JP2851088B2 - Biological signal pulse detector - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for detecting a pulse position of a heartbeat synchronization signal from a biological signal including a heartbeat synchronization signal derived from a heartbeat. is there.

2. Description of the Related Art For example, a pressure pulse wave collected from an artery of a living body, an impedance pulse wave collected based on a change in impedance of a part of a living body, a photoelectric pulse wave or a volume pulse wave collected photoelectrically from a capillary blood vessel. Such a heartbeat synchronous wave signal derived from a heartbeat is directly used as biological information, or is used to calculate a pulse rate, a blood pressure value, a degree of oxygen saturation in blood, and the like based on the information. In this case, the biological signal collected from the living body includes not only a heartbeat synchronous wave signal, but also a respiratory synchronous wave derived from respiration and noise caused by body movement of the living body. Value, to accurately calculate the oxygen saturation in the blood, or to determine the cutoff frequency of the digital filter processing for extracting the original heartbeat synchronization wave signal wave from the biological signal, There is a case where it is desired to accurately detect the position of the pulse.

Problems to be Solved by the Invention By the way, not only a heartbeat synchronizing wave signal but also a respiratory synchronizing wave signal and other noises are generally mixed into a biological signal, which affects the type or physical state of a living body. If the heartbeat synchronization wave signal component is weaker than the other components, the pulse position of the heartbeat synchronization wave cannot be accurately detected from the biological signal, and the pulse rate, blood pressure value, or blood In some cases, it was extremely difficult to measure the oxygen saturation or the like. For example, the neonatal arteries are extremely thin and the subcutaneous tissue is thick, so the signal collected from the neonate has a low S / N ratio of the regular heartbeat synchronizing signal. There was a case that could not be done.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a biological signal capable of reliably detecting a pulse position of a heartbeat synchronous wave signal from a biological signal obtained from a living body. A pulse detection device is provided.

Means for Solving the Problems To achieve the above object, the gist of the present invention is to detect a pulse position of a heartbeat synchronizing wave signal included in a biological signal as shown in the claim correspondence diagram of FIG. (A) an added value of a continuous group of signals among the original signals of the heartbeat synchronous wave signal continuously obtained with the passage of time, and the group of signals; Means for successively obtaining a difference from an added value of another group of signals successively shifted by a predetermined time to perform smooth differentiation on the original signal; and (b) obtaining a signal obtained by performing the smooth differentiation on the original signal. When a change state in a predetermined fixed direction within a first movement section having a predetermined width is determined, a difference between a maximum value and a minimum value in the movement section is sequentially determined, and a change state in a certain direction within the first movement section. When not A first moving section processing for generating a first processing signal by setting it to zero; and (c) when the first processing signal shows an extreme value in the center of a second moving section having a predetermined width, the extreme value is represented as A second moving section process for generating a second processing signal consisting of a predetermined pulse train by setting the magnitude of the first processing signal in the indicated portion to be zero when the extremum is not shown at the center in the second moving section; (D) pulse detection means for determining a pulse generation position of a heartbeat synchronization wave based on regularly generated pulses among the pulses constituting the second processing signal.

