JP2003135434A - Signal processing method and pulse wave signal processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To display and compute an accurate oxygen saturation by eliminating noise such as body motions superimposed on pulse wave signals. SOLUTION: The pulse waves of infrared light and red light are detected with a probe 1 and respective spectrums are obtained by Fourier transform. By using the ratio of the difference of the spectrums and one of the spectrums or the sum, a fundamental frequency of the pulse waves is obtained. Then, a filter using the fundamental frequency of the pulse waves and the frequency of a higher harmonic is formed and the pulse waves of the infrared light and the red light are respectively filtered by using the filter. The filtered pulse waves are displayed on a display part 11 and used for computing the accurate oxygen saturation.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a signal processing method, and more particularly to pulse wave noise removal that can be used in oxygen saturation measurement by pulse photometry.

[0002]

2. Description of the Related Art A pulse oximeter has been conventionally used for non-invasive continuous measurement of oxygen saturation of arterial blood. In this pulse oximeter, a probe is attached to a fingertip or an earlobe of a subject, and light of different wavelengths of red and infrared is radiated from a probe to a living body in a time-sharing manner to obtain transmitted or reflected light of two different wavelengths Ratio of pulsating component of absorbance Φ
Is to measure the oxygen saturation S. A reference wavelength of, for example, 660 nm is used for red light, and a wavelength of, for example, 940 nm is used for infrared light. Two light emitting diodes emitting these wavelengths and one photodiode for receiving light are built in the probe. Has been done. Now, assuming that the pulsation component of the absorbance of the red light wavelength is ΔA1 and the pulsation component of the absorbance of the infrared light wavelength is ΔA2, different 2
The wavelength absorption ratio Φ is given by the following equation. Φ = ΔA1 / ΔA2 The oxygen saturation S can be calculated as a function f of this absorbance ratio Φ. S = f (Φ)

In such a pulse oximeter, when a patient moves during measurement, a noise component is mixed in the pulse wave detected by the probe, and the oxygen saturation S is accurately measured.
Cannot be measured. Therefore, various attempts have conventionally been made to remove the influence of such noise.

Japanese Unexamined Patent Publication (Kokai) No. 2-172443 describes a technique of detecting a pulse in an oximeter and performing frequency filtering on the photoelectric pulse wave signal based on the frequency of the pulse. Here, as a specific method of detecting the pulse wave number, the output of the light receiving unit is logarithmically converted, the pulse wave signal obtained from the logarithmically converted output is binarized, and further, the predetermined number of binarized pulses. It is described that the period of the wave signal is obtained and the pulse rate is obtained from the reciprocal of the period. Furthermore, frequency filtering uses a bandpass filter with a low selection frequency when the pulse rate is small, and a bandpass filter with a medium selection frequency when the pulse rate is medium,
It is described that a bandpass filter with a high selection frequency is used when the pulse rate is large.

However, such an invention has the following problems. <Problem of Binarization> When body movement occurs, the pulse wave signal is mixed with noise caused by body movement and is buried.
If logarithmically calculating the pulse wave signal containing this body motion noise,
The ratio of the logarithmically calculated body motion component to the logarithmically calculated pulse wave component is smaller than the ratio of the logarithmically calculated body motion component to the logarithmically calculated pulse wave component. In the above-mentioned conventional technique, the pulse wave signal in which the body motion noise logarithmically calculated is mixed is distributed to 1 or 0 by a certain threshold value by the binarization calculation. However, since the noise component is not necessarily smaller than the pulse wave component, it is extremely likely that the noise component will be assigned to 1 by binarization. With this, the pulse rate cannot be detected accurately. <Problem of bandpass filter> When only a bandpass filter is used to remove noise, noise is mixed in the pulse wave when the noise component is in the pass band. That is, the pulse wave cannot be extracted well.

[0006]

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and its purpose is to:
(1) The fundamental frequency of the signal is accurately calculated, and (2) the fundamental frequency and its high frequency are used as filtering frequencies to perform signal processing, thereby extracting the characteristic frequency component of the signal To form. In particular, in the application to the measurement of oxygen saturation of arteries, even if the pulse wave is superimposed with noise due to body movement,
Alternatively, even if the pulse wave fluctuates, the pulse wave signal is processed by accurately measuring the fundamental frequency of the pulse and using the fundamental frequency and its harmonic frequency as the filtering frequency. Form a pulse wave signal by extracting characteristic frequency components from the pulse wave on which is superimposed,
Noise can be removed and the original pulse wave can be reproduced.
Another object of the present invention is to provide a device for accurately measuring oxygen saturation by pulse photometry from a pulse wave from which noise has been removed.

[0007]

A signal processing method according to claim 1 of the present invention is a signal processing method for processing a continuous first signal IR and second signal RD having the same fundamental frequency, For each of the first signal IR and the second signal R 1, a frequency spectrum or a frequency power spectrum is calculated in a predetermined period, and the first signal IR is calculated.
The first spectrum Spc.IR of IR and the second spectrum of the second signal RD
A spectrum calculation step for calculating the spectrum Spc.RD, the first spectrum Spc.IR and the second spectrum Spc.RD calculated in the spectrum calculation step
On the frequency axis, the first signal IR and the second signal IR are calculated based on the result of the spectrum calculation for obtaining the mutual difference or the spectrum calculation for obtaining the mutual difference and further performing the normalizing spectrum calculation. And a fundamental frequency calculation step of calculating the fundamental frequency of the signal RD. This is to accurately calculate the fundamental frequency of the signal even if the signal contains noise.

A signal processing method according to claim 2 of the present invention comprises:
In the signal processing method according to claim 1, in the fundamental frequency calculation step, (Spc.IR- Spc.RD) /Spc.RD, (Spc.IR- Spc.RD) /Spc.IR, (Spc.IR). − (Spc.RD) / (Spc.IR + Spc.RD), {1- (Spc.RD / Spc.IR)} / (Spc.RD / Spc.I
R), {1- (Spc.RD / Spc.IR)} / (Spc.IR / Sp
c.IR), or {1- (Spc.RD / Spc.IR)} / {1+ (Spc.RD /
Spc.IR)}, the fundamental frequency is calculated based on the result of the spectrum calculation. This is to make it easier to extract the fundamental frequency.

The signal processing method according to claim 3 of the present invention is
A continuous first signal IR and a second signal having the same fundamental frequency
In the signal processing method for processing the signal RD, the frequency spectrum or the frequency power spectrum is calculated for each of the first signal IR and the second signal RD in a predetermined period, and the first signal IR 1 spectrum Spc.IR from the second signal RD to the second spectrum Sp
c.RD spectrum calculation step, spectrum addition and average calculation step of adding and averaging the first spectrum Spc.IR and the second spectrum Spc.RD for a predetermined number of times by the spectrum calculation step, and the spectrum addition and average Using the first spectrum Av.Spc.RD and the second spectrum Av.Spc.RD that are averaged and added in the calculation step, the spectrum calculation for calculating the mutual difference or the mutual difference is performed on the frequency axis. The first signal IR and the second signal RD are calculated based on the result of the spectrum calculation for normalization in addition to the spectrum calculation.
And a fundamental frequency calculation step of calculating the fundamental frequency of. This is because the fundamental frequency of the signal is calculated by suppressing the noise appearing in the spectrum by using the arithmetic mean of the spectrum.

A signal processing method according to claim 4 of the present invention comprises:
The signal processing method according to claim 3, wherein in the fundamental frequency calculation step, (Av.Spc.IR-Av.Spc.RD) /Av.Spc.RD, (Av.Spc.IR-Av.Spc.RD). RD) / Av.Spc.IR, (Av.Spc.IR − Av.Spc.RD) / (Av.Spc.IR + Av.Spc.
RD), {1- (Av.Spc.RD / Av.Spc.IR)} / (Av.Spc.RD /Av.S
pc.IR), {1- (Av.Spc.RD /Av.Spc.IR)} / (Av.Spc.IR / A
v.Spc.IR), or {1- (Av.Spc.RD / Av.Spc.IR)} / {1+ (Av.Spc.RD /
Av.Spc.IR)} based on the result of the spectrum calculation. This is to make it easier to extract the fundamental frequency.

A signal processing method according to claim 5 of the present invention is
The signal processing method according to any one of claims 1 to 4, further comprising: a filter formation that forms a filter using the fundamental frequency calculated by the fundamental frequency calculation step and its harmonic frequency. And a filtering step of filtering at least the first signal IR or the second signal RD by the filter formed by the filter forming step. This is to form a signal by extracting the fundamental frequency component, which is a characteristic frequency component of the signal, and the frequency components of its harmonics from the signal from which the noise is removed.

A signal processing method according to claim 6 of the present invention comprises:
The signal processing method according to claim 5, wherein the filter forming step uses each fundamental frequency and its respective harmonic frequency when there are a plurality of fundamental frequencies computed by the fundamental frequency computing step. It is characterized by forming a filter. This is because even if the fundamental frequency changes, each fundamental frequency component and the frequency component of each harmonic thereof are extracted to form a signal.

A signal processing method according to claim 7 of the present invention comprises:
The signal processing method according to claim 5 or 6, wherein in the filter forming step, the characteristic of the filter is a rectangular shape having a width centered on the fundamental frequency and its harmonic frequency, respectively. Characterize. This is because a signal is formed by taking in the frequency components around the fundamental frequency and the frequencies of its harmonics.

A signal processing method according to claim 8 of the present invention comprises:
The signal processing method according to claim 5 or 6, wherein in the filter forming step, the filter is a Gaussian characteristic filter centered on the frequencies of the fundamental frequency and its harmonics. This is because the frequency components around the fundamental frequency and its higher harmonic frequencies are captured in a Gaussian distribution to form a signal.

A signal processing method according to claim 9 of the present invention comprises:
The signal processing method according to any one of claims 1 to 8, wherein the first signal IR and the second signal RD are both pulse wave signals. This is because the fundamental frequency is obtained from the pulse wave signal of the living body and the characteristic frequency component is extracted to form the pulse wave.

A signal processing method according to a tenth aspect of the present invention is the signal processing method according to any one of the fifth to eighth aspects, wherein the first signal IR and the second signal RD are included. Are pulse wave signals, and the step of displaying the filtered signal of the first signal IR or the second signal RD in the filtering step is further included. This is because the characteristic frequency component of the pulse wave is extracted from the pulse wave signal of the living body to form the pulse wave for display.

The signal processing method according to claim 11 of the present invention is the signal processing method according to any one of claims 5 to 8, wherein the first signal IR is transmitted through an artery of a living body or The second signal RD is a pulse wave signal obtained by reflected infrared light, the second signal RD is a pulse wave signal obtained by red light transmitted through or reflected by an artery of a living body, and the first signal in the filtering step. Filtering both the signal IR and the second signal R 2, and
An oxygen saturation calculation step of calculating an oxygen saturation using the first signal IR or the second signal RD filtered in the filtering step. This is to accurately calculate the oxygen saturation level using the filtered pulse wave.

A signal processing method according to a twelfth aspect of the present invention is a signal detection step of detecting a continuous signal having periodicity, and a fundamental frequency calculation for computing a fundamental frequency of the signal detected by the signal detection step. A step, a filter forming step of forming a filtering frequency by using the fundamental frequency of the signal calculated by the fundamental frequency calculating step and the frequency of its harmonic, and the filter using the filter formed by the filter filter, And a filtering step of filtering. This is for forming a signal by extracting the frequency components of the fundamental frequency and the harmonics thereof with respect to a continuous signal having periodicity.

