JP2002078695A - Electrocardiogram measuring instrument - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an electrocardiogram measuring instrument capable of efficiently utilizing clinical data of electrocardiogram by a conventional standard 12 induction method by removing only base line oscillation effectively while holding useful cardiogram waveform information in an electrocardiogram for bathtub. SOLUTION: A signal processing means has a subtracter 18, a magnification converter 20 and an integrator 19. An electrocardiogram signal 42 to be processed is input into the positive side input of the subtracter 18, and then a signal is input into the negative side input of the subtracter, which is obtained by performing either of a magnification conversion processing by a magnification converter 20 or an integration processing by a integrator 19 at first and subsequently performing the other for the output signal from the subtracter 18, thereby an electrocardiogram signal 43 in which the base line oscillation is removed from the signal 42 is output from the subtracter 18.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for improving the accuracy of an electrocardiogram measurement in an electrocardiograph such as a bathtub electrocardiograph having an electrode structure in which an electrocardiogram easily fluctuates in a baseline.

[0002]

2. Description of the Related Art In a medical field and physiology research, an electrocardiogram measuring method called a standard 12-lead method is widely used, and enormous amounts of clinical data are accumulated. Furthermore, in order to accumulate the measured ECG as clinical data, it is necessary for the ECG that is the measurement result to have a certain quality, so that errors and measurement variations due to the measurement environment and the ECG that is the measuring device are minimized. In order to limit the size of the electrode, there are fine restrictions on the electrode mounting position and the electrode material.

For example, an expensive Ag-AgCl electrode is used for an electrode material. This has the effect of suppressing the fluctuation of the baseline of the electrocardiogram by preventing the body surface potential from fluctuating due to the electrochemical reaction occurring between the electrodes on the body surface and the sweat (salt). Accordingly, errors and measurement variations due to the influence of sweat on the body surface of the subject can be suppressed to a minimum.
Further, noise removal in the case of processing the body surface potential signal (electric signal) appearing on the electrode can be performed by simple filter processing.

[0004] Further, due to the minute restrictions on the electrode mounting position, such electrode mounting work is extremely difficult for persons other than medical staff. This means
Although the electrocardiogram measurement is a useful method for diagnosing a heart disease and evaluating the degree of relaxation and physical condition by analyzing the electrocardiogram, it is a factor that significantly limits the industrial use range.

On the other hand, a bathtub electrocardiograph has been developed as a means for measuring a subject's electrocardiogram in a non-invasive and non-invasive manner without having to directly attach electrodes to the subject. In this bathtub electrocardiograph, a plurality of electrodes are arranged on the inner wall surface of the bathtub where the subject takes a bath, and by utilizing the relatively good electrical conductivity of hot and cold water in the bathtub, the electrodes are directly placed on the body surface of the subject. It is possible to continuously measure the electrocardiogram of the subject in the bathtub without requiring any labor for wearing.

[0006]

However, in the above bathtub electrocardiograph, since hot and cold water is interposed between the electrode and the surface of the test subject, the electrocardiogram shows not only the motion of the test subject but also the movement of the baseline due to breathing. A large baseline sway due to sway or the like occurs, which is a lead method different from the standard 12-lead method, and the obtained electrocardiogram waveform is different from the electrocardiogram waveform obtained by the standard 12-lead method. As a result, it is not possible to compare with the enormous amount of clinical data collected and accumulated by clinical medicine and physiology researchers, and the inconvenience that the clinical data cannot be used effectively for diagnosis of heart disease and evaluation of relaxation and physical condition. there were.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and has as its object to provide useful electrocardiogram waveform information in an electrocardiogram measuring apparatus such as a bathtub electrocardiograph which can cause a large baseline sway. An object of the present invention is to provide an electrocardiogram measurement apparatus capable of effectively utilizing clinical data of an electrocardiogram by a conventional standard 12-lead method by effectively removing only the base line sway while holding the data.

[0008]

A first characteristic configuration of an electrocardiogram measuring apparatus according to the present invention for achieving this object is a plurality of electrocardiographic electrodes as described in claim 1 of the claims. An electrocardiogram signal generating means for generating and processing an electrocardiogram signal from electric signals appearing on the plurality of electrocardiogram electrodes; and a signal processing for performing a predetermined signal processing on the processed electrocardiogram signal generated by the electrocardiogram signal generating means. Means, the signal processing means has a subtractor, a magnification converter and an integrator, inputs the processed electrocardiogram signal to the positive side input of the subtractor, and for the output signal from the subtractor By inputting a signal obtained by performing one of the magnification conversion processing by the magnification converter and the integration processing by the integrator first and the other to the negative input of the subtractor, Processing electrocardiogram Certain ECG signal has been removed or reduced baseline upset from the signal at the point that is configured to output from said subtractor.

Here, the processed electrocardiogram signal to be processed by the signal processing means may be an electrocardiogram signal generated by the electrocardiogram signal generation means, or may be a predetermined pre-processing for the electrocardiogram signal, for example, It may be an electrocardiogram signal that has been subjected to noise removal processing and the like performed by a conventional electrocardiogram measurement device. Further, the pre-processing may be performed by the electrocardiogram signal generating means.