In this way, when the biological signal including the heartbeat synchronization wave signal is smoothed and differentiated by the smoothing / differentiating means, the original signal of the heartbeat synchronization wave signal continuously obtained over time is obtained. The difference between the added value of a continuous group of signals and the added value of the other group of signals that are shifted from each other by a predetermined time is sequentially obtained, so that the slope of the biological signal is expressed. Thus, the slow component becomes inconspicuous and the component lower than the frequency of the heartbeat synchronous wave is removed. Next, the first processing signal generated by the first moving section processing means is a difference value in the first moving section when the slope of the signal after smooth differentiation is uniform in the first moving section. , The slope of the portion corresponding to the convex or concave section of the biological signal of the signal after the smooth differentiation.
Subsequently, the second processing signal generated by the second moving section processing means has the magnitude of the first processing signal when it exhibits an extreme value at the center of the first processing signal in the second moving section. At this time, a pulse shape having the same height as that of the first processing signal is obtained, and the pulse shows an upper peak or a lower peak of a biological signal. It corresponds to the upper or lower peak of the sexual synchronization wave. Therefore, the pulse detecting means can accurately determine the generation position of the heartbeat synchronization wave included in the biological signal based on the regularly generated pulse. When the pulse (peak) position of the heartbeat synchronization signal is accurately detected in this manner, the period, frequency or pulse rate of the heartbeat synchronization signal can be accurately calculated based on the pulse position. The blood pressure value can be calculated from the relationship stored in advance based on the magnitude of the heartbeat synchronization signal corresponding to the pulse position, and based on the amplitude of the heartbeat synchronization signal corresponding to the pulse position. The oxygen saturation in the blood can be accurately calculated from the stored relationship.
Alternatively, the cutoff frequency of the digital filter processing for extracting the original heartbeat synchronous wave signal wave from the biological signal is determined based on the frequency of the heartbeat synchronous wave signal. The heartbeat synchronous wave signal which is the basis of the measurement of the blood pressure value and the oxygen saturation in blood is preferably used after the above digital filter processing, and the heartbeat synchronous wave is weak and difficult to measure with a conventional measuring device. In this case, measurement can be performed, and relatively high measurement accuracy can be obtained.

Embodiment Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

In FIGS. 1 and 2, reference numeral 10 denotes a probe, which is mounted on a body surface 12 such as a finger having a relatively high density of peripheral blood vessels of a living body by a mounting means (not shown) with a predetermined pressing force. Is done. The probe 10 has a bottomed cylindrical member 14 having a relatively shallow depth, a light receiving element 16 formed of a photodiode, a phototransistor, or the like provided at the center of the bottom inner surface of the bottomed cylindrical member 14, and a bottomed cylindrical member. The first light-emitting element 18 and the second light-emitting element 20 each comprising, for example, eight LEDs or the like are provided alternately at predetermined intervals on a circle having the same radius around the light-receiving element 16 on the bottom inner surface of the shape member 14.
And a light receiving element integrally provided in the bottomed cylindrical member 14.
16 and a transparent resin 22 covering the light emitting elements 18 and 20, provided between the light receiving element 16 and the light emitting elements 18 and 20 in the bottomed cylindrical member 14, the light emitted from the light emitting elements 18 and 20 A cylindrical light-shielding member 24 that shields reflected light from the body surface 12 toward the light receiving element 16, and a main body 26 that is provided so as to cover the outer peripheral surface and the bottom outer surface of the bottomed cylindrical member 14 are configured. I have.

The first light emitting element 18 emits red light having a wavelength of, for example, about 660 mμ, and the second light emitting element 20 emits infrared light having a wavelength of, for example, about 820 mμ, but is not necessarily limited to these wavelengths. Instead, it is only necessary to emit light having a wavelength at which the absorption coefficient of hemoglobin and the absorption coefficient of oxyhemoglobin are significantly different, and light having a wavelength at which both the absorption coefficients are substantially the same. The first light emitting element 18 and the second light emitting element 20 emit light at a predetermined frequency in order for a certain period of time, and the reflected light from the vascular bed in the body surface 12 of the light emitted from both light emitting elements 18 and 20 is Common light receiving element
16 respectively.

The light receiving element 16 outputs an electric signal SV having a magnitude corresponding to the amount of received light to the low-pass filter 32 via the amplifier 30. This electric signal SV corresponds to the photoelectric pulse wave signal of the present embodiment, and includes a fluctuation component due to arterial pulsation. The low-pass filter 32 removes noise having a frequency higher than the frequency of the pulse wave from the input electric signal SV, and demultiplexes the noise-removed signal SV.
Output to 34. The demultiplexer 34 includes a switching signal SC described later.
Is switched in synchronism with the light emission of the first light emitting element 18 and the second light emitting element 20, whereby the red reflected light signal SV _R
Are sequentially supplied to the I / O port 40 via the A / D converter 38, and the infrared reflected light signal SV _IR is sequentially supplied to the I / O port 40 via the A / D converter 44.