A signal processing method according to a thirteenth aspect of the present invention is a signal processing method for processing a first signal and a second signal which have the same fundamental frequency and are continuous, wherein the first signal and the second signal are processed. Of the signal, the basic frequency calculation step of calculating the basic frequency of the signal using at least one of the first signal and the second signal, and the basic frequency calculation step. A filter forming step of forming a filter using the fundamental frequency and the frequencies of its harmonics. This is to form a filter for forming a signal by extracting the frequency components of the fundamental frequency and the harmonic frequency of two consecutive signals having the same fundamental frequency.

A signal processing method according to a fourteenth aspect of the present invention is the signal processing method according to the thirteenth aspect, wherein in the filter forming step, there are a plurality of the basic frequencies calculated by the basic frequency calculating step. Is
It is characterized in that a filter is formed by using each fundamental frequency and the frequency of each harmonic. This is because even if the fundamental frequencies of the two signals fluctuate, a filter for forming the signals by extracting the frequency components of the respective fundamental frequencies and their respective harmonic frequencies is formed.

A signal processing method according to a fifteenth aspect of the present invention is the signal processing method according to the thirteenth or fourteenth aspect, wherein in the filter forming step, the characteristic of the filter is the fundamental frequency and its harmonics. It is characterized in that it has a rectangular shape with a width centered on each of the frequencies. This is because a signal is formed by taking in the frequency components around the fundamental frequency and the frequencies of its harmonics.

A signal processing method according to a sixteenth aspect of the present invention is the signal processing method according to the thirteenth or fourteenth aspect, wherein in the filter forming step, the filter has the fundamental frequency and its harmonic frequencies. Is a Gaussian characteristic filter centered at This is because the frequency components around the fundamental frequency and its higher harmonic frequencies are captured in a Gaussian distribution to form a signal.

A signal processing method according to a seventeenth aspect of the present invention is the signal processing method according to any one of the twelfth to sixteenth aspects, in which the first signal and the second signal are both Is also a pulse wave signal. With this, the fundamental frequency is obtained from the pulse wave signal of the living body,
This is because a characteristic frequency component is extracted to form a pulse wave.

A signal processing method according to an eighteenth aspect of the present invention is the signal processing method according to any one of the twelfth to sixteenth aspects, in which the first signal and the second signal are both Is also a pulse wave signal, and the step of displaying the filtered signal of the first signal or the second signal in the filtering step is further included. This is for displaying the filtered pulse wave signal.

A signal processing method according to a nineteenth aspect of the present invention is the signal processing method according to any one of the twelfth to sixteenth aspects, wherein the first signal is transmitted or reflected through an artery of a living body. Pulse signal obtained by infrared light, the second signal is a pulse wave signal obtained by red light transmitted through or reflected by an artery of a living body, and the first signal and An oxygen saturation calculation step of calculating any of the second signals, and further calculating an oxygen saturation using the first signal or the second signal filtered in the filtering step. It is characterized by This is for accurately calculating the oxygen saturation using the filtered pulse wave.

A pulse wave signal processing method according to a twentieth aspect of the present invention is a pulse wave signal processing method for processing a first pulse wave signal and a second pulse wave signal of a living body having the same fundamental frequency,
A first pulsating component detecting step for detecting a pulsating component ΔA1 of the first pulsating wave signal; a second pulsating component detecting step for detecting a pulsating component ΔA2 of the second pulsating wave signal; A pulsating component calculating step for calculating a ratio Φ of the pulsating component of the pulsating signal and the pulsating component of the second pulsating signal, and a frequency spectrum or a frequency power spectrum for a predetermined period to calculate a fundamental frequency of the pulsating signal. And a fundamental frequency calculation step for calculating. With this, by calculating the difference of the pulsating component,
This is for obtaining the fundamental frequency of the pulse wave.

A pulse wave signal processing method according to a twenty-first aspect of the present invention is a pulse wave signal processing method for processing a first pulse wave signal and a second pulse wave signal of a living body having the same fundamental frequency,
A first pulsating component detecting step for detecting a pulsating component ΔA1 of the first pulsating wave signal; a second pulsating component detecting step for detecting a pulsating component ΔA2 of the second pulsating wave signal; A pulsating component calculating step for calculating the ratio Φ of the pulsating component of the pulsating wave signal and the pulsating component of the second pulsating wave signal, and a spectrum calculating step for calculating the frequency spectrum or the frequency power spectrum a predetermined number of times in a predetermined period. And a spectrum averaging calculation step of averaging the spectrum calculated for the predetermined number of times, and a fundamental frequency computation step of computing a fundamental frequency of the pulse wave signal from the spectrum averaging. To do. Thereby, in order to calculate the difference between the pulsation components and obtain the fundamental frequency of the pulse wave, the arithmetic mean of the spectrum is calculated to make it easier to extract the fundamental frequency of the pulse wave.

A pulse wave signal processing method according to a twenty-second aspect of the present invention is the pulse wave signal processing method according to any one of the twentieth and twenty-first aspects, in which the pulse frequency signal processing step is performed by the fundamental frequency operation step. At least the first pulse wave signal or the second pulse wave signal is filtered by a filter forming step of forming a filter using the fundamental frequency and its harmonic frequency, and a filter formed by the filter forming step. And a filtering step to perform. This is to form a filter for forming the pulse wave by extracting the frequency components of the fundamental frequency and the harmonics thereof from the pulse wave signal.

The pulse wave signal processing method according to claim 23 of the present invention is the pulse wave signal processing method according to claim 22,
In the filter forming step, when there are a plurality of basic frequencies calculated by the basic frequency calculating step, a filter is formed by using each basic frequency and the frequency of each harmonic thereof. This is to form a filter for forming a pulse wave by extracting frequency components of the respective fundamental frequencies and their respective harmonics even when the fundamental frequency of the pulse wave signal fluctuates.

A pulse wave signal processing method according to a twenty-fourth aspect of the present invention is the pulse wave signal processing method according to the twenty-second aspect or the twenty-third aspect, in the filter forming step,
The characteristic of the filter is a rectangular shape having a width centered on the fundamental frequency and its harmonic frequency. This is because the pulse wave signal is formed by taking in the frequency components around the fundamental frequency and the harmonics thereof.

A pulse wave signal processing method according to a twenty-fifth aspect of the present invention is the pulse wave signal processing method according to the twenty-second or the twenty-third aspect, in the filter forming step,
It is characterized in that the filter is a Gaussian characteristic filter centered on the fundamental frequency and its harmonic frequencies. This is because the frequency component around the fundamental frequency and the frequency of its harmonics is taken in in a Gaussian distribution to form a pulse wave signal.

The pulse wave signal processing method according to claim 26 of the present invention is the signal processing method according to any one of claims 22 to 25, further comprising the first signal in the filtering step. Or displaying the filtered one of the second signals. This is for displaying the filtered pulse wave signal.

A pulse wave signal processing method according to a twenty-seventh aspect of the present invention is the signal processing method according to any one of the twenty-second to twenty-fifth aspects, wherein the first pulse wave signal is an artery of a living body. Is a pulse wave signal obtained by red light transmitted or reflected, the second pulse wave signal is a pulse wave signal obtained by infrared light transmitted or reflected by an artery of a living body,
In the filtering step, both the first pulse wave signal and the second pulse wave signal are filtered, and further, the first pulse wave signal or the second pulse wave signal filtered in the filtering step. And an oxygen saturation calculating step of calculating the oxygen saturation by using. This is for accurately calculating the oxygen saturation using the filtered pulse wave.

A pulse wave signal processing method according to a twenty-eighth aspect of the present invention is a fundamental frequency detecting step of detecting a frequency of a pulse or a heartbeat in a living body, and a frequency of the pulse or the heartbeat detected by the fundamental frequency detecting step. And a filter forming step of forming a filtering frequency by using the frequency of its harmonic, and a filtering step of filtering the pulse wave signal by using the filter formed by the filter filter. With this, the fundamental frequency is detected from the pulse or heartbeat frequency, the pulse wave is filtered by a filter that uses the frequency of this fundamental frequency and its harmonics, and the characteristic frequency component of the pulse wave is extracted to form the pulse wave. This is because

The pulse wave signal processing method according to claim 29 of the present invention is the pulse wave signal processing method according to claim 28,
In the filter forming step, when there are a plurality of basic frequencies calculated by the basic frequency calculating step, a filter is formed by using each basic frequency and the frequency of each harmonic thereof. This allows
This is because even if the fundamental frequency of the pulse wave fluctuates, each fundamental frequency component and the frequency components of its harmonics are extracted to form a pulse wave.

A pulse wave signal processing method according to claim 30 of the present invention is the pulse wave signal processing method according to claim 28 or claim 29, wherein in the filter forming step,
The characteristic of the filter is a rectangular shape having a width centered on the fundamental frequency and its harmonic frequency. This is because the pulse wave signal is formed by taking in the frequency components around the fundamental frequency and the harmonics thereof.

The pulse wave signal processing method according to claim 31 of the present invention is the pulse wave signal processing method according to claim 28 or claim 29, wherein in the filter forming step,
It is characterized in that the filter is a Gaussian characteristic filter centered on the fundamental frequency and its harmonic frequencies. This is because the frequency component around the fundamental frequency and the frequency of its harmonics is captured in a Gaussian distribution to extract the pulse wave signal.

A pulse wave signal processing method according to a thirty-second aspect of the present invention is the signal processing method according to any one of the twenty-eighth through thirty-first aspects, further comprising the pulse wave signal filtered in the filtering step. And a step of displaying. This is for displaying the pulse wave filtered by this.

A pulse wave signal processing method according to a thirty-third aspect of the present invention is the signal processing method according to any one of the twenty-eighth to thirty-first aspects, wherein the pulse wave signal passes through an artery of a living body or A first pulse wave signal obtained by the reflected infrared light and a second pulse wave signal obtained by the red light transmitted or reflected by the artery of the living body, wherein the first pulse wave and the first pulse wave signal are obtained in the filtering step. An oxygen saturation calculation step of calculating any of the two pulse wave signals, and further calculating an oxygen saturation using the first pulse wave and the second pulse wave filtered in the filtering step. It is characterized by This is for accurately calculating the oxygen saturation using the filtered pulse wave.

[0040]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of a pulse wave signal processing and a pulse oximeter using the same according to the present invention will be described in detail with reference to the drawings. FIG. 1 is a pulse oximeter according to an embodiment of the present invention. The probe 1 includes a light emitting unit 2 and a light receiving unit 3, and a fingertip (living tissue) 4 is sandwiched between them. The light emitting unit 2 is
It has two light emitting diodes which respectively emit red light (wavelength λ1: 660 nm) which is the first wavelength light and infrared light (wavelength λ2: 940 nm) which is the second wavelength light. The light emitting section 2 is driven by the light emitting section drive circuit 5, and red light and infrared light are emitted alternately. The light receiving section 3 includes a photodiode, receives the transmitted light from the fingertip, and outputs an electric signal according to the intensity of the transmitted light. The output signal of the light receiving unit 3 is amplified by the received light signal amplification circuit 6 and demodulated by the demodulation circuit 7. The demodulation circuit 7 separately outputs signals corresponding to red light and infrared light. These signals are amplified by amplifiers 9a and 9b,
The signals are converted into digital signals by the / D converters 10a and 10b and input to a CPU (central processing unit) 8. The CPU 8 is the demodulation circuit 7
And the A / D converter 1 while controlling the light emitting unit drive circuit 5.
The signals given from 0a and 10b are processed and the result is output to the display unit 11. The present invention is characterized by pulse wave noise removal processing in this signal processing. The results output to the display unit are the pulse wave waveform from which noise has been removed, the pulse rate, and the SpO2 value (oxygen saturation).

