 of Non-Patent Document 1).

JP 2014-39666 A

Misawa et al., "Non-invasive biometric measurement using non-contact microwave Doppler sensor", J. Cardiol, Vol. 20, No. 1, Page. 209-216 (1990)

  Among various heart diseases, for example, right heart hypertrophy, tricuspid stenosis, tricuspid regurgitation, and contractile pericarditis are special conditions that are difficult to diagnose. In order to diagnose these conditions, it is usually necessary to monitor the jugular venous pressure (JVP) of the internal jugular vein of the patient's neck. Generally, the internal jugular vein pressure JVP is regarded as a representative index for evaluating the right heart function of a patient's heart. However, the venous pressure is small compared to the arterial pressure, the signal level is small and it is easily buried in noise, and it is difficult to measure the internal jugular venous pressure JVP accurately enough to diagnose the pathology of heart disease Met.

  For example, in the conventional method using a diaphragm-type pressure transducer, a detection area is small, and it is necessary to specify a position with high sensitivity, and it cannot be used without a skilled doctor. The vein is located closer to the surface than the artery, and the pressure transducer is a contact-type sensor, so placing the sensor on the patient's neck would deform the blood vessels and make accurate measurement difficult. .

  Further, in Non-Patent Document 1, although a signal similar to the jugular vein wave is measured, the signal obtained by the measurement has a duplication of the carotid artery wave, and the venous wave signal and the arterial wave are accurately determined so that the disease state can be diagnosed. It has not yet been possible to extract a vein wave signal by separating it from a wave signal (see FIG. 7 of Non-Patent Document 1).

  The present invention has been made to solve the above problems, and an object of the present invention is to provide a data analysis device and a data analysis method for accurately estimating a venous wave signal from a microwave signal reflected from a vein of a subject. Is to do.

  The present invention for achieving the above object includes the following aspects. As described in Non-Patent Document 2, according to the study by the present inventors, there is a high correlation between the speed of minute movement of the body surface due to the heartbeat and the non-contact heartbeat measurement by microwave. (S. Suzuki, et al., “An approach to remote monitoring of heart rate variability (HRV) using microwave radar during a calculation task”, Journal of Physiological Anthropology 30 (6), 241-249, 2011.). That is, according to the configuration described below, a reliable waveform of the venous wave signal can be estimated without directly contacting the subject.

(Claim 1)
A data analysis device including a waveform processing unit that generates a vein wave signal of the vein from a microwave signal reflected from a vein of the subject,
The waveform processing unit,
By performing an autocorrelation process on the microwave signal reflected from the vein of the subject, a reconstruction unit for reconstructing the arterial wave signal of the subject,
An estimating unit that estimates a vein wave signal of the subject based on the microwave signal and the arterial wave signal,
Data analysis device including:

(Claim 2)
The reconstruction unit,
An autocorrelation processing unit that performs the autocorrelation processing on the microwave signal to obtain an autocorrelation waveform,
A waveform cutout unit that cuts out a predetermined section of the autocorrelation waveform and obtains a representative waveform representing the microwave signal in the section,
By connecting the representative waveform obtained for each section in the time axis direction, a waveform connection unit for reconstructing the arterial wave signal,
Item 1. The data analysis device according to item 1, comprising:

(Claim 3)
The estimator,
A difference calculation unit that calculates a difference between the normalized microwave signal and the arterial wave signal, and obtains a candidate signal that is a candidate for the vein wave signal.
Based on the phase difference between the candidate signal and the arterial wave signal, a phase difference determination unit that determines the quality of the candidate signal,
Including
3. The data analysis device according to item 1 or 2, wherein the candidate signal determined to be good is estimated as the venous wave signal.

(Claim 4)
An aperiodic component cut filter unit that determines an aperiodic signal component by performing an independent component analysis on the microwave signal and removes the aperiodic signal component from the microwave signal. The data analyzer according to any one of Items 1 to 3.

(Claim 5)
Item 5. The data analyzer according to any one of Items 1 to 4, wherein the vein is an internal jugular vein.

(Claim 6)
Item 6. The data analysis device according to any one of Items 1 to 5, further comprising a disease determination unit that determines whether or not the subject has a disease based on the venous wave signal.

(Claim 7)
Item 7. A program for causing a computer to function as each unit of the data analysis device according to any one of Items 1 to 6.

(Claim 8)
An oscillator for generating microwaves,
An antenna that emits the microwave into a subject's vein,
A receiver for receiving the microwave reflected on the vein,
Items of the data analysis device according to any one of Items 1 to 6,
A vein wave monitoring device including the device.