The I / O port 40 is connected to the CPU 46 and ROM via the data bus line.
48, RAM 50 and display 52 are connected respectively. CPU46
Performs an oxygen saturation measurement operation by processing an input signal according to a program predetermined in the ROM 48 while utilizing the storage function of the RAM 50. That is, I / O port 40
Outputs the irradiation signal SLD to the drive circuit 54 to output the first light emitting element 1
8 and the second light emitting element 20 are sequentially emitted at a predetermined frequency for a predetermined time, and the switching signal SC is output in synchronization with the light emission of the first light emitting element 18 and the second light emitting element 20 to cause the demultiplexer 34 to emit light. By switching, the red reflected light signal SVR is _sent to the A / D converter 38, and the infrared reflected light signal _{SVIR is}
To the A / D converter 44. Also, the CPU 46
By processing the input signal according to a pre-stored program, the oxygen saturation in the blood flowing through the peripheral blood vessels is determined, and the determined oxygen saturation is displayed on the display 52.

Next, the operation of measuring the oxygen saturation of the pulse oximeter configured as described above will be described with reference to the flowchart of FIG.

First, by executing step S1 in FIG. 3,
The irradiation of the red light from the first light emitting element 18 and the infrared light from the second light emitting element 20 in order at a predetermined frequency is started. The predetermined frequency is the set frequency to obtain sufficiently high resolution of the formed red reflected optical signal SV _R and the infrared light reflected signals SV _IR by a series of data points. Then, the red reflected optical signal SV _R and the infrared light reflected signals SV _IR receiving element 1 shows the intensity of the reflected light from the vascular beds
Since the sequentially detected by 6, their signals SV _R and SV _IR sequentially read in step S2.

In step S3, the pulse position of the red reflected optical signal SV _R and the infrared light reflected signals SV _IR, i.e. under the peak position in the present embodiment is detected according to a subroutine shown in Figure 4. In step S1 of FIG. 4, in order to remove the relatively low frequency components, such as respiratory synchronization wave, to read red reflected optical signal SV _R and the infrared light reflected signals SV _IR, the following equation (1) Is applied. That is, the added value of a continuous group of signals of the original signal D (i) of the heartbeat synchronous wave signal continuously obtained with the passage of time and another continuous group shifted by a predetermined time from the group of signals. Of the original signal D
(I) is subjected to smooth differentiation, and the signal S (n) after the smooth differentiation is obtained.
Is required. In two-stage on the fifth diagram, heart of the red reflected optical signal taken from newborns can hardly be grasped synchronized wave visually SV _R and the infrared light reflected signals SV _IR
Are shown in the lower two stages of FIG.
_The signal SV _R ′ obtained by subjecting _R and SV _IR to the above-described smoothing differentiation process
And SV _IR ′ are shown, respectively.

Here, D (i) indicates the original signal, and S (n) indicates the signal after smooth differentiation.

In the following step SS2, the first moving section processing is performed on the waveform S (n) that has been subjected to the smooth differential processing as described above. That is, for the waveform S (n), a certain first movement section i
= I = (n−W, n), which gives the minimum value within (n−W, n), that is, the minimum value within the first moving section i = (n−W1, n) on the time axis going to the right, for example, The function Y (n) having the value shown in the following expression (2) when located at the right end, but having the value shown in the following expression (3) is generated in other cases.

Y (n) = MAX _{n−w1, n} S (n) −MIN _{n−w1, n} S
(N) (2) where MAX _{n−w1, n} S (n) is the first moving section i = (n
−W1, n) and MIN _{n−w1, n} S (n) represent the minimum value in the first moving section i = (n−W1, n), respectively.