FIG. 2 shows a rough processing flow of this apparatus. When the measurement is started (S10), red light and infrared light emitted from the light emitting unit 2 of the probe 1 are alternately emitted,
Red light and infrared light transmitted through the living tissue 4 are alternately received by the light receiving portion 3
Detected by. Then, the received light signal amplification circuit 6 amplifies the signal, and the demodulation circuit 7 separates it into red light and infrared light.
A / D conversion is performed by the converters 10a and 10b. As a result, the pulse wave is detected (S11). Next, with respect to each pulse wave of infrared light and red light, Fourier transform for a predetermined period is performed, and the absolute value of each spectrum and the value of the power spectrum are obtained. Although not described in detail in this specification, the method of obtaining the power spectrum can also be obtained by performing Fourier transform after obtaining the autocorrelation of each of the infrared light and the red light in a predetermined period.
Then, the fundamental frequency of the pulse wave is calculated based on the spectrum calculation of infrared light and red light (S1).
2). This corresponds to the frequency of the pulse. Then, a matched filter that transmits the fundamental frequency of the pulse wave and the harmonic frequency thereof is formed (S13). The pulse wave is filtered using this matched filter (S14). The filtered pulse wave is displayed on the display unit 11, and the pulse rate and the SpO2 value (oxygen saturation) obtained from the filtered pulse wave are also displayed on the display unit 11. This series of processes is continuously repeated. Hereinafter, the calculation of the fundamental frequency of the pulse wave S12 in FIG. 2 will be described in detail.

[0042] The first of the first aspect of the embodiment of the embodiment will be described with reference to FIGS. 3-7. The pulse intervals for infrared light (IR) and red light (RD) have sampling intervals of 16 msec. At the top of FIG. 3, noise-containing infrared light (IR) and red light (RD) are shown. The period [i] is the period corresponding to the number of 1024 samples, which is 16.384 seconds (= 1024 *
16 msec). First, Fourier transform is performed using 1024 data samples of each of infrared light (IR) and red light (RD) in the period [i] to obtain the absolute value of the spectrum. This spectrum is shown in Fig. 4.
It is (a) and FIG.4 (b). Infrared light (I
R) spectrum (Spc.IR) and red light (RD) spectrum (Spc.RD) are shown in FIG. As can be seen from this, until the Fourier transform is performed on the pulse wave to obtain the spectrum, the spectrum of the noise component is large and it is difficult to detect the fundamental frequency of the pulse wave. Therefore, the following (1)-
It has been found that the fundamental frequency of the pulse wave can be easily detected by performing any of the calculations in (3). The calculation here is the sum quotient calculation of each spectrum for the same frequency. What is common to the operations (1) to (3) is to use the difference between the infrared light (IR) spectrum (Spc.IR) and the red light (RD) spectrum (Spc.RD) as the molecule. . (1) (Spc.IR- Spc.RD) /Spc.RD (2) (Spc.IR- Spc.RD) /Spc.IR (3) (Spc.IR- Spc.RD) / (Spc.IR + Spc.RD
) In addition, as an equivalent arithmetic expression of these, the following (1) '
(3) 'may be used. (1) '{1- (Spc.RD / Spc.IR)} / (Spc.R
D / Spc.IR) (2) '{1- (Spc.RD / Spc.IR)} / (Spc.I
R / Spc.IR) (3) '{1- (Spc.RD / Spc.IR)} / {1+
(Spc.RD / Spc.IR)} Here, what is common to the operations (1) 'to (3)' is
As a molecule, {1- (Spc.RD / Spc.IR)} is used. In order to prevent the denominator from becoming zero, a predetermined value may be added to the denominator in advance, and the calculation result is substantially the same as the calculation result in which the predetermined value is not added. The same applies to the calculation of the spectrum shown below.

An example of this result is shown in FIGS. 5 (a) and 5 (b).
Shown in. FIG. 5 (a) is a graph of the calculation result of the spectrum when the pulse hardly changes in the period [i]. In the graph of FIG. 5A, the calculation results of the above (1), (2) and (3) are shown. As shown in this graph, the pulse hardly fluctuates in the period [i], so that the spectrum at the fundamental frequency of the pulse wave is outstanding, and the frequency is determined by almost one point. Since the second high frequency is also emphasized for the normalizing operation, it is easy to determine the fundamental frequency. Moreover, it is shown that the fundamental frequency of the pulse wave can be obtained by any of the calculation results of (1), (2), and (3). FIG. 5B is a graph of the calculation result of the spectrum when the pulse greatly fluctuates in the period [i]. In the graph of FIG. 5B, the calculation results of (1), (2), and (3) are shown. As shown in this graph, since the pulse fluctuates greatly in the period [i], it can be seen that the spectrum at the fundamental frequency of the pulse wave stands out, but the frequency has a range.

Here, it will be explained that the graph of the calculation result of this spectrum becomes as shown in FIG. There is a difference between the amplitudes of the pulse waves of infrared light (IR) and red light (RD) that do not contain noise, and the amplitude of the pulse wave of infrared light (IR) is usually larger. On the other hand, the amplitude of noise due to body movement or the like may be larger than the amplitude of the pulse wave of infrared light (IR) or red light (RD). Often the noise amplitudes are about the same. Therefore, when Fourier transform is performed on each pulse wave of infrared light (IR) and red light (RD) containing noise, the spectrum is obtained at the fundamental frequency of the pulse wave in the infrared light (IR) pulse wave spectrum. Red light (R
D) Greater than that in the pulse wave spectrum. On the other hand, the noise spectrum has a relatively small difference. Therefore,
If the difference between the infrared light (IR) pulse wave spectrum and the red light (RD) pulse wave spectrum is taken for each frequency, the infrared light (IR) and red light (RD) that are originally in the spectrum of the fundamental frequency of the pulse wave A difference in the spectrum appears due to the difference between the amplitudes of the respective pulse waves of the. Further, the difference in the frequency of the noise component is suppressed because the noise amplitudes of infrared light (IR) and red light (RD) are often about the same. Therefore, in the calculations of (1) to (3) above, the numerator (
Spc.IR − Spc.RD). In addition, Spc.RD, Spc.IR, or (Spc.IR + S
pc.RD), the difference of the noise component spectrum is divided by the noise component spectrum,
Furthermore, the spectrum of the noise component is suppressed. on the other hand,
Even if the spectral difference at the fundamental frequency of the pulse wave is divided by Spc.RD, Spc.IR, or (Spc.IR + Spc.RD) as the denominator, it is not suppressed as much as noise but stands out. However, when the pulse greatly changes during the period of the data that is the target of the Fourier transform, it means that the fundamental frequency of the pulse wave changes.
As shown in (b), the fundamental frequency of the pulse wave has a width. Since the obtained spectrum is normalized, it can be expressed in the range of approximately 0.4 to 1 and has a role as a discriminant function.

Next, the "formation of filter" S in FIG.
Step 13 will be described. As shown in FIG. 5A, when the pulse hardly changes, the fundamental frequency (f
s) to form a filter. As shown in FIG. 6A, the filtering frequencies are 2 fs, 3 fs, and 4 f, which are the fundamental frequency fs of the pulse wave and the harmonics thereof.
s, ..., N · fs. Alternatively, as shown in FIG. 6B, the filtering frequencies are the fundamental frequency fs of the pulse wave and 2fs, 3fs, and 4 which are the frequencies of the harmonics thereof.
A rectangular shape having a predetermined width (band) around fs, ..., N · fs may be used. Alternatively, as shown in FIG. 6C, a Gaussian characteristic filter centering on the fundamental frequency fs of the pulse wave and its harmonic frequencies 2fs, 3fs, 4fs, ..., Nfs can be used. Good. In addition,
Although various methods of detecting fs are conceivable, in the calculation result shown in FIG. 5A, for example, (1) a peak comes from a frequency range corresponding to the maximum possible range of fluctuation of the pulse rate in the living body. (2) Provide a sufficient threshold that will appear as a spectrum value if it is the fundamental frequency of the pulse wave statistically,
Of the frequencies whose spectrum values exceed this threshold, the first frequency with the lowest frequency is obtained. (3) Sufficient threshold that will appear if it is the fundamental frequency of the pulse wave statistically There is a method in which a frequency is provided, and among the frequencies whose spectrum values exceed this threshold, the frequency that appears as an integral multiple and is recognizable as a harmonic is the lowest frequency. Figure 5
When the pulse fluctuates greatly as shown in (b), the fundamental frequency of the pulse wave has a width of two. In the frequency band having this width, the maximum frequency is fhigh and the minimum frequency is flow. Then, with the center frequency as fc, fc = √ (f low) × √ (f high) is defined. Although various methods of detecting fhigh and flow are conceivable, in the calculation result shown in FIG. 5B, for example, from (1) the frequency range corresponding to the maximum possible range of fluctuations of the pulse rate in the living body, , Provide a sufficient threshold that will appear as a spectrum value if it is the fundamental frequency of the pulse wave statistically, and the lowest frequency band in the frequency band where the spectrum value exceeds this threshold. The minimum frequency value of the
Set the maximum frequency value to fhigh. (2) Provide a sufficient threshold that will appear if it is the fundamental frequency of the pulse wave statistically, and set the threshold of the frequency band where the spectrum value exceeds this threshold. Among them, there is a method of setting the minimum frequency value to flow and the maximum frequency value to fhigh of the lowest frequency band among the frequency bands in which the frequency bands appear to be almost integral multiples and can be recognized as harmonics. Also, fc
Instead of the above, fc = (f low + f high) / 2 may be used for the calculation of

There are various methods of forming the filtering frequency, and here, seven methods shown in FIG. 7 will be described. The range of the filtering frequency of the method shown in FIG. 7A is fhigh to flow, and the range of this harmonic is 2fhigh to 2flow and 3fhigh to 3f.
It is a rectangular shape in which low, ..., N · fhigh to n · flow are set. The range of the filtering frequency of the method shown in FIG. 7B is fhigh to flo including the center frequency fc.
The range of w and the range of this harmonic is the range with the width of fhigh to flow centered around 2fc and 3f.
The shape is such that a range having the same width as fhigh to flow centered on c is set, ..., A range having the same width as fhigh to flow centered on n · fc. In the method shown in FIG. 7C, fhigh to flow are divided into a predetermined number, and the center frequencies of the divided frequency widths are fs-1, fs-2, fs-3 ,.
-N. And the range of filtering frequency is
The range from fhigh to flow and the range of this harmonic are 2
fs-1, 2fs-2, 2fs-3, ..., 2fs-
Range with the same width as the frequency width divided above with m as the center frequency, 3fs-1, 3fs-2, 3fs-
3, ..., 3 fs-m as a center frequency, a range having the same width as the divided frequency width, ..., Nf
s-1, n-fs-2, n-fs-3, ..., N-f
It has a rectangular shape in which a range having the same width as the divided frequency width is set with s-m as the center frequency. FIGS. 7D to 7F are examples of the case where a Gaussian characteristic filter is provided instead of the rectangular shape in FIGS. 7A to 7C. That is, FIG. 7D shows a range of fhigh to flow which is a band of the filtering frequency of FIG. 7A, 2fhigh to 2flow, 3fhigh to 3flow, ...
, And n · fhigh to n · flow areas are the same, and a Gaussian characteristic filter is set corresponding to each band. FIG. 7E shows fhigh to f including the center frequency fc which is the band of the filtering frequency of FIG. 7B.
Range of low, range with the same width as width of fhigh to flow centered on 2fc, fhigh to f centered on 3fc
The range with the same width as the low width, ..., The area of the range with the same width as fhigh to flow centered on n · fc is the same, corresponding to each band, The Gaussian characteristic filter is set. FIG. 7 (f) shows
The range of fhigh to flow divided into a predetermined number, which is the band of the filtering frequency of FIG. 7C, and 2fs-1,
2fs-2, 2fs-3, ..., 2fs-m, with a center frequency divided into ranges having the same width as the frequency width, 3fs-1, 3fs-2, 3fs-3 ,.
Area having the same width as the frequency width divided with fs-m as the center frequency, ..., Area of the range having the same width as the frequency width divided with n · fs-m as the center frequency Similarly, the Gauss characteristic filter is set corresponding to each band. FIG. 7 (g) shows
The filtering frequency is set at the level of the Fourier transform resolution. Here, the data length used for the Fourier transform is 16.384 [seconds] (= 1024 × 16 ms).
ec), the resolution is 0.061 [Hz] = 1 /
It is 16.384 [seconds]. So from low to high
Frequency of every 0.061 [Hz] which is the resolution up to, ie, flow, flow +0.061 [Hz],
flow +0.061 [Hz] x2, ..., flow +
The fundamental frequency of the pulse wave is up to 0.061 [Hz] × n, ..., Hhigh. Next, a frequency that is an integral multiple of each fundamental frequency of these pulse waves is taken as a harmonic frequency. Ie 2
Frequency from flow to 2fhigh 0.061 [Hz] x frequency interval of 2 0.0 from 3flow to 3fhigh
The frequency having a frequency interval of 61 [Hz] × 3 and the frequency having a frequency interval of 0.061 [Hz] × n from n · flow to n · fhigh are the frequencies of the harmonics. The fundamental frequency and the harmonic frequency are adopted as the filtering frequency. The basic concept of forming the filtering frequency described with reference to FIGS. 7A to 7G is that when a plurality of basic frequencies occur due to fluctuations in the pulse wave, the filtering frequencies are the respective basic frequencies and their harmonics. It is the frequency obtained by ORing the frequencies of.