(Claim 9)
A data analysis method for generating a vein wave signal of a vein from a microwave signal reflected from a vein of the subject,
Reconstructing the subject's arterial wave signal by performing autocorrelation processing on the microwave signal reflected from the subject's vein,
Estimating a venous wave signal of the subject based on the microwave signal and the arterial wave signal,
Data analysis method including.

(Claim 10)
The reconfiguring comprises:
Performing the autocorrelation process on the microwave signal to obtain an autocorrelation waveform,
Cutting out a predetermined section of the autocorrelation waveform, and acquiring a representative waveform representing the microwave signal in the section,
Concatenating the representative waveform acquired for each section in the time axis direction, and reconstructing the arterial wave signal;
Item 10. The data analysis method according to Item 9, comprising:

(Claim 11)
The estimating step includes:
Calculating a difference between the normalized microwave signal and the arterial wave signal to obtain a candidate signal that is a candidate for the venous wave signal;
Based on the phase difference between the candidate signal and the arterial wave signal, determining the quality of the candidate signal,
Estimating the candidate signal determined to be good as the venous wave signal,
11. The data analysis method according to item 9 or 10, wherein

(Claim 12)
Clauses 9 to 11 further comprising: performing an independent component analysis on the microwave signal to determine an aperiodic signal component and removing the aperiodic signal component from the microwave signal. The data analysis method according to any of the above.

(Claim 13)
Item 13. The data analysis method according to any one of Items 9 to 12, wherein the vein is an internal jugular vein.

(Claim 14)
Item 14. The data analysis method according to any one of Items 9 to 13, further comprising a step of determining whether the subject has a disease based on the venous wave signal.

(Clause 15)
Generating a microwave;
Emitting the microwave into a subject's vein;
Receiving the microwave reflected at the vein;
Each step of the data analysis method according to any one of Items 9 to 14,
A vein wave monitoring method including:

  According to the present invention, an autocorrelation process is performed on a microwave signal reflected from a vein of a subject to reconstruct an arterial wave signal, and a venous wave signal is estimated based on the microwave signal and the arterial wave signal. ing. Therefore, an arterial wave signal having a high signal intensity is mixed in the microwave signal reflected from the vein, but according to the present invention, the signal intensity is lower in the presence of such an arterial wave signal mixing. The venous wave signal can be appropriately separated and extracted.

FIG. 1 is a schematic diagram showing a schematic configuration of a vein wave monitoring device 10 according to first and second embodiments of the present invention. 6 is an exemplary sample waveform of each signal obtained when the waveform processing unit 5 performs data processing. 5 is a flowchart illustrating a data processing procedure performed by a data analysis unit 12; FIG. 3 is a block diagram illustrating a configuration and a data flow of a waveform processing unit 5. 5 is a flowchart illustrating a procedure of an arterial wave reconstruction process performed by an arterial wave reconstruction unit 52. FIG. 9 is a schematic diagram for explaining an arterial wave reconstruction process performed by an arterial wave reconstruction unit 52. 6 is a flowchart illustrating a procedure of a vein wave estimation process performed by a vein wave signal estimation unit 53. FIG. 11 is a block diagram illustrating a configuration and a data flow of a waveform processing unit 5 ′ according to a second embodiment of the present invention. It is a schematic diagram showing a schematic configuration of a vein wave monitoring device 10 ′ according to a third embodiment of the present invention. FIG. 4 is a schematic diagram showing an exemplary waveform of a venous pressure JVP of the internal jugular vein. FIG. 3 is a block diagram illustrating a configuration and a data flow of a disease determination unit 7. FIG. 7 is a schematic diagram for explaining an example of a determination process performed by a disease presence / absence determination unit 72.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description and drawings, the same reference numerals indicate the same or similar components, and therefore, the description of the same or similar components will be omitted.

(First Embodiment)
In the first embodiment of the present invention, a venous wave signal is estimated from a microwave signal reflected from a vein of a subject.

<Schematic configuration of vein wave monitor>
FIG. 1 is a schematic diagram showing a schematic configuration of a venous wave monitoring device 10 according to the first and second embodiments of the present invention. The vein wave monitoring device 10 includes a microwave radar device 11 and a data analysis unit 12. The microwave radar device 11 irradiates a microwave to a vein of a subject (not shown) and receives the microwave reflected from the vein. The data analysis unit 12 processes the signal received by the microwave radar device 11 to estimate and present the waveform of the venous wave signal. The microwave radar device 11 is, for example, a known microwave radar module of a Doppler system, and the data analysis unit 12 is, for example, a known personal computer.

  The microwave radar device 11 includes an antenna 1, a microwave oscillation / receiver 2, an amplifier 3, and an analog-digital converter 4. The microwave oscillation / receiver 2 generates a microwave and receives the microwave reflected in the vein of the subject in an analog form. The microwave is emitted to the vein of the subject through the antenna 1. The amplification unit 3 amplifies the microwave signal received by the microwave oscillation / receiver 2. The analog-to-digital converter 4 converts the amplified microwave signal in analog format to digital format.

  The data analysis unit 12 includes the waveform processing unit 5. Further, the data analysis unit 12 can optionally include the display unit 6.

  The microwave received by the microwave oscillation / receiver 2 includes information such as a frequency change, a phase change, or an intensity change of the microwave reflected on the vein. The vein wave signal is estimated from the microwave signal by performing data processing according to the estimation principle. The display unit 6 displays the waveform of the estimated venous wave signal.

  When implemented using a known personal computer, the data analysis unit 12 includes, for example, a CPU, a memory, a hard disk device, an input / output interface, and the like. The CPU uses the memory as a work area, and records data necessary for processing (eg, intermediate data during processing) in a memory or a hard disk device as appropriate. In the present embodiment, a program for performing processing to be described later is recorded in the hard disk device in advance in, for example, an execution format (for example, generated by converting from a programming language by a compiler), and the data analysis unit 12 includes By reading the program recorded in the hard disk device into the memory, it functions as the waveform processing unit 5 and performs the processing described later. The program may be installed in the data analysis unit 12 from a computer-readable and non-temporary tangible recording medium such as a CD-ROM, or the data analysis unit 12 may be connected to the Internet (not shown). Then, the program code of the program may be downloaded via the Internet. Note that the waveform processing unit 5 may be realized by hardware such as a logic circuit.

<Principle of vein wave estimation>
FIG. 2 shows an exemplary sample waveform of each signal obtained when the waveform processing unit 5 performs data processing. In estimating the vein wave to be monitored from the microwave signal reflected in the vein of the subject, the microwave signal Sraw obtained by the microwave radar device 11 includes a signal Svein related to the vein wave, It is assumed to include information on a signal Sartery related to arterial waves, a noise signal Nartifact related to body movement of the subject, and a disturbance or other noise signal Ndisturbance.

The relationship between signal strengths is assumed as follows.

  Here, the artery wave signal Sartery and the vein wave signal Svein are signals originating from the heartbeat, and are both signals having periodicity. Also, as can be understood from the blood circulation process, the arterial wave signal has a greater signal intensity than the venous wave signal. Therefore, the arterial wave signal can be distinguished from the venous wave signal by the difference in signal strength.

  In addition, since the subject is in a stationary state at the time of monitoring, the noise signal Nartifact related to the body movement does not basically occur. Furthermore, among the body movements of the subject, the respiratory activity is random and non-periodic, and therefore, the noise signal Nartifact related to the body movement can be distinguished from the arterial wave signal and the venous wave signal depending on the presence or absence of the periodicity. It is possible. Even if periodicity is recognized in the respiratory activity, since the body motion due to the respiratory activity has a lower frequency than the cycle of the heartbeat, these can be distinguished by an arbitrary filter circuit. In addition, since the noise signal Ndisturbance related to disturbance or the like does not have periodicity, it can be distinguished from the arterial wave signal and the venous wave signal depending on the presence or absence of the periodicity.

  In the present invention, focusing on the point that both the arterial wave signal and the venous wave signal have periodicity and the point that the arterial wave signal has a higher signal strength than the venous wave signal, The venous wave signal is estimated from the signal. That is, for the arterial wave signal having a high signal intensity, the arterial wave signal Sartery is estimated by signal reconstruction using the autocorrelation function, and the difference between the estimated arterial wave signal Sartery and the estimated signal Sraw of the microwave, which is the original signal, The vein wave signal Svein is estimated. Here, the noise signal Nartifact related to body motion and the noise signal Ndisturbance related to disturbance have no periodicity and can be decorrelated by superimposition. Appropriately removed by configuration.

<Procedure for estimating venous waves>
FIG. 3 is a flowchart illustrating a data processing procedure performed by the data analysis unit 12, and FIG. 4 is a block diagram illustrating a configuration and a data flow of the waveform processing unit 5.