Y (n) = 0 (3) In the first moving section processing, the first moving section i = (n
−W1, n) is set to the value shown in the above equation (2) when i which gives the minimum value in n is equal to n. Processing is performed, and the function Y (n) has a shape in which a waveform having a peak temporally corresponding to the lower peak of the original waveform D (i) is discontinuously connected. In this embodiment, the time width W1 of the first moving section is set to about 100 _msec , so that the heartbeat synchronous wave appears remarkably. Note that a sharp change (high-frequency component) becomes remarkable as the time width W1 decreases, and a gradual change (low-frequency component) becomes prominent as the time width W1 increases.

In the subsequent second moving section processing in step SS3, the first
The function Y (n) in the fixed second moving section i = (n−W2, n + W2) is performed on the function Y (n) on which the moving section processing has been performed.
Is the same as n, that is, for example, on the rightward time axis, the second moving section i =
When the maximum value in (n−W2, n + W2) is located at the center, the value becomes the value shown in the following equation (4). Otherwise, the function G (n) having the value shown in the following equation (5) is generated. Let it.

G (n) = Y (n) (4) G (n) = 0 (5) In the second moving section processing, the peak value of the waveform of the function Y (n) is changed to the second moving section. Only when it is located at the center of i = (n−W2, n + W2), the function Y (n) is set as the peak value. Therefore, the function G (n) has the lower peak of the current waveform D (i). At a position corresponding to time, it is composed of a pulse train of the same size as this position Y (n). Pulse train other than a continuous waveform shown in the lower two stages of FIG. 5, the smoothing differential processing to the above two stages of the red reflected optical signal SV _R and the infrared light reflected signals SV _IR, first movement section process, and This is obtained by performing each of the second moving section processes. In the present embodiment, half W2 of the time width of the second moving section is about 10
This is set to 0 _msec , so that a peak occurring at a close position that is not physiologically possible as a heartbeat synchronization wave is removed. The smaller the W2 is, the more the peak located closer is removed.

Next, the positions of the respective pulse trains constituting the function G (n) are set as pulse position candidates, and in step SS4 and subsequent steps, those pulse position candidates are selected according to predetermined conditions. In addition,
Let T (k) be the time of occurrence of the pulse portion where the value of the function G (n) is not zero.

First, in step SS4, it is already determined whether the time interval [T (k) -T (k-1)] of the successively obtained pulse generation time T (k) is normal or irregular. Average pulse time tb obtained from the pulse PRD of the living body
Is determined based on whether it is within 25%. In this step SS4, the time interval of the above-mentioned pulse [T (k)
-T (k-1)] is determined not to be within 25% of the average interpulse time tb, in step SS5, the criterion value th for determining the pulse size is set to the previous value. This routine is terminated after being re-determined to half the value (th / 2). However, if it is determined in step SS4 that the time interval [T (k) -T (k-1)] of the pulse is within 25% of the average interpulse time tb, the function is determined in step SS6. Whether or not the magnitude G (k) of each pulse of G (n) is sufficiently large is determined based on the criterion value th. This criterion value th is determined in the previous cycle in step SS7 described below. In step SS7, the criterion value th is determined by the following equation (6) for the pulse G (k) satisfying the determinations in steps SS4 and SS6.

 th = G (k) / s (6) where s is determined according to the following equation (7).

 s = 2.5 × fs / [T (k) −T (k−1)] (7) where fs is a sampling frequency.

In the following step SS8, the pulse interval [T (k) -T
(K-1)] is Tmax, and the minimum value is Tmin, the difference ΔT (Tmax−Tmin) is the average time interval Tave [= T (i) −T (i− 1) 1/4]
It is determined whether or not: That is, the predetermined section Μ
The maximum variation of the pulse within the average time interval Tave is 25%
It is determined whether it is within the range. Note that, for the predetermined section Μ, the larger value of [T (k) -T (k-5)] or 3 seconds is adopted.