[0010] A second feature of the present invention is that an electrocardiogram signal generator for generating an electrocardiogram signal from a plurality of electrocardiogram electrodes and an electric signal appearing on the plurality of electrocardiogram electrodes, as described in claim 2 of the claims. Means, and signal processing means for performing predetermined signal processing on the processed electrocardiogram signal generated by the electrocardiogram signal generation means, wherein the signal processing means comprises three rectangular waves having a unit amplitude. The time ratio of each polarity is arranged to be 1: 2: 1 in the order of negative polarity / positive polarity / negative polarity or positive polarity / negative polarity / positive polarity, and the time width of three rectangular waves corresponds to the electrocardiogram signal to be processed. A cross-correlation calculation unit that performs a cross-correlation process between a template waveform set to have a time width of one or more characteristic signal components included therein and the electrocardiogram signal to be processed is provided.

Here, the processed electrocardiogram signal to be processed by the signal processing means may be an electrocardiogram signal generated by the electrocardiogram signal generation means, or may be a predetermined pre-processing, for example, It may be an electrocardiogram signal which has been subjected to noise removal processing performed by a conventional electrocardiogram measurement apparatus or baseline sway removal or reduction processing by the signal processing means of the first characteristic configuration. Further, the pre-processing may be performed by the electrocardiogram signal generating means.

[0012] In the third feature configuration, in addition to the second feature configuration, the cross-correlation output of the cross-correlation calculation unit may be equal to or more than a predetermined threshold value. The present invention is characterized in that an occurrence time phase detecting means for detecting the occurrence time phase of the characteristic signal component is provided based on the local peak detection time.

[0013] The fourth characteristic configuration is that, in addition to the second or third characteristic configuration, the characteristic signal component is a QRS complex in addition to the second or third characteristic configuration. It is in.

According to a fifth aspect of the present invention, in addition to the second, third or fourth aspect of the present invention, the signal processing means includes the above-mentioned signal processing means. The point is that a predetermined trigger signal is output when a signal component is continuously detected a predetermined number of times or more.

The sixth characteristic configuration is that, as described in claim 6 of the claims, at least three electrocardiogram electrodes separately installed at predetermined positions on the inner wall surface of the bathtub where the subject takes a bath, An electrocardiogram signal generating means for generating an electrocardiogram signal as a pseudo standard limb lead from electric signals appearing on at least three electrocardiogram electrodes, wherein each of the electrocardiogram electrodes has a first electrode at the base of the right arm of the subject. The second electrode is located at a position near the outside of the left base of the subject's left arm, and the third electrode is located at a position near the outside of the left base of the subject's left foot. The difference signal components of the electric signal between the first electrode and the second electrode, between the first electrode and the third electrode, and between the second electrode and the third electrode are respectively Differential amplifier to amplify separately It lies in the fact to be equipped with.

The seventh characteristic configuration is that, as described in claim 7 of the claims, at least three electrocardiogram electrodes separately installed at predetermined positions on the inner wall surface of the bathtub where the subject takes a bath, An electrocardiogram signal generating means for generating an electrocardiogram signal as a pseudo standard limb lead from electric signals appearing on at least three electrocardiogram electrodes, wherein each of the electrocardiogram electrodes has a first electrode at the base of the right arm of the subject. A second electrode is located at a position close to the outside, near the outside of the left base of the subject's left arm, and a third electrode is located at a position near the outside of the subject's left knee and at the same depth as the second electrode. Wherein the electrocardiogram signal generating means is provided between the first electrode and the second electrode, between the first electrode and the third electrode, and between the second electrode and the third electrode. Signal difference between the electrical signals The lies in comprising a differential amplifying means for amplifying each other.

According to an eighth feature of the present invention, in addition to the sixth or seventh feature of the present invention, the at least three electrocardiogram electrodes include the first electrocardiogram electrode. In addition to the second and third electrodes, a fourth electrode disposed at a position symmetrical with respect to the third electrode on the inner wall surface of the bathtub with respect to the body of the subject in the bathtub. Wherein the differential amplifying means comprises three independent differential amplifiers, and the reference potentials of the three differential amplifiers are all set to the potential of the fourth electrode.

According to a ninth feature of the present invention, in addition to the first, second, third, fourth or fifth feature of the present invention, as described in claim 9 of the claims, The present invention is characterized in that a physical condition estimating means for estimating a physical condition of a subject based on a processing result of the processing means is provided.

The operation and effect of each of the above features will be described below. According to the first feature configuration of the electrocardiogram measuring apparatus according to the present invention, the output of the subtractor of the signal processing means is fed back to the negative input of the subtractor as an estimated value of the baseline sway after the magnification conversion processing and the integration processing. Thus, the baseline sway is removed from the ECG signal including the baseline sway by the subtractor. As a result, useful electrocardiogram information included in the electrocardiogram signal, for example, a P-wave generated in connection with the movement of the heart,
The waveforms of characteristic signal components such as Q wave, R wave, S wave, T wave, QRS group, generation time phase and generation period, intervals between adjacent waves, and abnormal waveforms generated between them are held. In this state, the baseline fluctuation that hinders the detection of these useful ECG information can be effectively removed. As a result, the electrocardiogram waveform after removal of the baseline fluctuation is closer to the electrocardiogram waveform obtained by the conventional standard 12-lead method, and clinical data of the electrocardiogram obtained by the standard 12-lead method can be effectively used.