Here, the reason for using the fundamental frequency of the pulse wave and the frequencies of its harmonics as a method of forming the filtering frequency will be described. The FFT has low resolution at low frequencies and good resolution at high frequencies. Therefore, when the wave is fluctuating, the fluctuation cannot be extracted by filtering with a low frequency, but can be extracted by filtering with a harmonic frequency. Therefore, assuming that the fundamental frequency of the pulse wave also fluctuates,
Fluctuations of the pulse wave can also be extracted by filtering by the frequency of the harmonic.
Therefore, in order to extract the pulse wave waveform as faithfully as possible, it is necessary to perform filtering up to the frequency of a considerable harmonic wave. On the other hand, in order to measure the SpO2 value, it is sufficient if the amplitude of the pulse wave can be detected as a calculation, so filtering up to the frequency of the harmonic is not always necessary,
Filtering of the second to fourth order is sufficient. What high frequency to use for filtering may be appropriately determined. Note that if the filtering frequency is too wide, the filtered pulse wave will likely contain noise, so the filtering frequency is set appropriately.

The matched filter obtained from 1024 pieces of data of the period [i] in FIG. 3 is the period before the period [i] and the period [i + 1] forming the next matched filter are started. Used to filter [I] pulse waves. Then, the matched filter obtained from the data of the period [i + 1] forming the next matched filter is the period [i] and the period [I + 1 before the period forming the next matched filter is started. ] Is used to filter the pulse wave. The setting of the period for forming the matched filter may be set for each predetermined period, or the biological information parameter such as the heart rate obtained by the electrocardiogram measurement, the pulse rate obtained by the pulse wave measurement is more than a predetermined value. It may be provided if it fluctuates.

In this embodiment, the sampling interval is 16 ms.
ec, the period for forming the matched filter is 1024
The period is equal to the number of samples, but the period is not limited to this.

Second Embodiment Next, a second embodiment will be described with reference to FIG. Similar to the first embodiment, the sampling interval of the pulse wave is 16 msec and the number of data used to form the matched filter is 1024. Then, Fourier transform is performed from the data in the period [i] to obtain the absolute value of the spectrum, and the following (1) to (3)
It is the same as the first embodiment that the fundamental frequency of the pulse wave is obtained by performing any one of the operations to form the matched filter. (1) (Spc.IR − Spc.RD) / Spc.RD (2) (Spc.IR − Spc.RD) / Spc.IR (3) (Spc.IR − Spc.RD) / (Spc.IR + Spc.R
D) In addition, as the equivalent arithmetic expressions of these, the following (1) '
(3) 'may be used. (1) '{1- (Spc.RD / Spc.IR)} / (Spc.R
D / Spc.IR) (2) '{1- (Spc.RD / Spc.IR)} / (Spc.I
R / Spc.IR) (3) '{1- (Spc.RD / Spc.IR)} / {1+
(Spc.RD / Spc.IR)} The feature of the second embodiment is that the period for forming the matched filter is continuous. That is, regarding FIG. 8, the matched filter formed using the data in the period [i] is used to filter the pulse wave in the period [i]. Then, for consecutive periods [i +
In 1], a new matched filter is formed using the data in that period, and it is used to filter the pulse wave in that period [i +1]. Next consecutive period
The same applies to [i + 2]. In this way, the matched filter is continuously formed for each predetermined period, and the pulse wave is filtered using the matched filter.

Third Embodiment Next, a third embodiment will be described with reference to FIG. Similar to the first embodiment, the sampling interval of the pulse wave is 16 msec and the number of data used to form the matched filter is 1024. Then, Fourier transform is performed from the data in the period [i] to obtain the absolute value of the spectrum, and the following (1) to (3)
It is the same as the first embodiment that the fundamental frequency of the pulse wave is obtained by performing any one of the operations to form the matched filter. (1) (Spc.IR- Spc.RD) /Spc.RD (2) (Spc.IR- Spc.RD) /Spc.IR (3) (Spc.IR- Spc.RD) / (Spc.IR + Spc .RD) In addition, as the equivalent arithmetic expressions of these, the following (1) '
(3) 'may be used. (1) '{1- (Spc.RD / Spc.IR)} / (Spc.R
D / Spc.IR) (2) '{1- (Spc.RD / Spc.IR)} / (Spc.I
R / Spc.IR) (3) '{1- (Spc.RD / Spc.IR)} / {1+
(Spc.RD / Spc.IR)} The feature of the third embodiment is that the processing is performed while partially overlapping the periods for forming the matched filter. That is, regarding FIG. 9, the matched filter formed using the data in the period [i] is used to filter the pulse wave in the period [i]. The starting point of the period [i + 1] for forming the next new matched filter is after the starting point of the period [i], for example, a period corresponding to 64 data samples. The matched filter formed using the data of the period [i + 1] filters the pulse wave of 64 samples in the period [I + 1] from the end of the period [i] to the end of the period [i + 1]. Used to do. Further, the starting point of the period [i + 2] for forming the next new matched filter is after the period of the same data 64 samples at the starting point of the period [i + 1]. period[
The matched filter formed using the data of i + 2] filters the pulse wave of 64 samples in the period [I + 2] from the end of the period [i + 1] to the end of the period [i + 2]. Used to do. Similarly, the period [I
+ 3], the matched filter used to filter the pulse wave of 64 samples of data is
+ 2], which is formed from the data of the period [i + 3] that starts with a delay of 64 samples from the start point. In this way, while shifting the period for forming the matched filter by the predetermined number of samples, the formed matched filter is used to filter the terminal period for the shifted period.

In this embodiment, the sampling interval is 16 ms.
ec, the period for forming the matched filter is 1024
Although the period corresponding to the number of individual samples and the period during which the matched filter is formed are shifted to the period corresponding to the number of 64 samples, the period is not limited to this.

Fourth Embodiment Next, a fourth embodiment will be described with reference to FIG. Similar to the first embodiment, the pulse wave sampling interval is 16 msec and the number of data units for obtaining the spectrum is 1024. The feature of the fourth embodiment is that the periods for obtaining the absolute value of the spectrum of the pulse wave are set by shifting for a plurality of periods, the obtained absolute values of the spectra are averaged, and then a matched filter is formed. It is in. Period [i] and period [i + 1] shown in FIG.
, Period [i + 2], ..., period [i + n], ...
Is a period set by shifting by 64 samples of data. Period [i], Period [i + 1], Period [i + 2
], ..., In the period [i + n], ..., 1024 data samples are used to perform Fourier transform to obtain the absolute value of the spectrum. An example showing the spectrum (absolute value; the same applies hereinafter) is the same as FIG. 4A and FIG. 4B shown in the first embodiment. Figure 4 (a)
Fig. 4 shows the spectrum of infrared light (IR) (Spc.IR).
A red light (RD) spectrum (Spc.RD) is shown in (b). Period [i] ~ Period [i + in period [I]
Each of the n spectra obtained by the Fourier transform in [n] is added and averaged for each frequency component. That is, infrared light (IR) spectra are added and averaged (Av.Sp
c.IR) and the spectrum of the red light (RD) spectrum are averaged (Av.Spc.RD). And this average infrared light spectrum (Av.Spc.IR) and average red light spectrum (Av.Sp
c.RD) is used to detect the fundamental frequency of the pulse wave by performing any of the following operations (4) to (6). The calculation here is the sum quotient calculation of each spectrum for the same frequency. What is common to the operations (4) to (6) is to use the difference between the average infrared light spectrum (Av.Spc.IR) and the average red light spectrum (Av.Spc.RD) as the numerator. . (4) (Av.Spc.IR-Av.Spc.RD) /Av.Spc.RD (5) (Av.Spc.IR-Av.Spc.RD) /Av.Spc.IR (6) (Av.Spc.IR) Spc.IR − Av.Spc.RD) / (Av.Spc.IR ＋
Av.Spc.RD) Note that the following (4) '-
(6) 'may be used. (4) '{1- (Av.Spc.RD / Av.Spc.IR)} / (Av.Spc.R
D / Av.Spc.IR) (5) '{1- (Av.Spc.RD /Av.Spc.IR)} / (Av.Sp
c.IR/Av.Spc.IR) (6) '{1- (Av.Spc.RD / Av.Spc.IR)} / {1+ (Av.
Spc.RD / Av.Spc.IR)}

In the period [I], when the pulse hardly changes, the spectrum at the fundamental frequency of the pulse wave stands out, so it is decided at almost one point. When the pulse greatly changes, the spectrum at the fundamental frequency of the pulse wave changes. The fact that there is a wide range of frequencies, which is conspicuous,
5 (a) and 5 (b) shown in the embodiment of FIG. Also, regarding the method of forming the matched filter,
This is the same as the method described in the first embodiment with reference to FIGS. 6A to 6C and FIGS. 7A to 7G. Then, the obtained matched filter is used for filtering the pulse wave in the period [I].

Next, the period [i +] in the period [I + 1]
1] to the period [i + n +1] in each of the n spectrums obtained by Fourier transform, the infrared light (IR) and the red light (RD) are added and averaged for each frequency component,
An average infrared light spectrum (Av.Spc.IR) and an average red light spectrum (Av.Spc.RD) are calculated. Then, by using the average infrared light spectrum (Av.Spc.IR) and the average red light spectrum (Av.Spc.RD), any one of the above (4) to (6) is calculated. , Detect the fundamental frequency of pulse wave. Regarding the method of forming the matched filter using the fundamental frequency of the pulse wave, as described above, FIG.
This is the same as the method described in the first embodiment using (c) and FIGS. 7 (a) to 7 (g). The matched filter formed using the data of this period [I + 1] is
From the end of [] to the end of period [I + 1] [N + 1
] Is used to filter 64 samples of data.

In this embodiment, the sampling interval is 16 ms.
ec, the unit for obtaining the spectrum is the period corresponding to the number of 1024 samples, and the period for shifting the period for forming the matched filter is the period corresponding to the number of 64 samples, but the present invention is not limited to this. Further, the period for which the period for forming the matched filter is shifted may be the period for forming the matched filter (the period corresponding to the number of 1024 samples). At this time, the period of filtering by the matched filter is 1024 pieces of data.

Here, the advantage of obtaining the fundamental frequency of the pulse wave by using the arithmetic mean of the spectrum is that the noise due to the body motion unrelated to the pulse wave is canceled and disappears by performing the arithmetic mean of n times. , The S / N ratio is improved by √ (n) times.