If the determination in step SS8 is affirmative, in step SS9, the average time interval T
The pulse generation position obtained from ave is set as a pulse candidate PRb. However, if the determination in step SS8 is denied, in step SS10, the pulse candidate PRb
Search for That is, the time width Tu (= T (i) −T (i−
j): Out, for Tu ≦ 2 seconds, j ≧ 1), the following equation (8)
T that satisfies the condition of [1] and is closest to [T (i) -Tu].
(M) is searched, and if T (m) is found, a new Tb is obtained according to the following equation (9), and the operation of searching for the next (m) using T (m) as a new T (i) is performed in the section [T (i) -T
(M): However, within 4 seconds].

T (i) −Tu−L ≦ T (m) ≦ T (i) −Tu + L (8) where L = 0.25 Tu and L ≦ 0.8 seconds Tb = T (i) −T (m) / N (9) where N is the number of searches of T (m), and the last found T (m) is T (Μ), G
Let (m) be G (Μ) and Tb be TB.

T (m) is not found at all,
(I) If −T (m) ≦ 2 seconds, 1 is added to j and the search for T (m) is executed again.

Then, T (m) is found and T (i) −T
If (m) ≧ 2 seconds, the pulse is calculated with the interpulse time as TB, and the pulse is calculated as a pulse candidate PRb.

As described above, when the pulse candidate PRb is determined in step SS9 or SS10, in step SS11,
It is determined whether or not the pulse candidate PRb satisfies physiological conditions of a living body, for example, the following three conditions (1), (2), and (3).

(1) 30 pm≦PRb≦240bpm where bpm is the number of beats per minute (2) 0.8 × last-p ≦ PRb ≦ 1.2 × last-p where last-p is the previous pulse (time of last pulse (3) T (i) -last-ct ≧ 0.8 × interval interval of determined pulse PRD, where last-ct is the previous pulse generation time (immediate pulse generation time) and the above three conditions If all the conditions are satisfied, the generation time T (i) of the current pulse is updated as last-ct, the current pulse candidate PRb is updated as last-p, and the current pulse candidate PRb is updated as last-p. After updating half of the size as the judgment reference value hit, the function G (i) in this cycle is recognized as the occurrence of a pulse (lower peak). If only the above conditions (1) and (2) are satisfied, the current pulse candidate PR
After updating b as last-p and updating half of the magnitude of the current G (の) as the judgment reference value hit, the function G (i) in the current cycle is recognized as the occurrence of a pulse (lower peak). .
If only the condition (3) is satisfied, the pulse candidate PR
When b satisfies one of the following equations (10) and (11), the pulse candidate PRb is updated as last-p, and the function G (i) at the time of the current cycle is determined from the judgment reference value hit. If it is larger, the generation time T (i) of the current pulse is
After updating as last-ct, the function G in this cycle
(I) is recognized as the occurrence of a pulse (lower peak). Also, (1)
If the condition (2) is not satisfied, the time T (i) of the current pulse is updated as last-ct, and the function G (i) of the current cycle is pulsed. (Lower peak).

PRb <0.4 × last-p (10) PRb> 0.6 × last-p (11) When the occurrence of a pulse is recognized as described above, the pulse is determined in step S4 in FIG. with the number of pulses from the occurrence time is calculated, the cutoff frequency of the digital filter process is determined, and facilities digital filtering of the cut-off frequency in the red reflected optical signal SV _R and the infrared light reflected signals SV _IR Is done. Then, based on the red reflected light signal V _R and the infrared reflected light signal SV _IR that have been subjected to the digital filter processing in this manner, the red reflected light signal SV
The upper peak value V _dR (corresponding to the reflected light intensity during diastole) and the lower peak value V _sR (corresponding to the reflected light intensity during systole) of one pulse waveform represented by _R are determined, and infrared reflection is performed. peak value V _dir and below the peak value V _SIR on one pulse waveform represented by the optical signal SV _IR is determined. FIG. 6 is a diagram showing an example of a pulse waveform of the intensity of the red reflected light and a pulse waveform of the intensity of the infrared reflected light, and both pulse waveforms are indicated by the same pulse waveform for convenience.