  As shown in FIG. 4, the waveform processing unit 5 includes a noise filter 51, an arterial wave reconstruction unit 52, a normalization / difference calculation unit 53a, and a phase difference determination unit 53b, and these units cooperate with each other. To estimate the venous wave signal. The normalization / difference calculation unit 53a and the phase difference determination unit 53b constitute a vein wave signal estimation unit 53. The noise filter 51 has an arbitrary configuration. These components constituting the waveform processing unit 5 are functional blocks realized by causing the data analysis unit 12 to execute the program according to the present invention, and include a noise filter 51, an arterial wave reconstruction unit 52, and a vein wave signal. Each of the estimating units 53 corresponds to each of steps S1 to S3 described below. Alternatively, the noise filter 51, the arterial wave reconstruction unit 52, the normalization / difference calculation unit 53a, and the phase difference determination unit 53b included in the waveform processing unit 5 may be realized by hardware.

  First, in step S1, a noise filter is applied to the microwave signal Sraw which is the original signal. The noise filter 51 is, for example, a high-pass filter, and removes a noise signal Nartifact related to body motion from the microwave signal Sraw. The noise signal Nartifact related to body motion is, for example, the respiratory activity of the subject and has a lower frequency than the pulse wave signal, and thus is appropriately removed from the microwave signal. Hereinafter, the microwave signal after the application of the noise filter is referred to as Sfltd. Since the noise filter 51 has an arbitrary configuration, the microwave signal Sfltd to which the noise filter has been applied is appropriately replaced with the microwave signal Sraw when the noise filter is not applied in the subsequent steps.

  Next, in step S2, an autocorrelation process is performed on the microwave signal Sfltd to reconstruct the arterial wave signal Sartery, and in step S3, based on the microwave signal Sfltd and the reconstructed artery wave signal Sartery. Then, the vein wave signal Svein is estimated. Specific processing procedures of the reconstruction processing in step S2 and the estimation processing in step S3 will be described in more detail with reference to FIGS.

  Finally, in step S4, the waveform of the estimated vein wave signal Svein is presented. The waveform of the vein wave signal Svein is displayed on the display unit 6, for example.

FIG. 5 is a flowchart showing a procedure of an arterial wave reconstruction process (step S2) performed by the arterial wave reconstruction unit 52, and FIG. It is a schematic diagram for demonstrating a wave reconstruction process.

In step S21, an autocorrelation waveform is obtained by performing an autocorrelation process on the microwave signal Sfltd. The correlation function is a function expressing the similarity (correlation degree) between two different waveforms as a function of the shifted time (delay time τ) while shifting one of the waveforms on the time axis. The autocorrelation process is a process of obtaining such a correlation function for one signal waveform, and thereby, a repetitive portion (ie, periodicity) in one signal waveform can be obtained. The autocorrelation waveform is obtained by the following equation, for example.

  Here, f (t) is the original waveform, which is the microwave signal Sfltd shown in FIG. n is the number of data to be subjected to the autocorrelation analysis, and is the number of reference data in the area indicated by the symbol L in FIG. If the data in the area indicated by the symbol L is, for example, data of 10 seconds at a sampling rate of 1 k / s, n = 10000. k is a scanning amount. For example, if sampling is 1 k / s, k = 1 msec. The scan amount is also called a lag. μ is the average value of the original waveform f (t), that is, the average value of the microwave signal Sfltd. μ is introduced to make the offset 0 (zero). σ is the standard deviation of the original waveform f (t), that is, the standard deviation of the microwave signal Sfltd. Normalization is performed by dividing by σ. Since μ and σ are simply normalized, they have an arbitrary configuration in the equation, and an equation excluding μ and σ may be used.

  The processing in this step will be described more specifically with reference to FIG. First, of the microwave signal Sfltd shown in FIG. 6A which is the original waveform, attention is paid to a section L (for example, 10 seconds) in which the waveform is considered to have reproducibility. When the autocorrelation processing is performed on the signal Sfltd, an autocorrelation waveform shown in FIG. 6B is obtained. The horizontal axis of the graph shown in FIG. 6B is the delay time τ, and the vertical axis is the degree of correlation.

In step S22, a predetermined section of the autocorrelation waveform is cut out to obtain a representative waveform representing the microwave signal Sfltd in the section. For example of the autocorrelation waveform shown in FIG. 6 (b), by cutting the second unit waveform, a representative waveform Rep1 interval l ₁ signal Sfltd microwave.

In step S23, the representative waveform acquired for each section is connected in the time axis direction to reconstruct the arterial wave signal Sartery. After the processing in step S22, shifting the interval L of interest only sections l _1, to newly set the interval L of interest. Then, the processing of steps S21 to S22 is repeatedly performed for the newly set section L. This gives a representative waveform Rep2 new interval l _2. Similarly thereafter, shifted by further interval l ₂ a section L of interest, and newly set the interval L of interest, a plurality of representative waveforms RepN representing each interval _{l N (N = 1, 2} , 3, .. To get.) A plurality of representative waveforms RepN acquired for each section are connected in the time axis direction to reconstruct an arterial wave signal Sartery.

  Upon completion of the process in the step S23, the vein wave signal estimating process in the step S3 is continuously performed.

Estimation of the vein wave signal Svein FIG. 7 is a flowchart showing the procedure of the vein wave estimation process (step S3) performed by the vein wave signal estimation unit 53.

  First, in step S31, a difference between the microwave signal Sfltd and the reconstructed arterial wave signal Sartery is calculated. Since the signal level (magnitude of amplitude) is different between the microwave signal Sfltd and the reconstructed arterial wave signal Sartery, it is necessary to normalize the signal level before calculating the difference between these signals. desirable. A signal obtained by calculating the difference is a signal Sveincd that is a candidate for a venous wave signal.

  In step S32, the quality of the candidate signal Sveincd is determined based on the phase difference between the candidate signal Sveincd obtained from the difference and the reconstructed arterial wave signal Sartery. Considering the blood circulation process, it is considered that a phase difference occurs between the arterial wave and the venous wave in principle. That is, when focusing on a single heartbeat, it is considered that a phase difference of a predetermined magnitude occurs between the pulse wave in the artery and the pulse wave in the vein generated by the heartbeat. In this step, based on this phase difference, it is determined whether or not the candidate signal Sveincd is appropriate as a venous wave (for example, “good” or “not good”). As a method of determining the phase difference, for example, a peak interval between two signals may be used as a phase difference, or a cross-correlation between two signals may be obtained and a shift (lag) may be used as a phase difference.

  In step S33, the candidate signal Sveincd determined to be “good” in step S32 is adopted as an estimated signal of the vein wave Svein. The estimated waveform of the vein wave signal Svein is displayed on, for example, the display unit 6 in step 4.

(Second embodiment)
In the second embodiment of the present invention, the data analysis unit 12 includes a waveform processing unit 5 'instead of the waveform processing unit 5. The waveform processing unit 5 'uses a non-periodic component cut filter 51' for removing non-periodic signal components from the microwave signal Sraw, instead of the noise filter 51 using a high-pass filter.

  FIG. 8 is a block diagram showing the configuration and data flow of the waveform processing unit 5 'according to the second embodiment of the present invention. As shown in FIG. 8, the waveform processing unit 5 'according to the second embodiment includes an aperiodic component cut filter 51' instead of the noise filter 51 of the first embodiment. The aperiodic component cut filter 51 'includes an independent component analyzer 51'a and an aperiodic component determiner 51'b.

  When the microwave radar device 11 is a Doppler system, the output signal Sraw of the microwave radar device 11 is separated into two components, an I signal having an in-phase component and a signal having a quadrature component. Can be expressed as These two signals, the I signal and the Q signal, are input to the independent component analyzer 51'a, and the independent component analyzer 51'a performs independent component analysis. By the independent component analysis, the input two signals of the I signal and the Q signal are separated into two signal components of a non-periodic noise signal component and a component including the arterial wave signal Sartery and the vein wave signal Svein. Is done. The non-periodic component determination unit 51′b selects a component including the arterial wave signal Sartery and the venous wave signal Svein from the two signal components separated by the independent component analysis, and selects the microwave after applying the noise filter. Is output as the signal Sfltd. As a method of determining a component including the arterial wave signal Sartery and the venous wave signal Svein, for example, the determination is performed using a frequency component or an eigenvalue. When a frequency component is used, a signal component whose peak can be confirmed around 0.8 to 3 Hz (preferably 0.8 to 2 Hz), which is a heartbeat component at rest, is generally determined by frequency analysis. It is determined that the component does not include the component of the periodic noise signal and includes the arterial wave signal Sartery and the vein wave signal Svein.

  A series of processes of the arterial wave reconstruction unit 52 and the vein wave signal estimation unit 53 (the normalization / difference calculation unit 53a and the phase difference determination unit 53b) after the application of the aperiodic component cut filter 51 'are shown in FIG. 7 is the same as the data processing procedure shown in FIG. 7, and a detailed description thereof will be omitted.

(Third embodiment)
FIG. 9 is a schematic diagram showing a schematic configuration of a vein wave monitoring device 10 'according to the third embodiment of the present invention.

  In the third embodiment of the present invention, the data analysis unit 12 'further includes a disease determination unit 7, and the disease determination unit 7 determines the presence or absence of a disease based on the estimated venous wave signal Svein. The disease determination unit 7 may be realized as software by the CPU of the data analysis unit 12 'executing a program, or may be realized as hardware by a logic circuit or the like. In the present embodiment, a case where the vein to be measured is an internal jugular vein will be described as an example. The configuration of the data analysis unit 12 'other than the disease determination unit 7 is the same as the configuration of the data analysis unit 12 according to the first embodiment or the second embodiment, and therefore, detailed description is omitted. I do.

  FIG. 10 is a schematic diagram showing an exemplary waveform of the venous pressure JVP of the internal jugular vein. As shown in FIG. 10, a typical internal jugular vein JVP waveform has a characteristic peak or valley described below.

・ A-wave Pressure rise due to transient venous extension due to contraction of right atrium ・ c-wave Pressure when tricuspid valve closes ・ v-wave From systemic circulation when tricuspid valve closes during systole to right atrium Pressure change due to venous regurgitation to the valley ・ Following x-valley c-wave, pressure drop caused by downward movement of the tricuspid valve due to contraction of the right ventricle ・ Y-valley dilatation The tricuspid valve opens early and quickly from the right atrium to the right ventricle Decreased right atrial pressure due to outflow of blood

  In the present embodiment, the presence / absence of a disease is determined based on these feature amounts regarding the JVP waveform. For example, when a remarkable a-wave exists in the estimated vein wave signal Svein of the subject, it is determined that there is a possibility of right heart hypertrophy or tricuspid valve stenosis. When a remarkable v-wave exists, it is determined that there is a possibility of tricuspid regurgitation. If there is a significant y-valley, it is determined that there is a possibility of contractile pericarditis. Furthermore, if the JVP value itself is increasing, it is determined that there is a possibility of heart failure, tricuspid valve disease, pulmonary artery stenosis, or pericardial disease as a condition in which the right heart pressure increases, and the JVP value itself is increased. If is decreased, it is determined that there is a possibility that the amount of circulating blood such as dehydration has decreased.

  FIG. 11 is a block diagram illustrating a configuration and a data flow of the disease determination unit 7. As shown in FIG. 11, the disease determination unit 7 includes a feature amount extraction unit 71 and a disease presence / absence determination unit 72.

  The feature amount extraction unit 71 includes a peak detection unit 71a. Further, the feature amount extraction unit 71 can optionally include a parameter calculation unit 71b and a statistical value calculation unit 71c. The vein wave signal Svein estimated by the vein wave signal estimation unit 53 is input to the peak detection unit 71a, and the positions of the a-wave, c-wave, v-wave, x-valley, and y-valley in the JVP waveform and the respective peak heights (Hereinafter, referred to as a feature amount). For example, when the first-order differentiation of the JVP waveform shown in FIG. 10 is performed, a waveform of a pressure change amount (pressure change speed) per unit time can be obtained. The a-wave, c-wave, v-wave, x-valley, and y-valley in the JVP waveform correspond to each inflection point of the first-order differential waveform. That is, inflection points are determined based on the waveform of the pressure change amount, and the a-wave, c-wave, v-wave, x-wave are obtained from the position of each inflection point on the time axis and the pressure value at that time. Each position and each peak height (characteristic amount) of the valley and the y-valley can be determined.

  The parameter calculating unit 71b and the statistical value calculating unit 71c can calculate more detailed parameters and various known statistical values for each feature amount of the a-wave, c-wave, v-wave, x-valley, and y-valley. . The calculated parameters and various statistical values are input to the disease presence / absence determination unit 72, and can be used to enhance the determination accuracy of the determination process of the disease presence / absence determination unit 72 described later.

  The disease presence / absence determining unit 72 includes a comparing unit 72a and a steady value holding unit 72b. The steady value holding unit 72b stores in advance, for example, feature amount data on a typical JVP waveform of a healthy person. The comparing unit 72a compares the feature amount of the subject's JVP waveform obtained by the feature amount extracting unit 71 with the feature amount of the healthy subject's JVP waveform recorded in the steady value holding unit 72b. .

  FIG. 12 is a schematic diagram for explaining an example of the determination process performed by the disease presence / absence determination unit 72. In the figure, the waveform shown in (a) is the JVP waveform of a healthy person, and the waveform shown in (b) is the JVP waveform of the subject. In the example shown in FIG. 12, it can be confirmed that the peak position of the v-wave is shifted. In such a case, the disease presence / absence determination unit 72 determines that there is a deviation of the peak position that exceeds the threshold value of the predetermined size, and determines that there is a possibility of tricuspid regurgitation for the subject. Can be determined.

(Appendix)
As described above, the present invention has been described with reference to the specific embodiments. However, the present invention is not limited to the above-described first to third embodiments.

  In the first to third embodiments, the microwave radar device 11 of the Doppler system is used. However, the detection system of the microwave radar device 11 is not limited to this, and instead of the Doppler system, for example, FMCW A method or a pulse wave may be used. In addition, the frequency of the microwave used by the microwave radar device 11 is not limited, and if it is within the range of use restriction by the Radio Law, for example, 1 GHz, 2.4 GHz, 10.525 GHz, 24.15 GHz, 228 GHz, or the like. Frequency can be used. The output of the microwave is not limited, and a microwave of any output may be used. However, in consideration of the influence on the body of the subject, it is desirable to set the upper limit of the output according to the frequency to be used, and to set the output below the upper limit. For example, when using a microwave having a frequency of 10 GHz or more, it is preferable to suppress the output to 10 mW or less.

  In the first to third embodiments, the estimated waveform of the venous wave signal is displayed on the display unit 6 included in the data analysis unit 12, but the storage device included in the data analysis unit 12 (see FIG. (Not shown) may be stored as data. Alternatively, the data may be stored as data from the data analysis unit 12 on a server (not shown) installed in a medical institution via the Internet.

  In the first to third embodiments, the noise filter is applied to the microwave signal Sraw in step S1, but the noise filter 51 has an arbitrary configuration. That is, in the process after step S1, the microwave signal Sraw as the original signal may be used as it is instead of the microwave signal Sfltd after the application of the noise filter.

  Further, in the first or third embodiment, the noise filter 51 of the waveform processing unit 5 uses a high-pass filter. However, the noise filter 51 is not limited to a high-pass filter, and may be a low-pass filter or a band-pass filter. Can be used. For example, if a noise signal due to respiratory activity is removed, a high-pass filter that removes low-frequency noise components is used, and if a high-frequency noise component that is higher than the frequency of the heartbeat is removed, a low-pass filter may be used. A band pass filter integrating these high pass filter and low pass filter may be used.

In the above-mentioned first to third embodiments, in step S22, among the autocorrelation waveform shown in FIG. 6 (b), by cutting the second unit waveforms, interval l ₁ signal Sfltd microwave Although the representative waveform is Rep1, the unit waveform to be cut out is not limited to the second and may be the first or the third.

  In the third embodiment, the case where the vein to be measured is an internal jugular vein is described as an example. However, the vein to be measured is not limited to the internal jugular vein, and all veins of the subject are measured. Can be targeted. The diseases for which the disease determination unit 7 determines the presence / absence also include right heart hypertrophy, tricuspid stenosis, tricuspid regurgitation, constrictive pericarditis, heart failure, tricuspid valve disease, pulmonary artery stenosis, pericardial disease, dehydration, tricuspid The disease is not limited to valve insufficiency or the like, and can be any disease as long as it can be determined based on a change in a venous wave signal.

  In the third embodiment, the disease presence / absence determination unit 72 determines the possibility of the presence or absence of a disease in the subject based on the shift of the peak position of the v-wave. The feature value is not limited to the v-wave, and is not limited to the shift of the peak position. The disease presence / absence determination unit 72 appropriately determines the positions of the a-wave, c-wave, v-wave, x-valley, and y-valley, and the respective peak heights (feature amounts) based on various known diagnostic methods for the JVP waveform. Can be used for judgment. Also, the feature quantity used for the determination is not limited to the waveform of the pressure change obtained by the first-order differentiation of the JVP waveform shown in FIG. 10, but the change of the pressure change rate per unit time (pressure) obtained by the second-order differentiation. The waveform of the change acceleration) may be calculated and used.

  As described above, according to the present invention, an autocorrelation process is performed on a microwave signal reflected from a vein of a subject to reconstruct an arterial wave signal, and a venous wave signal is generated based on the microwave signal and the arterial wave signal. Estimated. Therefore, an arterial wave signal having a high signal intensity is mixed in the microwave signal reflected from the vein, but according to the present invention, the signal intensity is lower in the presence of such an arterial wave signal mixing. The venous wave signal can be appropriately separated and extracted.

  Further, since the microwave radar device 11 can measure microwaves in a non-invasive and non-contact manner, it is possible to more easily observe a venous wave signal as compared with a conventional pressure transducer which is a contact type sensor. Becomes possible.

  In addition, since the use of the microwave radar device 11 does not require a high level of skill by the user, it is possible to observe the venous wave signal more easily than before in this regard. Since the measurement of the microwave and the estimation of the venous wave signal can be performed in real time, the venous wave signal can be observed in real time, which is useful for, for example, daily medical care of a cardiovascular system by a medical worker.

DESCRIPTION OF SYMBOLS 1 Antenna 2 Microwave oscillation / receiver 3 Amplification part 4 Analog-digital conversion part 5, 5 'Waveform processing part 6 Display part 7 Disease determination part 10, 10' Vein wave monitor apparatus 11 Microwave radar apparatus 12, 12 'Processing Unit 51 noise filter 51 'aperiodic component cut filter 51'a independent component analyzer 51'b aperiodic component determination unit 52 arterial wave reconstruction unit 53 venous wave signal estimation unit 53a normalization / difference calculation unit 53b phase difference Determination unit 71 Feature extraction unit 71a Peak detection unit 71b Parameter calculation unit 71c Statistical value calculation unit 72 Disease presence determination unit 72a Comparison unit 72b Steady value holding unit
Sraw microwave signal
Svein vein signal
Sartery Arterial wave signal
Nartifact noise signal (body motion related)
Ndisturbance noise signal (disturbance, etc.)
Sfltd Microwave signal after applying noise filter
Sveincd vein wave candidate signal

Claims (14)

A data analysis device including a waveform processing unit that generates a vein wave signal of the vein from a microwave signal reflected from a vein of the subject,
The waveform processing unit,
By performing an autocorrelation process on the microwave signal reflected from the vein of the subject, a reconstruction unit for reconstructing the arterial wave signal of the subject,
An estimating unit that estimates a vein wave signal of the subject based on the microwave signal and the arterial wave signal,
Data analysis device including: The reconstruction unit,
An autocorrelation processing unit that performs the autocorrelation processing on the microwave signal to obtain an autocorrelation waveform,
A waveform cutout unit that cuts out a predetermined section of the autocorrelation waveform and obtains a representative waveform representing the microwave signal in the section,
By connecting the representative waveform obtained for each section in the time axis direction, a waveform connection unit for reconstructing the arterial wave signal,
The data analysis device according to claim 1, comprising: The estimator,
A difference calculation unit that calculates a difference between the normalized microwave signal and the arterial wave signal, and obtains a candidate signal that is a candidate for the vein wave signal.
Based on the phase difference between the candidate signal and the arterial wave signal, a phase difference determination unit that determines the quality of the candidate signal,
Including
The data analysis device according to claim 1, wherein the candidate signal determined to be good is estimated as the vein wave signal.   An aperiodic component cut filter unit that determines an aperiodic signal component by performing an independent component analysis on the microwave signal and removes the aperiodic signal component from the microwave signal. The data analysis device according to any one of claims 1 to 3.   The data analysis device according to claim 1, wherein the vein is an internal jugular vein.   The data analysis device according to any one of claims 1 to 5, further comprising a disease determination unit that determines whether the subject has a disease based on the venous wave signal.   A program for causing a computer to function as each unit of the data analysis device according to claim 1. An oscillator for generating microwaves,
An antenna that emits the microwave into a subject's vein,
A receiver for receiving the microwave reflected on the vein,
Each part of the data analysis device according to any one of claims 1 to 6,
A vein wave monitoring device including the device. A data analysis method for generating a vein wave signal of a vein from a microwave signal reflected from a vein of the subject,
Reconstructing the subject's arterial wave signal by performing autocorrelation processing on the microwave signal reflected from the subject's vein,
Estimating a venous wave signal of the subject based on the microwave signal and the arterial wave signal,
Data analysis method including. The reconfiguring comprises:
Performing the autocorrelation process on the microwave signal to obtain an autocorrelation waveform,
Cutting out a predetermined section of the autocorrelation waveform, and acquiring a representative waveform representing the microwave signal in the section,
Concatenating the representative waveform acquired for each section in the time axis direction, and reconstructing the arterial wave signal;
The data analysis method according to claim 9, comprising: The estimating step includes:
Calculating a difference between the normalized microwave signal and the arterial wave signal to obtain a candidate signal that is a candidate for the venous wave signal;
Based on the phase difference between the candidate signal and the arterial wave signal, determining the quality of the candidate signal,
Estimating the candidate signal determined to be good as the venous wave signal,
The data analysis method according to claim 9, comprising:   The method according to claim 9, further comprising performing independent component analysis on the microwave signal to determine a non-periodic signal component, and removing the non-periodic signal component from the microwave signal. The data analysis method according to any one of the above.   The data analysis method according to claim 9, wherein the vein is an internal jugular vein. Generating a microwave;
Emitting the microwave into a subject's vein;
Receiving the microwave reflected at the vein;
Each step of the data analysis method according to any one of claims 9 to 13 ,
A vein wave monitoring method including:

Images