Fifth Embodiment Next, a fifth embodiment will be described with reference to FIGS. 3, 11 and 12. A feature of the present embodiment is that the difference between the pulse waves of infrared light (IR) and red light (RD) is Fourier-transformed to obtain the spectrum when obtaining the fundamental frequency of the pulse wave. The pulse wave sampling interval for infrared light (IR) and red light (RD) is 16 msec,
The number of data used to form the matched filter is 10
It is the same as that of the first embodiment that the number is 24. First, the difference between the pulse waves of infrared light (IR) and red light (RD) in the period [i] is calculated, and this is taken as aI-aR (= infrared light (IR) pulse wave-red light (RD) pulse). Let's call it a wave
The difference pulse wave is Fourier transformed to obtain the absolute value of the spectrum. An example of this is shown in FIG. As you can see,
Until the Fourier transform is performed on the pulse wave difference to obtain the spectrum, the spectrum of the noise component is large, and it is difficult to detect the fundamental frequency of the pulse wave. Therefore, the following (7) to (9)
It has been found that the fundamental frequency of the pulse wave can be easily detected by performing any one of the calculations. The calculation here is the sum quotient calculation of each spectrum for the same frequency. What is common to the operations (7) to (9) is, as a numerator,
This is to use a spectrum (Spc. (AI-aR)) obtained by Fourier transforming the difference between the pulse waves of infrared light (IR) and red light (RD). (7) (Spc. (AI − aR)) / Spc.aI (8) (Spc. (AI − aR)) / Spc.aR (9) (Spc. (AI − aR)) / (Spc.aI + Spc.aR
) Here, Spc.aI: spectrum of infrared light (IR) pulse wave Spc.aR: spectrum of red light (RD).

An example of the calculation result of (9) is shown in FIG.
Shown in. The example of FIG. 12 is a graph of the calculation result of the spectrum when the pulse hardly changes during the period [i]. Therefore, the spectrum at the fundamental frequency of the pulse wave is outstanding, and the frequency is determined by almost one point. When the pulse greatly fluctuates in the period [i], the spectrum at the fundamental frequency of the pulse wave stands out, but the frequency has a range.

If the fundamental frequency of the pulse wave can be determined by the above operations (7) to (9), the above-described first embodiment will be described with reference to FIGS. 6A to 6C and 7A to 7A. Determine the filtering frequency to form a matched filter, as described with g). That is, when the pulse hardly changes, as shown in FIG. 6 (a), the fundamental frequency and its harmonics are used as the filtering frequencies, or
Alternatively, as shown in FIGS. 6B and 6C, the filtering frequency is determined by using a rectangular shape in which the frequency of the fundamental frequency and its harmonics have a width or a Gaussian filter. When the pulse fluctuates significantly, the fundamental frequency of the pulse wave has a range. Therefore, as shown in FIGS. 7 (a) to 7 (g), the fundamental frequency of the pulse wave varies depending on the pulse frequency. Then, the filtering frequency is set by giving a width to the filtering frequency or by using a Gaussian filter together with the frequency of the harmonic.

The matched filter obtained by 1024 pieces of data of the period [i] in FIG. 3 is the period before the period [i] and the period [i + 1] forming the next matched filter are started. Used to filter [I] pulse waves. Then, the matched filter obtained from the data of the period [i + 1] forming the next matched filter is the period [i] and the period [I + 1 before the period forming the next matched filter is started. ] Is used to filter the pulse wave. The period for forming the matched filter may be set for each predetermined period, or may be set when the biometric information parameters such as the heart rate and the pulse rate change by a predetermined amount or more.

In this embodiment, the sampling interval is 16 ms.
ec, the period for forming the matched filter is 1024
The period is equal to the number of samples, but the period is not limited to this.

Sixth Embodiment Next, a sixth embodiment will be described with reference to FIG. The pulse wave sampling interval is 16 msec, the number of data used to form the matched filter is 1024, and the difference between the pulse waves of infrared light (IR) and red light (RD) is Fourier-transformed. The method of obtaining the fundamental frequency from the calculation of any of the above (7) to (9) using the obtained spectrum and the method of forming the matched filter from the fundamental frequency of the pulse wave are the same as in the fifth embodiment. is there. Fifth
The embodiment is characterized in that the period for forming the matched filter is continuous. That is, regarding FIG. 8, the matched filter formed using the data in the period [i] is used to filter the pulse wave in the period [i]. Then, in the continuous period [i + 1], a matched filter is newly formed using the data in that period, and it is used to filter the pulse wave in that period [i + 1]. Next consecutive period [i +
2] is also the same. In this way, the matched filter is continuously formed for each predetermined period, and the pulse wave is filtered using the matched filter.

Seventh Embodiment Next, a seventh embodiment will be described with reference to FIG. The pulse wave sampling interval is 16 msec, the number of data used to form the matched filter is 1024, and the difference between the pulse waves of infrared light (IR) and red light (RD) is Fourier-transformed. The method of obtaining the fundamental frequency from the calculation of any of the above (7) to (9) using the obtained spectrum and the method of forming the matched filter from the fundamental frequency of the pulse wave are the same as in the fifth embodiment. is there. 7th
The embodiment of (1) is to perform processing while partially overlapping the periods of forming the matched filter. That is, regarding FIG. 9, the matched filter formed using the data in the period [i] is used to filter the pulse wave in the period [i]. The starting point of the period [i + 1] to form the next new matched filter is
For example, after the start point of the period [i] is a period corresponding to the number of data samples of 64 samples. The matched filter formed using the data of the period [i +1] is from the end of the period [i] to the period [
It is used for filtering the pulse wave of 64 samples in the period [I + 1] until the end of [i + 1]. Further, the period [i for forming the next new matched filter]
The start point of + 2] is after the same number of 64 samples of data as the start point of the period [i + 1]. The matched filter formed using the data of the period [i + 2] is
period [I + 1] end to period [i + 2] end [I
+2] to filter the pulse wave for 64 samples. Similarly, the matched filter used to filter the pulse wave of 64 samples of data in the period [I + 3] is a period [i + which starts 64 samples later than the starting point of the period [i + 2]. 3
] Is formed from the data. In this way, while shifting the period for forming the matched filter by the predetermined number of samples, the formed matched filter is used to filter the terminal period for the shifted period.

In this embodiment, the sampling interval is 16 ms.
ec, the period for forming the matched filter is 1024
Although the period corresponding to the number of individual samples and the period during which the matched filter is formed are shifted to the period corresponding to the number of 64 samples, the period is not limited to this.

Eighth Embodiment Next, the eighth embodiment will be described with reference to FIG. Similar to the fifth embodiment, the pulse wave sampling interval is 16 msec and the number of data units for obtaining the spectrum is 1024. A feature of the eighth embodiment is that the periods for obtaining the absolute value of the spectrum of the pulse wave are set by shifting for a plurality of periods, the obtained absolute values of the plurality of spectra are averaged, and then a matched filter is formed. It is in. Period [i] and period [i + 1] shown in FIG.
, Period [i + 2], ..., period [i + n], ...
Is a period set by shifting by 64 samples of data. Period [i], Period [i + 1], Period [i + 2
], ..., Then, in the period [i + n], the calculation is performed for every 1024 data samples. Fourier transform of the difference (aI −aR) between the infrared light (IR) pulse wave and the red light (RD) pulse wave yields the spectrum Spc (a
I −aR). Infrared light (IR) pulse wave and / or red light (RD)
The spectrum Spc.IR is obtained by Fourier transforming the pulse wave.
And / or Spc.RD. Each period [i], period [i + 1], period [i + 2],
..., the spectrum Spc (aI for n pieces in the period [i + n]
-AR) is added and averaged to obtain Av.Spc (aI-aR). Each period [i], period [i + 1], period [i + 2],
..., n spectra Spc.IR for the period [i + n]
And / or Spc.RD are averaged to obtain Av.Spc.IR and / or Av.Spc.RD. The fundamental frequency of the pulse wave is obtained by calculating any of the following (10) to (12). (10) (Av.Spc. (AI-aR)) / Av.Spc.RD (11) (Av.Spc. (AI-aR)) / Av.Spc.IR (12) (Av.Spc. (AI) − AR)) / (Av.Spc.IR + Av.
Spc.RD) In common with the operations of (10) to (12) above, as a numerator, the spectrum of the difference between the infrared light (IR) pulse wave and the red light (RD) pulse wave is averaged (Av. Spc. (AI-aR)). The calculation result by the above (12) is almost as shown in FIG. 12, but since the noise component is canceled and disappears by averaging, the fundamental frequency of the pulse wave becomes easier to detect.

In the period [I], when the pulse hardly changes, the spectrum at the fundamental frequency of the pulse wave stands out, so it is decided at almost one point. When the pulse greatly changes, the spectrum at the fundamental frequency of the pulse wave changes. Fig. 5 shows that there is a wide range of frequencies, although it stands out.
(A) -It is the same as that of the 1st embodiment explained using Drawing 5 (b). The method of forming the matched filter is the same as the method described in the first embodiment using FIGS. 6A to 6C and FIGS. 7A to 7G.
Then, the obtained matched filter is used for filtering the pulse wave in the period [I].

Next, the method of forming the matched filter in the period [I + 1] is the same as the method of forming the matched filter in the period [I]. That is, each period [i +
1] ~ Using the data of the period [i + n + 1], the above ~
Is performed to obtain the fundamental frequency of the pulse wave. Then, using this, a matched filter is formed by the method described in the first embodiment with reference to FIGS. 6A to 7G. The matched filter formed by using the data of this period [I + 1] is
It is used for filtering 64 samples of data in the period [N + 1] until the end of [I + 1]. further,
Similarly, the matched filter in the period [I + 2] is used to filter the data in the period [N + 2], and the matched filter in the period [I + 3] filters the data in the period [N + 3]. Used to do. This process is repeated.

In this embodiment, the sampling interval is 16 ms.
ec, a unit for obtaining a spectrum is a period corresponding to the number of 1024 samples, and a period for shifting the period for forming the matched filter is a period corresponding to the number of 64 samples, but the present invention is not limited to this. Further, the period for which the period for forming the matched filter is shifted may be the period for forming the matched filter (the period corresponding to the number of 1024 samples).
At this time, the period of filtering by the matched filter is 1024 pieces of data. Here, points to be considered in designing the algorithm will be described. The resolution of the Fourier transform depends on the reciprocal of the data length used for the Fourier transform. In this embodiment, the data length used for Fourier transform is 16.384 [seconds] (= 1024
X16 msec), the resolution is 0.061 [H
z] = 1 / 16.384 [seconds]. Thus, the required Fourier transform resolution is taken into consideration in design. Further, since it is necessary to continuously display the filtered pulse wave on the display device, (1) the time length of the pulse wave displayed on the display device, and (2) the real time display of the pulse wave are allowed. It is also necessary to consider the time difference between the appearance of the pulse wave appearing in the living body and the display timing on the display, and (3) the time during which the successively detected pulse waves are stored and can be processed by the various embodiments described above.

Ninth Embodiment Next, a ninth embodiment will be described. After the infrared light (IR) pulse wave and the red light (RD) pulse wave are detected, the pulse rate conventionally used is determined using at least one of them. Then, using the pulse frequency as the fundamental frequency of the pulse wave, the matched filter forming method is formed by the matched filter forming method described in the first embodiment with reference to FIGS. The formed matched filter is used to filter the infrared light (IR) pulse wave and the red light (RD) pulse wave. Since the pulse rate is sequentially detected, the matched filter can be sequentially formed in accordance with the pulse rate, and the filtering frequency of the matched filter can be shifted according to the fluctuation of the pulse rate. Since the pulse rate obtained by the pulse wave measurement and the heart rate obtained by the electrocardiogram measurement are almost the same, in the present embodiment, in order to obtain the fundamental frequency of the pulse wave, the electrocardiogram measurement is performed, and the frequency of the heartbeat is calculated. It may be used as the fundamental frequency of the pulse wave.

In the embodiment, the above nine examples are shown.
These may be continuously switched and implemented depending on the situation. For example, at the initial stage of measurement, there is not enough data for averaging the spectra, so the averaging is not performed in the first to third embodiments, or the fifth to seventh and ninth modes.
The embodiment of may be adopted. Alternatively, in the fourth or eighth embodiment, the arithmetic mean number n may be reduced. Then, when sufficient data has been accumulated, the arithmetic mean number n may be gradually increased in the fourth or eighth embodiment. In any case, it can be said that the matched filter is adaptive in that the matched filter is formed according to the fundamental frequency of the pulse wave that changes with time.

In the above embodiment, the absolute value of the spectrum of the pulse wave of each wavelength is used in obtaining the basic period of the pulse wave, but a power spectrum may be used instead. The spectrum according to the present application is a concept including not only the absolute value of the spectrum but also the power spectrum.

Further, in the first to third embodiments, the following operations (1) to (3) are used to specify the fundamental frequency of the pulse wave. (1) (Spc.IR- Spc.RD) /Spc.RD (2) (Spc.IR- Spc.RD) /Spc.IR (3) (Spc.IR- Spc.RD) / (Spc.IR + Spc .RD) Instead of this, the pulsating component of the absorbance of the red light wavelength is ΔA
1. When the pulsating component of the absorbance of the infrared light wavelength is ΔA2, the difference between ΔA1 and ΔA2 on the time axis is obtained, and the difference signal is used by using a data sample in the same predetermined period as described above. The fundamental frequency may be detected by performing the Fourier transform to obtain the absolute value of the spectrum.

The infrared light (IR) pulse wave and the red light (RD) pulse wave filtered by the respective embodiments can be displayed by a display. Either one may be displayed on the display. Further, the SpO2 value (oxygen saturation) can be obtained by an existing method using the filtered infrared light (IR) pulse wave and red light (RD) pulse wave. That is, the pulsation component obtained from the filtered pulse wave of the red light is ΔA1, the pulsation component obtained from the filtered pulse wave of the infrared light is ΔA2, the ratio of absorbance Φ == AA1 / ΔA2 is determined, and the predetermined relation, The oxygen saturation is calculated from the oxygen saturation S = f (Φ). In the above embodiment, the fundamental frequency is obtained from the pulse wave signal, a matched filter is formed, and noise is removed by this filter. However, the calculation of the fundamental frequency in the invention according to the present application, and the application of the processing for removing the noise is not limited to the pulse wave signal, other biological signals, or not limited to the biological signal, and more broadly to industrial signals in general. Applicable. Particularly, it is suitable for application to a continuous signal having periodicity and two signals having the same fundamental frequency. Therefore,
In the above embodiment, the infrared light (IR) pulse wave and the red light (RD) pulse wave can be applied by replacing them with two industrial general signals having the same fundamental frequency. Accordingly, Spc.IR, Spc.RD, Av.Spc.IR, and Av.Spc.RD can be applied by substituting the spectrum obtained by Fourier transforming the industrial general signal.

[0075]

As described in detail above, the first aspect of the present invention
The signal processing method according to claim 1, wherein the first signal IR and the second signal RD have the same fundamental frequency and are continuous, the first signal IR and the second signal RD are processed.
For each of the above, a frequency spectrum or frequency power spectrum is calculated in a predetermined period, and the first spectrum is calculated.
First spectrum Spc.IR of the signal IR of the
Of the second spectrum Spc.RD, and the first spectrum Spc.IR and the second spectrum Sp calculated by the spectrum calculation step.
c. Using RD, on the frequency axis, the first signal IR and the first signal IR based on the result of the spectrum calculation for obtaining the difference between the A fundamental frequency computing step of computing a fundamental frequency of the second signal RD. As a result, since the first signal IR and the second signal RD have the same fundamental frequency, the first signal IR and the second signal RD are used to calculate (Spc.IR
-Spc.RD) or {1- (Spc.RD / Spc.IR)} makes it possible to accurately calculate the fundamental frequency of the signal even if the signal contains noise.

A signal processing method according to claim 2 of the present invention comprises:
In the signal processing method according to claim 1, in the fundamental frequency calculation step, (Spc.IR- Spc.RD) /Spc.RD, (Spc.IR- Spc.RD) /Spc.IR, (Spc.IR). − (Spc.RD) / (Spc.IR + Spc.RD), {1- (Spc.RD / Spc.IR)} / (Spc.RD / Spc.I
R), {1- (Spc.RD / Spc.IR)} / (Spc.IR / Spc.I
R), or {1- (Spc.RD / Spc.IR)} / {1+ (Spc.RD / Sp
c.IR)}, the fundamental frequency is calculated based on the result of the spectrum calculation. This makes it possible to suppress a component caused by noise and make it easier to extract the fundamental frequency.

A signal processing method according to claim 3 of the present invention is
A continuous first signal IR and a second signal having the same fundamental frequency
In the signal processing method for processing the signal R 1 of the above, the frequency spectrum or the frequency power spectrum is calculated for each of the first signal IR and the second signal RD in a predetermined period, and the first signal IR 1 1 spectrum Spc.IR from the second signal RD to the second spectrum
Spc.RD, a spectrum calculation step, a spectrum addition and average calculation step of adding and averaging the first spectrum Spc.IR and the second spectrum Spc.RD for a predetermined number of times by the spectrum calculation step, and the spectrum addition and average The first averaged by the arithmetic step
Using the spectrum Av.Spc.RD and the averaged second spectrum Av.Spc.RD, in addition to spectrum calculation for obtaining mutual difference or spectrum calculation for obtaining mutual difference on the frequency axis, a spectrum to be further normalized Based on the result of the calculation, the first signal IR and the second signal RD
And a fundamental frequency calculation step of calculating the fundamental frequency of. By using the arithmetic mean calculation of the spectrum in this way, the fundamental frequency of the signal can be calculated while suppressing the noise appearing in the spectrum.

The signal processing method according to claim 4 of the present invention is
The signal processing method according to claim 3, wherein in the fundamental frequency calculation step, (Av.Spc.IR-Av.Spc.RD) /Av.Spc.RD, (Av.Spc.IR-Av.Spc.RD). RD) / Av.Spc.IR, (Av.Spc.IR − Av.Spc.RD) / (Av.Spc.IR + Av.Spc.
RD), {1- (Av.Spc.RD / Av.Spc.IR)} / (Av.Spc.RD /Av.S
pc.IR), {1- (Av.Spc.RD /Av.Spc.IR)} / (Av.Spc.IR / A
v.Spc.IR), or {1- (Av.Spc.RD / Av.Spc.IR)} / {1+ (Av.Spc.RD /
Av.Spc.IR)} based on the result of the spectrum calculation. By using the spectrum averaged by this in the calculation,
A component caused by noise can be suppressed, and the fundamental frequency can be more easily extracted.

The signal processing method according to claim 5 of the present invention is
The signal processing method according to any one of claims 1 to 4, further comprising: a filter formation that forms a filter using the fundamental frequency calculated by the fundamental frequency calculation step and its harmonic frequency. And a filtering step of filtering at least the first signal IR or the second signal RD by the filter formed by the filter forming step. As a result, the signal can be formed by extracting the fundamental frequency component, which is the characteristic frequency component of the signal, and the frequency components of the harmonics thereof from the signal from which noise has been removed.

A signal processing method according to claim 6 of the present invention comprises:
The signal processing method according to claim 5, wherein the filter forming step uses each fundamental frequency and its respective harmonic frequency when there are a plurality of fundamental frequencies computed by the fundamental frequency computing step. It is characterized by forming a filter. Thereby, even when the fundamental frequency changes, each fundamental frequency component and the frequency component of each harmonic thereof can be extracted to form a signal.

A signal processing method according to claim 7 of the present invention comprises:
The signal processing method according to claim 5 or 6, wherein in the filter forming step, the characteristic of the filter is a rectangular shape having a width centered on the fundamental frequency and its harmonic frequency, respectively. Characterize. As a result, the signal can be formed by taking in the frequency components around the fundamental frequency and the frequencies of the harmonics thereof.

A signal processing method according to claim 8 of the present invention comprises:
The signal processing method according to claim 5 or 6, wherein in the filter forming step, the filter is a Gaussian characteristic filter centered on the frequencies of the fundamental frequency and its harmonics. As a result, it is possible to form a signal by taking in a Gaussian distribution of the frequency components around the fundamental frequency and its harmonic frequencies.

A signal processing method according to claim 9 of the present invention comprises:
The signal processing method according to any one of claims 1 to 8, wherein the first signal IR and the second signal RD are both pulse wave signals. This makes it possible to obtain the fundamental frequency from the pulse wave signal of the living body and extract the characteristic frequency component to form the pulse wave.

A signal processing method according to claim 10 of the present invention is the signal processing method according to any one of claims 5 to 8, wherein the first signal IR and the second signal RD are Are pulse wave signals, and the step of displaying the filtered signal of the first signal IR or the second signal RD in the filtering step is further included. This makes it possible to extract a characteristic frequency component of the pulse wave from the pulse wave signal of the living body, form the pulse wave, and display the pulse wave.

The signal processing method according to claim 11 of the present invention is the signal processing method according to any one of claims 5 to 8, wherein the first signal IR passes through an artery of a living body or The second signal RD is a pulse wave signal obtained by reflected infrared light, the second signal RD is a pulse wave signal obtained by red light transmitted through or reflected by an artery of a living body, and the first signal in the filtering step. Filtering both the signal IR and the second signal RD, and
An oxygen saturation calculation step of calculating an oxygen saturation using the first signal IR or the second signal RD filtered in the filtering step. Thereby, the oxygen saturation can be accurately calculated using the filtered pulse wave.

A signal processing method according to a twelfth aspect of the present invention is a signal detecting step of detecting a continuous signal having periodicity, and a basic frequency calculation for calculating a basic frequency of the signal detected by the signal detecting step. A step, a filter forming step of forming a filtering frequency by using the fundamental frequency of the signal calculated by the fundamental frequency calculating step and the frequency of its harmonic, and the filter using the filter formed by the filter filter, And a filtering step of filtering. This makes it possible to form a signal by extracting the frequency components of the fundamental frequency and the harmonics thereof with respect to a continuous signal having periodicity.

A signal processing method according to a thirteenth aspect of the present invention is the signal processing method for processing a first signal and a second signal which have the same fundamental frequency and which are continuous, wherein the first signal and the second signal are processed. Of the signal, the basic frequency calculation step of calculating the basic frequency of the signal using at least one of the first signal and the second signal, and the basic frequency calculation step. A filter forming step of forming a filter using the fundamental frequency and the frequencies of its harmonics. As a result, it is possible to form a filter for forming signals by extracting the frequency components of the fundamental frequency and the harmonics of the two consecutive signals having the same fundamental frequency.

A signal processing method according to a fourteenth aspect of the present invention is the signal processing method according to the thirteenth aspect, wherein in the filter forming step, there are a plurality of the basic frequencies calculated by the basic frequency calculating step. Is
It is characterized in that a filter is formed by using each fundamental frequency and the frequency of each harmonic. As a result, even if the fundamental frequencies of the two signals fluctuate, it is possible to form a filter for extracting the frequency components of the respective fundamental frequencies and the respective harmonic frequencies to form the signals.

A signal processing method according to a fifteenth aspect of the present invention is the signal processing method according to the thirteenth or fourteenth aspect, wherein in the filter forming step, the characteristics of the filter are the fundamental frequency and its harmonics. It is characterized in that it has a rectangular shape with a width centered on each of the frequencies. As a result, the signal can be formed by taking in the frequency components around the fundamental frequency and the frequencies of the harmonics thereof.

A signal processing method according to a sixteenth aspect of the present invention is the signal processing method according to the thirteenth or fourteenth aspect, wherein, in the filter forming step, the filter has the fundamental frequency and its harmonic frequencies. It is characterized by being a Gaussian characteristic filter centered on. As a result, it is possible to form a signal by taking in a Gaussian distribution of the frequency components around the fundamental frequency and its harmonic frequencies.

A signal processing method according to a seventeenth aspect of the present invention is the signal processing method according to any one of the twelfth to sixteenth aspects, in which the first signal and the second signal are both Is also a pulse wave signal. With this, the fundamental frequency is obtained from the pulse wave signal of the living body,
A characteristic frequency component can be extracted to form a pulse wave.

A signal processing method according to an eighteenth aspect of the present invention is the signal processing method according to any one of the twelfth to sixteenth aspects, wherein the first signal and the second signal are both Is also a pulse wave signal, and the step of displaying the filtered signal of the first signal or the second signal in the filtering step is further included. Thereby, the filtered pulse wave signal can be displayed.

A signal processing method according to a nineteenth aspect of the present invention is the signal processing method according to any one of the twelfth to sixteenth aspects, wherein the first signal is transmitted or reflected through an artery of a living body. Pulse signal obtained by infrared light, the second signal is a pulse wave signal obtained by red light transmitted through or reflected by an artery of a living body, and the first signal and An oxygen saturation calculation step of calculating any of the second signals, and further calculating an oxygen saturation using the first signal or the second signal filtered in the filtering step. It is characterized by Thereby, the oxygen saturation can be accurately calculated using the filtered pulse wave.

A pulse wave signal processing method according to claim 20 of the present invention is a pulse wave signal processing method for processing a first pulse wave signal and a second pulse wave signal of a living body having the same fundamental frequency,
A first pulsating component detecting step for detecting a pulsating component ΔA1 of the first pulsating wave signal; a second pulsating component detecting step for detecting a pulsating component ΔA2 of the second pulsating wave signal; A pulsating component calculating step for calculating a ratio Φ of the pulsating component of the pulsating signal and the pulsating component of the second pulsating signal, and a frequency spectrum or a frequency power spectrum for a predetermined period to calculate a fundamental frequency of the pulsating signal. And a fundamental frequency calculation step for calculating. Thus, the fundamental frequency of the pulse wave can be obtained by calculating the difference between the pulsating components obtained for each pulse.

A pulse wave signal processing method according to a twenty-first aspect of the present invention is a pulse wave signal processing method for processing a first pulse wave signal and a second pulse wave signal of a living body having the same fundamental frequency,
A first pulsating component detecting step for detecting a pulsating component ΔA1 of the first pulsating wave signal; a second pulsating component detecting step for detecting a pulsating component ΔA2 of the second pulsating wave signal; A pulsating component calculating step for calculating the ratio Φ of the pulsating component of the pulsating wave signal and the pulsating component of the second pulsating wave signal, and a spectrum calculating step for calculating the frequency spectrum or the frequency power spectrum a predetermined number of times in a predetermined period. A spectrum arithmetic mean calculation step of arithmetically averaging the spectrum calculated for the predetermined number of times, a fundamental frequency calculation step of computing a fundamental frequency of the pulse wave signal from the arithmetic mean, and the arithmetic mean To a fundamental frequency computing step of computing a fundamental frequency of the pulse wave signal. Accordingly, in calculating the fundamental frequency of the pulse wave by calculating the difference between the pulsating components, noise can be suppressed by calculating the arithmetic mean of the spectrum, and the basic frequency of the pulse wave can be more easily extracted.

A pulse wave signal processing method according to a twenty-second aspect of the present invention is the pulse wave signal processing method according to any one of the twentieth and twenty-first aspects, in which the basic frequency calculation step is performed. At least the first pulse wave signal or the second pulse wave signal is filtered by a filter forming step of forming a filter using the fundamental frequency and its harmonic frequency, and a filter formed by the filter forming step. And a filtering step to perform. As a result, it is possible to form a filter for forming the pulse wave by extracting the frequency components of the fundamental frequency at which the characteristics of the pulse wave signal appear and the harmonic frequency thereof.

The pulse wave signal processing method according to claim 23 of the present invention is the pulse wave signal processing method according to claim 22,
In the filter forming step, when there are a plurality of basic frequencies calculated by the basic frequency calculating step, a filter is formed by using each basic frequency and the frequency of each harmonic thereof. As a result, even when the fundamental frequency of the pulse wave signal varies, it is possible to form a filter for forming the pulse wave by extracting the frequency components of the respective fundamental frequencies and their respective harmonic frequencies.

A pulse wave signal processing method according to a twenty-fourth aspect of the present invention is the pulse wave signal processing method according to the twenty-second aspect or the twenty-third aspect, wherein in the filter forming step,
The characteristic of the filter is a rectangular shape having a width centered on the fundamental frequency and its harmonic frequency. As a result, it is possible to form a pulse wave signal by taking in frequency components around the fundamental frequency and its higher harmonic frequencies.

A pulse wave signal processing method according to a twenty-fifth aspect of the present invention is the pulse wave signal processing method according to the twenty-second aspect or the twenty-third aspect, wherein in the filter forming step,
It is characterized in that the filter is a Gaussian characteristic filter centered on the fundamental frequency and its harmonic frequencies. As a result, it is possible to form a pulse wave signal by taking in the frequency components around the fundamental frequency and its harmonic frequency in a Gaussian distribution.

The pulse wave signal processing method according to claim 26 of the present invention is the signal processing method according to any one of claims 22 to 25, further comprising the first signal in the filtering step. Or displaying the filtered one of the second signals. Thereby, the filtered pulse wave signal can be displayed.

A pulse wave signal processing method according to a twenty-seventh aspect of the present invention is the signal processing method according to any one of the twenty-second to twenty-fifth aspects, wherein the first pulse wave signal is an artery of a living body. Is a pulse wave signal obtained by red light transmitted or reflected, the second pulse wave signal is a pulse wave signal obtained by infrared light transmitted or reflected by an artery of a living body,
In the filtering step, both the first pulse wave signal and the second pulse wave signal are filtered, and further, the first pulse wave signal or the second pulse wave signal filtered in the filtering step. And an oxygen saturation calculating step of calculating the oxygen saturation by using. Thereby, the oxygen saturation can be accurately calculated using the filtered pulse wave.

A pulse wave signal processing method according to a twenty-eighth aspect of the present invention is a fundamental frequency detecting step of detecting a pulse or heartbeat frequency in a living body, and the pulse or heartbeat frequency detected by the fundamental frequency detecting step. And a filter forming step of forming a filtering frequency by using the frequency of its harmonic, and a filtering step of filtering the pulse wave signal by using the filter formed by the filter filter. With this, the fundamental frequency is detected from the pulse or heartbeat frequency, the pulse wave is filtered by a filter that uses the frequency of this fundamental frequency and its harmonics, and the characteristic frequency component of the pulse wave is extracted to form the pulse wave. can do.

A pulse wave signal processing method according to claim 29 of the present invention is the pulse wave signal processing method according to claim 28,
In the filter forming step, when there are a plurality of basic frequencies calculated by the basic frequency calculating step, a filter is formed by using each basic frequency and the frequency of each harmonic thereof. This allows
Even if the fundamental frequency of the pulse wave fluctuates, it is possible to extract the fundamental frequency components and the frequency components of their respective harmonics to form the pulse wave.

A pulse wave signal processing method according to claim 30 of the present invention is the pulse wave signal processing method according to claim 28 or claim 29, wherein in the filter forming step,
The characteristic of the filter is a rectangular shape having a width centered on the fundamental frequency and its harmonic frequency. As a result, the pulse wave signal can be extracted by taking in the frequency components around the fundamental frequency and the harmonics thereof.

The pulse wave signal processing method according to claim 31 of the present invention is the pulse wave signal processing method according to claim 28 or claim 29, wherein in the filter forming step,
It is characterized in that the filter is a Gaussian characteristic filter centered on the fundamental frequency and its harmonic frequencies. As a result, it is possible to form a pulse wave signal by taking in the frequency components around the fundamental frequency and its harmonic frequency in a Gaussian distribution.

A pulse wave signal processing method according to a thirty-second aspect of the present invention is the signal processing method according to any one of the twenty-eighth through thirty-first aspects, further comprising the pulse wave signal filtered in the filtering step. And a step of displaying. This is for displaying the pulse wave filtered by this.

A pulse wave signal processing method according to a thirty-third aspect of the present invention is the signal processing method according to any one of the twenty-eighth to thirty-first aspects, wherein the pulse wave signal passes through an artery of a living body or A first pulse wave signal obtained by reflected infrared light and a second pulse wave signal obtained by red light transmitted or reflected through an artery of a living body, wherein the first pulse wave and the first pulse wave signal are obtained in the filtering step. An oxygen saturation calculating step of filtering any of the two pulse wave signals, and further calculating an oxygen saturation using the first pulse wave and the second pulse wave filtered in the filtering step. Is characterized by. Thereby, the oxygen saturation can be accurately calculated using the filtered pulse wave.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of an oxygen saturation measuring device that performs pulse wave signal processing according to the present embodiment.

FIG. 2 is a diagram showing a flowchart of pulse wave signal processing according to the present embodiment.

FIG. 3 is a diagram showing a time chart of pulse wave signal processing according to the present embodiment.

FIG. 4A is a diagram showing a spectrum of an infrared light pulse wave;
(B) is a figure which shows the spectrum of a red light pulse wave.

5A is a diagram showing a calculation result of a spectrum for obtaining a fundamental frequency of a pulse wave, and FIG. 5B is a diagram showing a calculation result of a spectrum for obtaining a fundamental frequency of a pulse wave.

6A is a diagram showing a filtering frequency for filtering a pulse wave, FIG. 6B is a diagram showing a rectangular frequency characteristic filter for filtering a pulse wave, and FIG. 6C is a filter for filtering a pulse wave. It is a figure which shows the Gaussian characteristic filter for doing.

7A is a diagram showing a rectangular frequency characteristic filter for filtering a pulse wave, FIG. 7B is a diagram showing a rectangular frequency characteristic filter for filtering a pulse wave, and FIG.
FIG. 4 is a diagram showing a rectangular frequency characteristic filter for filtering a pulse wave. (D) is a diagram showing a Gaussian characteristic filter for filtering a pulse wave, (e) is a diagram showing a Gaussian characteristic filter for filtering a pulse wave, and (f) is a Gaussian characteristic filter for filtering a pulse wave. Showing the figure,
(G) is a figure which shows the filtering frequency for filtering a pulse wave.

FIG. 8 is a diagram showing a time chart of pulse wave signal processing according to the present embodiment.

FIG. 9 is a diagram showing a time chart of pulse wave signal processing according to the present embodiment.

FIG. 10 is a diagram showing a time chart of pulse wave signal processing according to the present embodiment.

FIG. 11 is a diagram showing a spectrum of a difference between an infrared light pulse wave and a red light pulse wave.

FIG. 12 is a diagram showing a calculation result of a spectrum for obtaining a fundamental frequency of a pulse wave.

[Explanation of symbols]

1 probe 2 Light emitting part 3 Light receiving part 4 Living tissue (fingers) 5 Light emission part drive circuit 6 Light receiving signal amplification circuit 7 Demodulation circuit 8 CPU 9a amplifier 9b amplifier 10a A / D converter 10b A / D converter 11 Display

Claims (33)

[Claims] 1. A signal processing method for processing a continuous first signal IR and second signal RD having the same fundamental frequency, wherein a predetermined value is set for each of the first signal IR and the second signal RD. A spectrum calculation step of calculating a frequency spectrum or a frequency power spectrum and calculating a first spectrum Spc.IR of the first signal IR and a second spectrum Spc.RD of the second signal RD in the period; The first calculated by the spectrum calculation step
Using the spectrum Spc.IR and the second spectrum Spc.RD, on the frequency axis, in addition to the spectrum calculation for obtaining the mutual difference or the spectrum calculation for obtaining the mutual difference, the result of further performing the spectrum calculation for normalization is obtained. Based on the first
And a fundamental frequency computing step of computing a fundamental frequency of the signal IR and the second signal RD. 2. The signal processing method according to claim 1, wherein in the fundamental frequency calculation step, (Spc.IR− Spc.RD) /Spc.RD, (Spc.IR− Spc.RD) /Spc.IR. , (Spc.IR- Spc.RD) / (Spc.IR + Spc.RD), {1- (Spc.RD / Spc.IR)} / (Spc.RD / Spc.I)
R), {1- (Spc.RD / Spc.IR)} / (Spc.IR / Spc.I
R), or {1- (Spc.RD / Spc.IR)} / {1+ (Spc.RD / Sp
c.IR)}, the signal processing method characterized in that the fundamental frequency is calculated based on the result of the spectrum calculation. 3. A signal processing method for processing a continuous first signal IR and second signal RD having the same fundamental frequency, wherein a predetermined value is set for each of the first signal IR and the second signal RD. A spectrum calculation step of calculating a frequency spectrum or a frequency power spectrum in the period and calculating a first spectrum Spc.IR from the first signal IR and a second spectrum Spc.RD from the second signal RD; The first spectrum according to the spectrum calculation step
Spc.IR and the second spectrum Spc.RD a predetermined number of times,
On the frequency axis, using a spectrum addition / averaging calculation step of addition / averaging, the first spectrum Av.Spc.RD and the second spectrum Av.Spc.RD added / averaged by the spectrum addition / averaging calculation step Then, based on the result of performing the spectrum calculation for obtaining the mutual difference or the spectrum calculation for obtaining the mutual difference and further performing the spectrum calculation for normalization, the fundamental frequencies of the first signal IR and the second signal RD. A signal processing method comprising: a fundamental frequency calculation step for calculating. 4. The signal processing method according to claim 3, wherein in the fundamental frequency calculation step, (Av.Spc.IR − Av.Spc.RD) /Av.Spc.RD, (Av.Spc.IR). − Av.Spc.RD) / Av.Spc.IR, (Av.Spc.IR − Av.Spc.RD) / (Av.Spc.IR + Av.Spc.
RD), {1- (Av.Spc.RD / Av.Spc.IR)} / (Av.Spc.RD /Av.S
pc.IR), {1- (Av.Spc.RD /Av.Spc.IR)} / (Av.Spc.IR / A
v.Spc.IR), or {1- (Av.Spc.RD / Av.Spc.IR)} / {1+ (Av.Spc.RD /
Av.Spc.IR)}, the signal processing method characterized in that the fundamental frequency is calculated based on the result of the spectrum calculation. 5. Any one of claims 1 to 4
In the signal processing method according to item 4, further, a filter forming step of forming a filter using the fundamental frequency calculated by the fundamental frequency calculating step and a frequency of its harmonic, and a filter formed by the filter forming step. And a filtering step of filtering at least the first signal IR or the second signal RD. 6. The signal processing method according to claim 5, wherein in the filter forming step, when there are a plurality of the fundamental frequencies calculated by the fundamental frequency calculating step, each fundamental frequency and its respective harmonics. A signal processing method characterized in that a filter is formed by using the frequencies of. 7. The signal processing method according to claim 5, wherein in the filter forming step, the characteristic of the filter is a quadrature having a width centered on the fundamental frequency and its harmonic frequency. A signal processing method having a shape. 8. The signal processing method according to claim 5, wherein in the filter forming step, the filter is
A signal processing method, which is a Gaussian characteristic filter centered on the fundamental frequency and its harmonic frequency, respectively. 9. Any one of claims 1 to 8.
5. The signal processing method according to claim 6, wherein both the first signal IR and the second signal RD are pulse wave signals. 10. The signal processing method according to claim 5, wherein each of the first signal IR and the second signal RD is a pulse wave signal, In the filtering step, the first
Signal IR or the second signal RD, whichever is filtered, is displayed. 11. The signal processing method according to claim 5, wherein the first signal IR is a pulse wave obtained by infrared light transmitted through or reflected by an artery of a living body. The second signal RD is a pulse wave signal obtained by red light transmitted or reflected through an artery of a living body, and the first signal IR in the filtering step.
And an oxygen saturation calculation step of filtering any of the second signal RD, and further calculating an oxygen saturation using the first signal IR or the second signal RD filtered in the filtering step. And a signal processing method comprising: 12. A signal detecting step of detecting a continuous signal having periodicity, a basic frequency calculating step of calculating a basic frequency of the signal detected by the signal detecting step, and a basic frequency calculating step of calculating the basic frequency. A filter forming step of forming a filtering frequency by using a fundamental frequency of the signal and its harmonic frequency, and a filtering step of filtering the signal by using the filter formed by the filter filter. Signal processing method. 13. A signal processing method for processing a continuous first signal and a second signal having the same fundamental frequency, the method comprising: detecting the first signal and the second signal; Of the fundamental frequency of the signal using at least one of the signal and the second signal, and the fundamental frequency calculated by the fundamental frequency calculating step and the frequency of its harmonics. A signal processing method comprising: a filter forming step of forming a filter. 14. The signal processing method according to claim 13, wherein in the filter forming step, when there are a plurality of the fundamental frequencies calculated by the fundamental frequency calculating step, each fundamental frequency and its respective harmonics. A signal processing method characterized in that a filter is formed by using the frequencies of. 15. The signal processing method according to claim 13, wherein in the filter forming step, the characteristic of the filter is a quadrature having a width centered on the fundamental frequency and its harmonic frequency. A signal processing method having a shape. 16. The signal processing method according to claim 13 or 14, wherein in the filter forming step, the filter is
A signal processing method, which is a Gaussian characteristic filter centered on the fundamental frequency and its harmonic frequency, respectively. 17. The signal processing method according to claim 12, wherein each of the first signal and the second signal is a pulse wave signal. Signal processing method. 18. The signal processing method according to claim 12, wherein each of the first signal and the second signal is a pulse wave signal, and the filtering is performed. In step 1
And a signal of the filtered one of the second signal and the second signal. 19. The signal processing method according to claim 12, wherein the first signal is a pulse wave signal obtained by infrared light transmitted through or reflected by an artery of a living body. And the second signal is a pulse wave signal obtained by red light transmitted through or reflected by an artery of a living body, and filtering both the first signal and the second signal in the filtering step. Further, a signal processing method comprising: an oxygen saturation calculating step of calculating oxygen saturation using the first signal or the second signal filtered in the filtering step. 20. A pulse wave signal processing method for processing a first pulse wave signal and a second pulse wave signal of a living body having the same fundamental frequency, wherein a pulse component ΔA1 of the first pulse wave signal is detected. 1 pulsation component detection step, a second pulsation component detection step of detecting the pulsation component ΔA2 of the second pulse wave signal, a pulsation component of the first pulse wave signal and the second pulse wave signal A pulsation component calculation step of calculating a difference between pulsation components, and a basic frequency calculation step of calculating a frequency spectrum or a frequency power spectrum in a predetermined period to calculate a basic frequency of the pulse wave signal. Pulse wave signal processing method. 21. A pulse wave signal processing method for processing a first pulse wave signal and a second pulse wave signal of a living body having the same fundamental frequency, wherein a pulsating component ΔA1 of the first pulse wave signal is detected. 1 pulsation component detection step, a second pulsation component detection step of detecting the pulsation component ΔA2 of the second pulse wave signal, a pulsation component of the first pulse wave signal and the second pulse wave signal A pulsation component calculation step for calculating the difference between pulsation components, a spectrum calculation step for calculating a frequency spectrum or a frequency power spectrum a predetermined number of times in a predetermined period, and a spectrum addition for adding and averaging the spectra calculated for the predetermined number of times. An average calculation step; and a basic frequency calculation step of calculating a basic frequency of the pulse wave signal from the added and averaged spectrum. Pulse wave signal processing method. 22. The pulse wave signal processing method according to claim 20, wherein a filter is provided using the fundamental frequency calculated by the fundamental frequency calculating step and the harmonic frequency thereof. And a filtering step of filtering at least the first pulse wave signal or the second pulse wave signal with the filter formed by the filter forming step. Wave signal processing method. 23. The pulse wave signal processing method according to claim 22, wherein, in the filter forming step, when there are a plurality of the basic frequencies calculated by the basic frequency calculating step, each basic frequency and each of the basic frequencies are calculated. A pulse wave signal processing method, characterized in that a filter is formed by using frequencies of harmonics. 24. The pulse wave signal processing method according to claim 22 or 23, wherein in the filter forming step, the characteristic of the filter has a width centered on the fundamental frequency and its harmonic frequency, respectively. A pulse wave signal processing method having a rectangular shape. 25. The pulse wave signal processing method according to claim 22 or 23, wherein in the filter forming step, the filter is
A pulse wave signal processing method, which is a Gaussian characteristic filter centered on the fundamental frequency and its harmonic frequency, respectively. 26. The signal processing method according to claim 22, further comprising the first step in the filtering step.
And a signal of the filtered one of the second signal and the second signal. 27. The signal processing method according to any one of claims 22 to 25, wherein the first pulse wave signal is a pulse wave obtained by red light transmitted or reflected through an artery of a living body. Signal, the second pulse wave signal is a pulse wave signal obtained by infrared light transmitted through or reflected by an artery of a living body, and the first pulse wave signal and the second pulse wave signal in the filtering step. An oxygen saturation calculation step of calculating oxygen saturation using either the first pulse wave signal or the second pulse wave signal filtered in the filtering step. A signal processing method comprising: 28. A fundamental frequency detecting step of detecting a pulse or heartbeat frequency in a living body, and a filtering frequency is formed by using the pulse or heartbeat frequency detected by the fundamental frequency detecting step and a harmonic frequency thereof. And a filtering step of filtering the pulse wave signal using the filter formed by the filter filter. 29. The pulse wave signal processing method according to claim 28, wherein in the filter forming step, when there are a plurality of basic frequencies calculated by the basic frequency calculating step, each basic frequency and its respective A pulse wave signal processing method, characterized in that a filter is formed by using frequencies of harmonics. 30. The pulse wave signal processing method according to claim 28 or claim 29, wherein in the filter forming step, the characteristics of the filter have a width centered on the fundamental frequency and its harmonic frequency, respectively. A pulse wave signal processing method having a rectangular shape. 31. The pulse wave signal processing method according to claim 28 or 29, wherein in the filter forming step, the filter is
A pulse wave signal processing method, which is a Gaussian characteristic filter centered on the fundamental frequency and its harmonic frequency, respectively. 32. The signal processing method according to claim 28, further comprising a step of displaying the pulse wave signal filtered in the filtering step. Pulse wave signal processing method. 33. The signal processing method according to claim 28, wherein the pulse wave signal is a first pulse wave obtained by infrared light transmitted through or reflected by an artery of a living body. A signal and a second pulse wave signal obtained by red light transmitted or reflected through an artery of a living body, wherein both the first pulse wave signal and the second pulse wave signal are filtered in the filtering step, and An oxygen saturation calculation step of calculating an oxygen saturation using the first pulse wave and the second pulse wave signal filtered in the filtering step, and a pulse wave signal processing method.