Next, by executing step S5, the step
Based on the peak value determined in S4, V _dR −V _sR , V _dR +
V _sR , V _dIR −V _sIR , V _dIR + V _sIR are calculated respectively, and the following ratios (12) and (13) are calculated respectively. By taking such a ratio, the light emitting elements 12, 20
The influence on the measurement due to the difference in the light emission intensity of the light-emitting element, the characteristics of the light-emitting element 16, the light absorption characteristics of the skin pigment, and the scattering and absorption of light in the vascular bed due to the wavelength of light is avoided. In the following step S6, the following ratio (14) is calculated.

(SV _dR −V _sR ) / (V _dR + V _sR ) (12) (V _dIR −V _sIR ) / (V _dIR + V _sIR ) (13) Next, step S7 is executed, and the ratio (14) actually calculated in step S6 from a predetermined and stored relationship between the value shown in the above equation (14) and the oxygen saturation of the artery. Is used to determine the actual oxygen saturation. Next steps
In S8, the oxygen saturation determined in step S7 is displayed on the display 52, and the oxygen saturation is continuously determined and displayed by repeatedly executing this routine.

As described above, according to this embodiment, when the red reflected optical signal SV _R and the infrared light reflected signals SV _IR is a biological signal is smoothed differentiated by the corresponding smooth differential means in step SS1, the formula (1 ), The added value of a continuous group of signals in the original signal D (i) of the heartbeat synchronous signal continuously obtained with the passage of time and the group of signals are successively shifted by a predetermined time. the difference between the sum of the other group of signals are obtained sequentially becomes the inclination of the signal SV _R and SV _IR is expressed, less components than the frequency of the heartbeat of synchronous wave becomes inconspicuous gentle ingredients Removed. Next, the first processed signal Y (n) generated by the corresponding first moving section processing means in step SS2 is obtained when the signal S (n) after the smooth differentiation has a uniform gradient in the first moving section. since the first is a difference in the movement zone of the inclination of the portion corresponding to the red reflected optical signal SV _R and the infrared light reflected signals SV section of downward convex shape of the _IR signal after the smoothing differential S (n) Will be represented. Subsequently, when the second processing signal G (n) generated by the corresponding second moving section processing means in step SS3 indicates an extreme value at the center of the first processing signal Y (n) in the second moving section. First processing signal Y
Since it is the size of the (n), but are intended to be of a pulse shape of the first processed signal Y (n) and the same height at that time, the pulse is reflected red light signal SV _R and the infrared shows the lower peak of the reflected optical signal SV _IR, periodic ones of these pulses corresponds to the upper peaks or the lower peaks of the heartbeat of synchronous wave. Therefore, step SS
By 4 to SS11 of the corresponding pulse detector, accurate generation position of the pulse of the heart of synchronous waves contained in the red reflected optical signal SV _R and the infrared light reflected signals SV _IR based on the pulse occurring regularly It can be determined to When the pulse (peak) position of the heartbeat synchronization signal is accurately detected in this manner, the period, frequency or pulse rate of the heartbeat synchronization signal can be accurately calculated based on the pulse position. The cutoff frequency of the digital filter processing for extracting the original heartbeat synchronous wave signal from the biological signal is determined based on the frequency of the heartbeat synchronous wave signal. Since the heartbeat synchronous wave signal used as the basis for the measurement of the oxygen saturation in blood is subjected to the above digital filter processing, the heartbeat synchronous wave contained in the biological signal is weak and cannot be measured with a conventional measurement device. Even when it is difficult, measurement can be performed, and relatively high measurement accuracy can be obtained.