More specifically, since the subtractor having the characteristic configuration basically subtracts the integrated value of the output from the input value, the subtractor acts so as to reduce the output value. By the way, since the magnification conversion process and the integration process, which are actual feedback elements, are performed with a fixed time constant, if this time constant is set to a predetermined range (for example, about 1 to 10 seconds), a predetermined period of time is set. The waveform of the characteristic signal component having a higher frequency than that of the base line sway is smoothed along the base line sway by the above-mentioned magnification conversion processing and integration processing, and an estimated value of the latest base line sway component is obtained. Since the high-frequency characteristic signal component has been removed from the estimated value, the subtraction processing does not cause the loss of the characteristic signal component from the processed electrocardiogram signal, but results in the subtraction processing of only the base line fluctuation of the low-frequency component. . Accordingly, when the time constant is extremely short, the estimated value includes a characteristic signal component, and the original signal component may be lost in addition to the baseline fluctuation. Here, the magnification converter is used to determine the cutoff frequency of the integrator.

The input and output of the signal processing means are not limited to analog signals, but may be digital signals. Therefore, in the case of a digital signal, a subtractor,
Each of the magnification converter and the integrator is constituted by a digital circuit.

According to the second characteristic configuration, useful P-wave, Q-wave, R-wave, S-wave, which are useful electrocardiogram information included in the electrocardiogram signal.
By focusing on a specific signal component among characteristic signal components such as a wave, a T-wave, and a QRS complex, it is possible to accurately confirm the state of occurrence. In addition, in particular, according to the third characteristic configuration, since the generation phase of the characteristic signal component can be accurately extracted from the cross-correlation output, the subject related to the generation phase of the extracted characteristic signal component can be extracted. Physiological information can be obtained.

Further, since only the waveform information of a specific signal component is extracted by this processing, the fluctuation of the base line and the fluctuation of the electrocardiogram signal included in the electrocardiogram signal to be processed are removed as a result. Since the characteristic signal component of the generation time phase which is substantially accurate is obtained for the electrocardiogram waveform obtained by the method, the standard 1 relating to the generation time phase of the characteristic signal component is obtained.
Clinical data of electrocardiogram by 2-lead method can be effectively used.

Incidentally, the cross-correlation processing of this characteristic configuration is performed by applying a technique known as a pattern matched filter. That is, cross-correlation processing is performed between a signal waveform (template waveform) having the same shape as the characteristic signal component to be extracted and the electrocardiogram signal to be processed, and a local peak at which the output value of the cross-correlation output is equal to or greater than a predetermined threshold value is detected. By detecting, the appearance time of the characteristic signal component to be extracted can be known based on the detection time. However, in general, the computational load of the cross-correlation processing is very heavy and lacks real-time characteristics. Therefore, in this characteristic configuration, three rectangular waves having a unit amplitude of negative, positive, negative or positive, negative, positive In order of sex, the time ratio of each polarity is arranged to be 1: 2: 1,
By using a template waveform that is set so that the time width of the three rectangular waves is equal to the time width of the characteristic signal component to be extracted, the cross-correlation calculation can be executed by simple addition and subtraction processing. The processing speed is greatly increased. Further, even if the arithmetic processing is simplified, signal components having a time width different from the time width of the three rectangular waves are canceled out in the cross-correlation arithmetic processing because the total time distribution of each of the positive polarity and the negative polarity is equal. Thus, necessary characteristic signal components can be extracted.

According to the fourth characteristic configuration, even if the processed electrocardiogram signal has a large baseline fluctuation, a fluctuation in the waveform shape near the R-wave peak, or other fluctuations, the appearance time of the QRS complex can be accurately detected. can do. As a result, physiological information of the subject such as heart rate variability based on the RR interval can be easily determined.

According to the fifth aspect, when the signal processing means continuously detects the characteristic signal component a predetermined number of times or more, it is possible to determine that the subject is in a stable state. By using the result of the signal processing means after the transmission for the determination of the physiological condition such as the physical condition of the subject, the determination can be made more accurately. In particular, in the case of a bathtub electrocardiograph, since a temporary heartbeat fluctuation and an increase in blood pressure occur immediately after bathing, the above determination can be executed by using a trigger signal without being affected by such initial fluctuation.

According to the sixth or seventh characteristic configuration, an electrocardiogram in which the positions of the electrodes are approximated to the electrocardiogram obtained by the standard 12-lead method can be obtained. By amplifying, it is possible to artificially generate an electrocardiogram signal by three types of bipolar derivation of standard limb lead which is a part of the standard 12 lead method. Further, from these three types of bipolar derived signals, an electrocardiogram based on three types of unipolar derived called a Goldberger derivation method can also be obtained by calculation. Therefore, various diagnoses based on past clinical data can be performed by comparing the electrocardiogram obtained by this feature configuration with the electrocardiogram obtained by the conventional standard 12-lead method.

According to the eighth characteristic configuration, the position of the body surface potential detected by the fourth electrode on the subject's body surface is different from that of the other three.
Since the body surface potential detected by one electrode is electrically substantially equidistant from the position on the subject body surface, the potential of the fourth electrode is set as the reference potential of the three differential amplifiers. A common power supply voltage can be supplied to each differential amplifier,
Each input voltage can also be suppressed within a certain range with respect to the reference potential, and reliable operation of each differential amplifier can be ensured, and the differential amplifier can be simply configured.

According to the ninth characteristic configuration, since the physical condition of the subject is estimated by the physical condition estimating means, it is not necessary for the expert to analyze the processing result of the signal processing means, and the subject himself or a third party can The subject's physical condition can be easily and accurately determined. It is also possible to automatically or semi-automatically adjust the environment of the subject based on the estimation result.

[0030]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of an electrocardiogram measuring apparatus according to the present invention (hereinafter, simply referred to as "the present invention apparatus") will be described with reference to the drawings for a bathtub electrocardiograph.

<First Embodiment> As shown in FIG. 1, the apparatus of the present invention comprises an electrocardiogram signal generating means for generating an electrocardiogram signal 41 from a plurality of electrocardiogram electrodes 10 and electric signals 40 appearing on the plurality of electrocardiogram electrodes 10. 11 and a signal processing unit 12 for performing predetermined signal processing on the processed electrocardiogram signal 42 generated by the electrocardiogram signal generation unit 11. In the first embodiment, the signal processing unit 12 directly processes the electrocardiogram signal 41 generated by the electrocardiogram signal generation unit 11 as the processed electrocardiogram signal 42.

As shown in FIG. 2, a plurality of ECG electrodes 10
Consists of four electrodes 10a to 10d, and each electrode 10a to 10d
10d is separately installed at a predetermined position on the inner wall surface of the bathtub 2 where the subject 1 takes a bath in the hot water. Specifically, the first
The electrode 10a is located at a position near the outside of the right base of the subject 1, the second electrode 10b is located at a position near the outside of the left base of the subject 1, and the third electrode 10c is located near the outside of the left base of the subject 1. The fourth electrode 10d is installed at a position near the outside of the right base of the subject's 1 foot.

As shown in FIG. 3, the electrocardiogram signal generating means 11 is basically composed of three differential amplifiers 13 to 15, and the electric signal 40a guided to the first electrode 10a is supplied to the first amplifier 10a. 13 and the negative input of the second amplifier 14
The electric signal 40b induced at the third electrode 10c is applied to the positive input of the first amplifier 13 and the negative input of the third amplifier 15, and the electric signal 40c induced to the third electrode 10c is applied to the second amplifier 14b. And the positive side input of the third amplifier 15. Further, each of the differential amplifiers 13 to 15 has a ground terminal electrically connected to the fourth electrode 10d, and a potential of the electrode 10d is provided as a reference potential of each of the differential amplifiers 13 to 15,
Each power supply is also supplied from a common power supply.

Depending on the configuration of the electrocardiogram signal generating means 11 and the positions of the electrocardiogram electrodes 10a to 10d, the first amplifier 13 outputs the pseudo I-lead ECG signal 41a and the second
From the pseudo-lead II electrocardiogram signal 41b
However, the pseudo third lead ECG signal 41c is output from the third amplifier 15, respectively. Here, each ECG signal 4
1a to 41c are called pseudo inductions because
Although it corresponds to three types of bipolar derived electrocardiograms of the standard limb lead in the standard 12-lead method, it takes into account that a baseline sway peculiar to the bathtub electrocardiograph described later is included.

Each of the electrocardiogram signals 41a to 41c can be expressed by the following equations 1 to 3 in the form of mathematical expressions. Here, V _I , V _II , and V _III are the signal voltages of the respective pseudo-lead electrocardiogram signals 41a to 41c, and V _R , V _L , and V _F are the first electrode 10a and the second electrode 10b, respectively. The third electrode 10
The potential of c.

[0036]

[Number 1] V _I = V _L - V _R

## EQU2 ## V _II = V _F −V _R

## EQU3 ## V _III = V _F −V _L

The electrocardiogram signals 41a to 41c are further amplified or filtered for unnecessary power supply noise and high frequency noise as required.

Further, as shown in FIG. 3, the electrocardiogram signal generating means 11 converts the electrocardiogram signals corresponding to the three types of unipolar derivation of the standard limb lead in the standard 12-lead method into the following equations 4 to 6:
The calculation means 16 is derived by calculation based on _{Here, aV R, aV L, aV} F is the signal voltage of the electrocardiogram signal corresponding to each of three monopolar derivation.

[0039]

Equation 4] _{aV R = V R -0.5 (V} L + V F) = - 0.5 (V I + V II)

AV _L = V _L -0.5 (V _R + V _F ) = 0.5 (V _I -V _III )

AV _F = V _F −0.5 (V _R + V _L ) = 0.5 (V _I + V _II )

The electrocardiogram signal generating means 11 having the above-described configuration allows the electrocardiogram signals 41 (V _I , V _II , V _III , aV _R , aV _L , aV _F ) of six standard limb leads in the standard 12 lead method.
Is generated.

The signal processing means 12 generates an electrocardiogram signal 41 generated by the electrocardiogram signal generation means 11 or an electrocardiogram signal obtained by subjecting the electrocardiogram signal 41 to predetermined preprocessing, for example, power noise removal processing. Are collectively referred to as the processed electrocardiogram signal 42).

As shown in FIG. 4, the baseline fluctuation removing means 17 includes a subtractor 18, an integrator 19, and a magnification converter 20. The processing circuit shown in FIG. 4 is for one signal. The positive input of the subtractor 18 is one of the processed electrocardiogram signals 42, and the output is the electrocardiogram signal 43 after the baseline oscillation is removed. By subjecting the electrocardiogram signal to magnification conversion processing and integration processing, a low-frequency baseline generated due to movement of hot water in the bathtub 2 or movement of the body of the subject 1 superimposed on the electrocardiogram signal 42 to be processed. An estimated value of the sway component is obtained, and the estimated value is fed back to the negative input of the subtractor 18, whereby an electrocardiogram signal 43 after the baseline sway is removed is obtained. Here, by setting the overall time constant of the integrator 19 and the magnification converter 20 to about 1 to 10 seconds, the characteristic signal components (P wave, Q wave, R wave, S wave, T
The waveform of the wave, the QRS complex, etc.) is smoothed along the baseline fluctuation by the above-described magnification conversion processing and integration processing, and an estimated value of the latest baseline fluctuation component is obtained. Since the high frequency characteristic signal component has been removed from the estimated value, the characteristic signal component is not lost from the processed electrocardiogram signal by the subtraction processing,
Only the base line fluctuation of the low frequency component is subtracted. Therefore, if the time constant is extremely short, a characteristic signal component is included in the estimated value, and the original signal component may be lost in addition to the baseline fluctuation. Here, the same effect can be obtained by replacing the arrangement of the integrator 19 and the magnification converter 20 with the configuration shown in FIG. 4 and reversing the order of the magnification conversion processing and the integration processing.

Further, the baseline fluctuation removing means 17 can be realized by digital signal processing in addition to analog signal processing. In this case, the integration process can be easily realized by the piecewise quadrature method. That is, the input digital values may be cumulatively added at each sampling interval starting from the initial value 0, and the values may be stored and output. More specifically, if the input digital value is 16 bits wide, cumulative addition is performed with 32 bits, and the value obtained by multiplying the cumulative added value by a coefficient indicating a magnification is the above-mentioned estimated value. Real-time processing can be easily realized by subtracting from the digitally converted electrocardiogram signal 42 to be processed. The processed electrocardiogram signal 4
The A / D conversion process 2 may be executed by either the signal processing unit 12 or the electrocardiogram signal generation unit 11.

In this embodiment, the baseline sway elimination process is performed on all of the six standard limb lead electrocardiogram signals 41 in the standard 12 lead method, thereby obtaining the results in clinical medicine and physiological research. Electrocardiogram signals 43 approximate to the six standard limb leads in the standard 12 lead method are obtained. As a result, it is possible to compare these data with clinical data accumulated in clinical medicine and physiology research sites, and to evaluate the effectiveness of such clinical data even for simple ECG measurement results using a bathtub electrocardiograph. Can be used.

[0045] FIG. 5 shows an example of a result of performing the treatment with baseline upset removing means 17 relative to the second inductive signal V _II of the processed electrocardiographic signal 42. From this, it can be seen that the removal of the baseline guidance is effectively performed.

Incidentally, the above-mentioned baseline fluctuation removal processing is not necessarily required to be executed for all of the six kinds of standard limb lead electrocardiogram signals 41, and attention is paid to specific test items such as the heart rate of the subject 1. May be performed by inputting only a part of the electrocardiogram signals to the baseline fluctuation removing means 17 and performing a baseline fluctuation removing process. In addition, since three types of electrocardiogram signals derived from unipolar derivation of the six standard limb leads are generated by arithmetic processing on three types of bipolar derived electrocardiogram signals, such arithmetic processing is performed by three types of electrocardiogram signals after the baseline sway is removed. It may be performed on a bipolar derived electrocardiogram signal. According to this processing procedure, the circuit configuration of the baseline fluctuation removing unit 17 can be simplified.

<Second Embodiment> As shown in FIG. 6, a plurality of electrocardiogram electrodes 10, electrocardiogram signal generating means 11 and signal processing means 12 are provided. In the second embodiment, the signal processing means 12 is provided with a characteristic signal component extracting means 21 in addition to the baseline fluctuation removing means 17, although there is no change in the typical configuration.
The plurality of electrocardiogram electrodes 10, the electrocardiogram signal generating means 11, and the baseline fluctuation removing means 17 are the same as in the first embodiment, and a description thereof will be omitted.

The characteristic signal component extracting means 21 converts the P-wave, the Q-wave, and the R-wave from the electrocardiogram signal 43 after the baseline oscillation removal generated by the baseline oscillation removing section 17 which becomes the electrocardiogram signal 44 to be processed.
It is configured to extract a specific signal wave from characteristic signal components such as a wave, an S wave, a T wave, and a QRS complex, and further extract and output electrocardiographic information included in the specific signal wave. I have.
Specifically, in the present embodiment, the QRS complex is extracted, its occurrence time phase is detected, and the heart rate and its time variation are detected in real time from the occurrence time interval. As shown in FIG. 7, the specific configuration is as follows: a cross-correlation calculation unit 22, an optimum window width setting unit 23, a peak time phase detection unit 24, an RR interval calculation unit 25, a heart rate variability spectrum analysis unit 26, a trigger The signal generator 27 is provided.

In general, an electrocardiogram signal includes a base line sway caused by respiration, body movement, fluctuations in a waveform shape near the R-wave peak, a change in the position of the heart, and a change in the amplitude value of the signal itself. Various fluctuations such as turbulence are mixed. The characteristic signal component extracting means 21 is basically a method known as a pattern matched filter as an operation principle, so that even when fluctuations in the processed electrocardiogram signal 44 are included, the occurrence phase of the QRS complex can be accurately detected. To the cross-correlation calculation unit 22
For cross-correlation processing. Therefore, the cross-correlation calculating unit 22 outputs the electrocardiogram signal 4 after the baseline oscillation removal generated by the baseline oscillation removing unit 17 to be the processed electrocardiogram signal 44.
3 and a template waveform simulating the waveform of the QRS complex to be extracted.

However, the normal cross-correlation processing using a pattern matched filter has a very heavy computational load and lacks real-time performance. Therefore, in this embodiment, three rectangular waves having a unit amplitude as shown in FIG. Are used such that the time ratio of each polarity is 1: 2: 1 in the order of negative polarity / positive polarity / negative polarity. The time width of the three rectangular waves is set to be the time width of the QRS complex included in the electrocardiogram signal 43. By using a square wave of unit amplitude as a template waveform, multiplication processing during digital arithmetic processing can be greatly reduced, and the cross-correlation value for the subsequent sampling value at a fixed sampling interval can be expressed by a recurrence formula, and the recalculation load Can be reduced. In addition, even if such arithmetic processing is simplified, since the total time distribution of each of the positive polarity and the negative polarity is equal,
Since signal components having time widths different from the time widths of the three rectangular waves are canceled in the cross-correlation calculation processing, necessary characteristic signal components can be extracted. Hereinafter, a description will be given based on a specific example.

The input electrocardiogram signal 44 to be processed has already been digitized, and the time-series data d
[T]. For simplicity of explanation, if the time width of the template waveform and the QRS complex is set to 12 samples, the template waveform w [t] is expressed as in Expression 8. In this case, the cross-correlation value φ [t] (for example, the first four samples φ ₀
To φ ₃ ) are obtained as shown in Equations 9 to 12.

[0052]

D [t] = [d0, d1, d2, d3,..., D _n-1 ]

## EQU8 ## w [t] = [-1, -1, -1,1,1,1,1]
1,1,1, -1, -1, -1, -1]

## EQU9 ## φ0 = −d0−d1−d2 + d3 + d4 + d5
+ D6 + d7 + d8-d9-d10-d11

## EQU10 ## φ1 = -d1-d2-d3 + d4 + d5 + d
6 + d7 + d8 + d9-d10-d11-d12

## EQU11 ## φ2 = -d2-d3-d4 + d5 + d6 + d
7 + d8 + d9 + d10-d11-d12-d13

## EQU12 ## φ3 = -d3-d4-d5 + d6 + d7 + d
8 + d9 + d10 + d11-d12-d13-d14

Here, Equations (10) to (12) are obtained by rewriting each cross-correlation value φ _i using φ _i−1 one sample before.
As shown in Expression 15, it is understood that the calculation load can be reduced because it is represented by a simple recurrence formula of only addition and subtraction of the above-mentioned cross-correlation value and four sample values of the processed electrocardiogram signal. Further, when the time width of the template waveform and the QRS complex is generalized to 4 m samples, Expressions 13 to 15 can be expressed as Expression 16.

[0054]

## EQU13 ## φ1 = φ0 + d0−2 × d3 + 2 × d9−d12

[Formula 14] φ2 = φ1 + d1-2 × d4 + 2 × d10−d13

[Formula 15] φ3 = φ2 + d2-2 × d5 + 2 × d11−d14

[Formula 16] φ _{i + 1} = φ _i + d _i −2 × d _{i + m} + 2 × d _{i + 3m} −d _{i + 4m}

The cross-correlation calculator 22 based on the calculation formula shown in Expression 16 can be realized with a simple configuration using a FIFO memory 28, a register 29, and an adder 30, as shown in FIG. The input and output bit widths of the FIFO memory 28, the register 29, and the adder 30 are determined according to the number of quantization bits of the signal value of the processed electrocardiogram signal 44. Optimal window width setting unit 23
Inputs the cross-correlation value 45 output from the cross-correlation calculator 22 and sets the time width of the template waveform at which the local peak value of the cross-correlation value 45 becomes maximum.

FIG. 10 shows an example of input / output signals of the cross-correlation calculator 22. The upper part is a processed electrocardiogram signal 44 which is an input signal, and the lower part is a cross-correlation value 45 which is an output signal. As described above, the cross-correlation calculator 22 can extract the waveform of the QRS complex from the processed electrocardiogram signal 44.

As shown in FIG. 7, the peak time phase detector 24
Detects the presence of a local peak from the time series data of the cross-correlation value 45, and outputs a pulse signal 46 at each detection time.
The RR interval calculator 25 sequentially receives the pulse signal 46 and calculates an RR interval (time interval from the QRS group to the next QRS group) of the electrocardiogram waveform based on the reception time interval. Is converted into real-time waveform information, and a heartbeat variability signal 47 is output. Heart rate variability spectrum analyzer 2
6 samples the heart rate variability signal 47 at a sampling rate of, for example, about 2 Hz and performs spectrum analysis.
The power spectrum of heart rate variability is in the low frequency range (0.04 ~
0.15 Hz) and a high frequency range (0.15 to 0.5 Hz), respectively.
Outputs the peak value in the low frequency range as the LF component 48 and the peak value in the high frequency range as the HF component 49.

The trigger signal generator 27 outputs a pulse signal 46
Is input, the number of inputs is counted, and a trigger signal 50 is output when the count value becomes equal to or greater than a predetermined value within a predetermined time. The count of the pulse signal 46 is
It is executed by the built-in up counter. When the power is turned on, the hot water supply to the bathtub 2 is started,
The count value is reset to “0” when the measurement mode is set and at regular time intervals. Since the trigger signal 50 indicates a state in which the QRS complex is continuously generated in the electrocardiogram waveform of the subject 1, the electrocardiogram signal 43 after the trigger signal 50 is output and the characteristic signal component extracting means 21 The output heart rate variability signal 47 and its LF component 48 and HF component 49 can be determined to be in a stable state that can be used for estimating the physical condition of the subject 1 described later, for example, and the output of the trigger signal 50 is confirmed. Thus, the reliability of the result of the physical condition estimation and the like using the various signals can be improved.

<Third Embodiment> As shown in FIG. 11, the third embodiment includes a physical condition estimating means 31 for estimating the physical condition of the subject 1 in addition to the configuration of the second embodiment shown in FIG. It is configured. The physical condition estimating means 31 includes an electrocardiogram signal 43 output from the baseline fluctuation removing means 17, a heart rate variability signal 47 output from the characteristic signal component extracting means 21, and its L
The F component 48 and the HF component 49 are received as electrocardiogram information of the subject 1. When receiving the trigger signal 50 output from the characteristic signal component extracting means 21, the physical condition of the subject 1 is estimated based on the electrocardiogram information. The physical condition estimation method in the physical condition estimating means 31 is performed by searching a database in which the relationship between the electrocardiogram information and the physical condition and symptoms is stored in advance based on clinical data and physiological research data accumulated by the standard 12-lead method. As a relationship between the electrocardiogram information and a physical condition or symptom, for example, a relationship as shown in Table 1 is used.

[0060]

[Table 1]

In bathing in a normal state of health, immediately after the input, the heart rate temporarily increases and the blood pressure also increases, but gradually returns to the original state. However, for example, when the above ECG information appears and the corresponding physical condition abnormality is estimated,
There is a possibility that the exercise load due to bathing is too strong for the current physical condition, and it is possible to perform a treatment for reducing the exercise load on the subject 1 at an early stage based on the physical condition estimation result.

Another embodiment will be described below.

<1> In the first embodiment, the configuration of the electrocardiogram signal generation means 11 is not necessarily limited to the configuration of the three differential amplifiers 13 to 15 described above. For example, for one differential amplifier, each electric signal 40 of the first, second and third electrodes 10a, 10b, 10c
Time division processing may be performed by switching and inputting a, 40b, and 40c at high speed.

<2> In the first embodiment, the arrangement of the plurality of electrocardiogram electrodes 10a to 10d is based on the arrangement method shown in FIG. The fourth electrode 10b is located at a position close to the outside of the left base of the left arm of the subject 1 and the third electrode 10c is located near the upper left knee of the subject 1 and at the same depth as the second electrode. 10d may be installed at a position close to the upper outside of the right knee and at the same depth as the second electrode. In any arrangement method, another electrode potential may be used as the reference potential of the differential amplifiers 13 to 15 without providing the fourth electrode 10d.

<3> In the second embodiment, since the baseline fluctuation of the electrocardiogram signal 44 to be processed can be removed at the same time by the cross-correlation processing of the cross-correlation calculator 22, the baseline fluctuation removing means 17 generated by the baseline fluctuation removing means 17 can be used. Later ECG signal 4
If you do not need to use 3 in the post process such as physical condition estimation,
Without providing the baseline fluctuation removing means 17 in the signal processing means 12,
The processed electrocardiogram signal 4 to be input to the baseline fluctuation removing means 17
2 may be input to the characteristic signal component extraction means 21.

<4> In the second embodiment, the characteristic signal component extracting means 21 is the heart rate variability spectrum analyzing section 2
6 and the trigger signal generator 27 may not be provided. If the heart rate variability spectrum analysis unit 26 is not provided, the RR interval calculation unit 25
The heart rate may be output at predetermined time intervals instead of the heart rate variability signal 47.

<5> In the second embodiment, the configuration of the cross-correlation calculator 22 is not limited to the configuration of the above-described embodiment. Further, even with the same circuit configuration, for example, a ring-shaped memory may be used instead of the FIFO memory 28. The signal wave to be extracted by the cross-correlation calculator 22 is not limited to the QRS complex. When extracting another signal wave, the time width of the template waveform may be set to the time width of the signal wave to be extracted.

<6> In each of the above embodiments, the case of a bathtub electrocardiograph has been described. However, even in a normal electrocardiograph, when a stainless steel electrode or an Ag electrode is used, the baseline of the electrocardiogram due to the influence of the subject's sweat is used. There is a problem that the oscillation easily occurs, and the baseline oscillation cannot be completely removed by a noise removing means such as a simple low-pass removing filter provided in a normal electrocardiograph. Incidentally, an expensive Ag-AgCl which is hardly affected by sweat is used for an electrode of an electrocardiograph used for general clinical medicine and physiological research. Accordingly, since the baseline fluctuation removing means 17 and the characteristic signal component extracting means 21 of the first embodiment have a function of effectively removing the baseline fluctuation,
The present invention may be applied to a normal electrocardiograph baseline removal process or a characteristic signal component extraction process using a stainless steel electrode or an Ag electrode.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a first embodiment of an electrocardiogram measuring apparatus according to the present invention.

FIG. 2 is an explanatory diagram showing an example of arrangement of a plurality of electrocardiogram electrodes.

FIG. 3 is a block diagram showing an electrocardiogram signal generating means.

FIG. 4 is a block diagram showing a baseline fluctuation removing unit.

FIG. 5 is a waveform chart showing an example of a processing result by a baseline fluctuation removing unit.

FIG. 6 is a block diagram showing a second embodiment of the electrocardiogram measuring apparatus according to the present invention.

FIG. 7 is a block diagram showing a characteristic signal component extracting unit.

FIG. 8 is a waveform chart showing an example of a template waveform.

FIG. 9 is a block diagram showing a cross-correlation calculating unit.

FIG. 10 is a waveform chart showing an example of input / output signals of a cross-correlation calculating unit.

FIG. 11 is a block diagram showing a third embodiment of the electrocardiogram measuring apparatus according to the present invention.

[Explanation of symbols]

 1: Subject 2: Bathtub 10: Electrocardiogram electrode 11: Electrocardiogram signal generating means 12: Signal processing means 13, 14, 15: Differential amplifier 16: Operation means 17: Baseline fluctuation removing means 18: Subtractor 19: Integrator 20: Magnification converter 21: Characteristic signal component extraction means 22: Cross-correlation calculation unit 23: Optimal window width setting unit 24: Peak time phase detection unit 25: RR interval calculation unit 26: Heart rate fluctuation spectrum analysis unit 27: Trigger signal Generation unit 28: FIFO memory 29: Register 30: Adder 31: Physical condition estimation means 40: Electric signal 41, 43: Electrocardiogram signal 42, 44: Electrocardiogram signal to be processed 45: Cross-correlation value 46: Pulse signal 47: Heart rate fluctuation signal 48: LF component 49: HF component 50: Trigger signal

 ──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Satoshi Fujita 4-1-2, Hirano-cho, Chuo-ku, Osaka-shi, Osaka Osaka Gas Co., Ltd. (72) Inventor Tomoaki Ueda 17 2D032 AA00 4C027 AA02 CC00 EE01 EE08 FF03 FF05 GG01 GG05 GG10 KK01

Claims (9)

[Claims] 1. A plurality of electrocardiogram electrodes, an electrocardiogram signal generation means for generating and processing an electrocardiogram signal from electric signals appearing on the plurality of electrocardiogram electrodes, and a processed electrocardiogram signal generated by the electrocardiogram signal generation means. Signal processing means for performing predetermined signal processing, the signal processing means has a subtractor, a magnification converter and an integrator, and inputs the processed electrocardiogram signal to a positive side input of the subtractor, With respect to the output signal from the subtractor, one of a magnification conversion process by the magnification converter and an integration process by the integrator is performed first, and a signal obtained by subsequently performing the other is a negative side of the subtractor. An electrocardiogram measuring apparatus, characterized in that an electrocardiogram signal obtained by removing or reducing baseline fluctuation from the processed electrocardiogram signal is output from the subtracter by inputting the input to an input. 2. A plurality of electrocardiogram electrodes, an electrocardiogram signal generating means for generating an electrocardiogram signal from electric signals appearing on the plurality of electrocardiogram electrodes, and a processed electrocardiogram signal generated by the electrocardiogram signal generating means. Signal processing means for performing predetermined signal processing, wherein the signal processing means outputs three rectangular waves of unit amplitude in the order of negative polarity, positive polarity, negative polarity or positive polarity, negative polarity, and positive polarity. Are arranged so that the time ratios of the signals are 1: 2: 1, and the time width of the three rectangular waves is set to be the time width of one or more characteristic signal components included in the processed electrocardiogram signal. An electrocardiogram measuring apparatus, comprising: a cross-correlation calculating unit that performs a cross-correlation process between a template waveform and a processed electrocardiogram signal. 3. An occurrence time phase detecting means for detecting an occurrence time phase of the characteristic signal component based on a local peak detection time at which a cross-correlation output of the cross-correlation calculation section is equal to or greater than a predetermined threshold value. The electrocardiogram measuring device according to claim 2, wherein: 4. The electrocardiogram measuring apparatus according to claim 2, wherein the characteristic signal component is a QRS complex. 5. The apparatus according to claim 2, wherein said signal processing means outputs a predetermined trigger signal when said characteristic signal component is continuously detected a predetermined number of times or more.
The electrocardiogram measuring device according to the above. 6. An electrocardiogram signal which is a pseudo standard limb lead from at least three electrocardiogram electrodes respectively installed at predetermined positions on an inner wall surface of a bathtub in which a subject takes a bath, and electric signals appearing at the at least three electrocardiogram electrodes. Wherein each of the ECG electrodes has a first electrode located at a position near the outside of the right base of the subject's right arm, and a second electrode at a position near the outside of the base of the left arm of the subject. , A third electrode is disposed at a position close to the left side of the subject's left foot, and the electrocardiogram signal generating means includes a first electrode and a second electrode.
, Between the first electrode and the third electrode, and
An electrocardiogram measuring apparatus, comprising: differential amplifying means for separately amplifying a difference signal component of the electric signal between the second electrode and the third electrode. 7. An electrocardiogram signal which is a pseudo standard limb lead from at least three electrocardiogram electrodes which are separately provided at predetermined positions on an inner wall surface of a bathtub in which a subject takes a bath, and electric signals appearing at the at least three electrocardiogram electrodes. Wherein each of the ECG electrodes has a first electrode located at a position near the outside of the right base of the subject's right arm, and a second electrode at a position near the outside of the base of the left arm of the subject. A third electrode is disposed at a position close to and above the left knee of the subject and at the same depth as the second electrode, and the electrocardiogram signal generating means includes a first electrode and a second electrode.
, Between the first electrode and the third electrode, and
An electrocardiogram measuring apparatus, comprising: differential amplifying means for separately amplifying a difference signal component of the electric signal between the second electrode and the third electrode. 8. The at least three ECG electrodes, in addition to the first, second, and third electrodes, with respect to the third electrode on the inner wall surface of the bathtub with respect to a body of a subject in the bathtub. And a fourth electrode disposed at a symmetrical position, wherein the differential amplifying means comprises three independent differential amplifiers, and all of the reference potentials of the three differential amplifiers are The electrocardiogram measuring apparatus according to claim 6, wherein the potential of the fourth electrode is set to the potential of the fourth electrode. 9. The electrocardiogram measuring apparatus according to claim 1, further comprising a physical condition estimating means for estimating a physical condition of a subject based on a processing result of the signal processing means. .