As mentioned above, although one Example of this invention was described based on drawing, this invention is applied to another aspect.

For example, in the foregoing embodiments, by sequential pulse detection routine is executed for read red reflected optical signal SV _R and the infrared light reflected signals SV _IR, but the pulse detected at real time it has been performed , The red reflected light signal SV _R for a certain period in advance
And infrared reflected light signal SV _IR is loaded, than it is also a batch process for performing pulse detection on this way read red reflected optical signal SV _R and the infrared light reflected signals SV _IR .

In the illustrated embodiment, the pulse detection performed by pulse detection routine of FIG. 4 with respect to read red reflected optical signal SV _R and the infrared light reflected signals SV _IR is performed, whereby blood Although the oxygen saturation inside can be calculated accurately, the pulse detection routine of FIG. 4 is executed for a pressure pulse wave signal (biological signal) obtained from a pressure sensor pressed on an artery. In this way, the blood pressure value may be calculated from a relationship stored in advance based on the magnitude of the heartbeat synchronizing wave signal corresponding to the pulse position.

Further, in the illustrated embodiment, had been detected by the lower peak of the red reflected optical signal SV _R and the infrared light reflected signals SV _IR pulse detection routine of FIG. 4 is executed to detect the upper peak You may do so. Step SS2 in this case
Then, i that gives the maximum value in the section i = (n-W1, n) is n
, Y (n) is set to the value of equation (2).

Further, in the above-described embodiment, the pulse detection process in the pulse oximeter has been described. However, for example, when determining the blood pressure value from the pressure pulse wave non-invasively collected from the radial artery, the pulse of the present invention may be used. Detection is applied.

In the above-described embodiment, the first light emitting element 18 and the second
A plurality of light-emitting elements 20 are provided, but are not necessarily required. For example, one light-emitting element may be provided.

In the above-described embodiment, the case where the present invention is applied to the probe of the reflection type pulse oximeter has been described. However, the present invention can be applied to the probe used for the transmission type pulse oximeter.

In addition, the present invention can be variously modified without departing from the spirit thereof.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a pulse oximeter to which the present invention is applied. FIG. 2 is a view of the probe in FIG. FIG. 3 is a flowchart for explaining the operation of the pulse oximeter of FIG. FIG. 4 is a flowchart showing a pulse detection routine of FIG. FIG. 5 is a diagram for explaining waveforms processed by the operation of FIG. FIG. 6 is a view for explaining the peak value determined in step S4 of FIG. FIG. 7 is a diagram corresponding to the claims of the present invention. SV _R : Red reflected light signal (biological signal) SV _IR : Infrared reflected light signal (biological signal) Step SS1: Smoothing differentiating means Step SS2: First moving section processing means Step SS3: Second moving section processing means Steps SS4 to SS4 SS11: pulse detection means

Claims (1)

(57) [Claims] 1. A biological signal pulse detecting device for detecting a pulse position of a heart rate synchronizing wave signal included in a biological signal, comprising: The smoothed differentiation is performed on the original signal by sequentially calculating the difference between the added value of a continuous group of signals among the signals and the added value of another continuous group of signals that are shifted from each other by a predetermined time. A smoothing differentiating means for sequentially determining a difference between a maximum value and a minimum value in the moving section when the signal subjected to the smoothing differentiation is in a predetermined fixed direction changing state in a predetermined first moving section; A first moving section process for generating a first processing signal by setting the value to zero when the state is not a change state in a fixed direction in the first moving section; In the center When the value indicates a value, the magnitude of the first processing signal in a portion indicating the extreme value is set, and when the value does not indicate an extreme value in the center of the second movement section, the value is set to zero, so that the second processing signal including a predetermined pulse train is obtained. And pulse detection means for determining a pulse generation position of a heartbeat synchronization wave based on regularly generated pulses among the pulses constituting the second processing signal. A biological signal pulse detection device characterized by including: