JP2008188216A - Biological information measurement apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a biological information measurement apparatus which determines a commercial frequency by carrying out no sampling at a high frequency and carries out such determinations even during measurement operation. <P>SOLUTION: An analog signal including a two-wave photoelectric pulse wave signal is converted into a digital signal at an A/D conversion section 25. A control processing section 26 has a biological information computing section 263 for calculating blood oxygen saturation on the basis of the two-wave photoelectric pulse wave signal. The control processing section 26 further has a noise extracting section 261 for extracting noise signals that are not derived from a living body from the photoelectric pulse wave signals, and a determination section 266 that determines whether commercial frequency of a measurement environment is 50 Hz or 60 Hz by detecting aliasing noises, which are periodical noise components, from the noise signals. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a biological information measuring apparatus that non-invasively detects various types of biological information from a human body, and more particularly to a biological information measuring apparatus having a commercial frequency detection function used in an environment where the biological information measuring apparatus is used.

  As a biological information measuring device that non-invasively detects biological information from a human body, a photoelectric pulse wave meter and a pulse oximeter for measuring the circulatory state of the living body are known. These measuring devices attach a probe including a light emitting unit and a light receiving unit to a subject's finger, project light emitted from the light emitting unit toward a living body (finger) at predetermined intervals, and pass through the living body. Then, a photoelectric pulse wave waveform is obtained by measuring a change in the amount of light detected by the light receiving unit as a pulse signal. Based on such a photoelectric pulse waveform, blood oxygen saturation and arrhythmia (atrial fibrillation / extra systole) can be detected, defibrillation monitor, autonomic nerve disorder, blood vessel age, and the like can be diagnosed.

  By the way, various noise signals not derived from a living body are superimposed on the light receiving signal output from the light receiving unit. The main thing is the induction noise of the fluorescent lamp which exists in use environments, such as a pulse oximeter, and the noise of the AC power supply frequency (it is called "commercial frequency" in this specification) of the commercial power supply system of 50 Hz or 60 Hz. . Such noise becomes an error factor in the calculation of biological information, and thus needs to be removed from the data signal.

  Conventionally, in order to remove this noise, an analog filter is provided on the circuit, or after determining whether the commercial commercial frequency is 50 Hz or 60 Hz, the filter characteristic of the digital filter is determined and filtering is performed. (See, for example, Patent Document 1). When determining the commercial frequency, the measurement data is generally sampled at a sampling frequency higher than 50 Hz or 60 Hz. In this case, when the sampling frequency is assumed to be a commercial frequency of 60 Hz, the discrimination is possible even at about 120 Hz.

  However, in order to accurately determine the commercial frequency, it is desirable to set the sampling frequency to about 10 times (600 Hz) or more. If the sampling frequency is about this level, as shown in FIG. 20, 10 sampling points can be obtained in one cycle in the case of 60 Hz. Therefore, the “60 Hz sine wave” is accurately identified from the obtained sampling data. It can be carried out.

On the other hand, a general sampling frequency in a pulse oximeter or the like is about 30 to 100 Hz, and a high frequency exceeding several hundred Hz is unnecessary. For this reason, in a pulse oximeter or the like, a photoelectric pulse wave measurement operation is performed after sampling at a high frequency as described above and determining a commercial frequency before starting measurement.
JP 2003-235823 A

  As described above, in the conventional pulse oximeter, since the commercial frequency is determined before the measurement is started, there is a problem that noise cannot be accurately removed when the environment of the commercial frequency is changed during the measurement. Therefore, it is conceivable that sampling can be performed at a high frequency during the measurement operation to enable determination of the commercial frequency. In this case, however, the amount of information processing increases, which places a heavy burden on the calculation unit and increases the current consumption. This will shorten the battery life.

  The present invention has been made in view of the above problems, and provides a biological information measuring apparatus capable of determining a commercial frequency without performing sampling at a high frequency and performing the determination even during a measurement operation. For the purpose.

  A biological information measuring apparatus according to an embodiment of the present invention includes a sensor unit that measures a parameter related to biological information and outputs an analog signal, and the analog signal is obtained from a first commercial frequency and a second commercial frequency that are different from each other. An A / D converter that converts a digital signal at a sampling frequency having a small Nyquist frequency and outputs the digital signal as a data signal, and an arithmetic processing unit that performs a predetermined data analysis process based on the data signal, The arithmetic processing unit extracts a noise signal that is not derived from a living body from the data signal, and detects aliasing noise of a periodic noise component from the noise signal. And a determination unit that identifies the difference between the two commercial frequencies (claim 1).

  Aliasing noise (folding noise) is a digital signal that is converted when an analog signal having a frequency exceeding the Nyquist frequency, which is half the sampling frequency, is input to the A / D converter during A / D conversion. This is noise mixed in the signal, and the signal appears on the frequency axis at the folded position with the Nyquist frequency as the axis of symmetry. Therefore, when the sampling frequency is the same, for example, when 50 Hz noise is mixed in the analog signal and when it is 60 Hz, aliasing noise of different frequency components is detected in the digital signal. By detecting such aliasing noise, the first and second commercial frequencies can be identified without sampling at a high frequency.

  In the above-described configuration, it is desirable that the analog signal of the sensor unit is generated by a photoelectric conversion element. According to this configuration, it is possible to perform measurement related to biological information using non-invasive light and obtain various biological information based on the electrical signal output from the photoelectric conversion element.

  In the above configuration, it is desirable that the determination unit detects aliasing noise by frequency-converting the noise signal to obtain a peak frequency thereof. According to this configuration, aliasing noise can be detected by a simple calculation method called frequency conversion.

  In any one of the configurations described above, it is desirable to further include a measurement control unit that controls the sampling frequency. According to this configuration, the sampling frequency can be freely changed at an appropriate time, for example, during the measurement operation.

  In this case, the measurement control unit may be configured to select the sampling frequency so that the frequency of the signal component derived from the living body is separated from the frequency of the aliasing noise (Claim 5). According to this configuration, it is possible to prevent the influence of aliasing noise from affecting the detection of signal components derived from a living body by adjusting the sampling frequency.

  In addition, it is desirable that the measurement control unit can change the sampling frequency according to the first and second commercial frequencies specified by the determination unit (claim 6). According to this configuration, the aliasing noise frequency of the first and second commercial frequency noises can be controlled by changing the sampling frequency, and the aliasing noise is transferred to a frequency convenient for noise removal, signal processing, and the like. be able to.

  In this case, it is preferable that the measurement control unit selects the sampling frequency so that the aliasing noise appears in an integer of a sampling frequency. According to this configuration, the aliasing noise period matches the sampling period, and the subsequent noise removal processing can be facilitated.

  In any one of the configurations described above, the digital filter further includes a digital filter that removes the noise signal from the data signal, and the filter characteristics of the digital filter are changed according to the first and second commercial frequencies specified by the determination unit. It is desirable that this is possible (claim 8). According to this configuration, the noise signal can be accurately removed according to the first and second commercial frequencies.

  In any one of the above-described configurations, it is desirable that the determination unit performs an operation of identifying the first and second commercial frequencies during the biological information measurement operation. According to this configuration, it is not necessary to determine the commercial frequency before starting the measurement, and it is possible to cope with the case where the environment of the commercial frequency is changed during the measurement.

  In any one of the configurations described above, the analog signal output from the sensor unit is preferably a one-wavelength photoelectric pulse wave signal or a two-wavelength photoelectric pulse wave signal. According to this configuration, it is possible to detect arrhythmia based on the one-wavelength photoelectric pulse wave signal, monitor during defibrillation, diagnosis of autonomic nerve damage, blood vessel age, etc. Medium oxygen saturation can be measured.

  According to the present invention, the sampling frequency for A / D conversion is set to a sampling frequency having a Nyquist frequency smaller than the commercial frequency, so that a large burden is not imposed on the biological information measuring device for discrimination of the commercial frequency. Therefore, current consumption of the biological information measuring device can be suppressed. In many cases, a relatively low sampling frequency is sufficient for measurement of biological information. However, since the commercial frequency can be determined at such a low sampling frequency, the operation for determining the commercial frequency is performed during the measurement of biological information. It is easy to make. For this reason, even when the environment of the commercial frequency changes during measurement, noise can be accurately removed according to the commercial frequency after the change.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a plan view showing a pulse oximeter 1 which is an embodiment of the biological information measuring device of the present invention, and shows a state in which the pulse oximeter 1 is worn on a subject's hand. The pulse oximeter 1 is for measuring the blood oxygen saturation (SpO ₂ ; an example of biological information) of a subject.

  The pulse oximeter 1 includes a pulse oximeter main body 21 that can be attached to the subject's wrist and a probe 22 that can be attached to the fingertip of the subject, and both are electrically connected by a signal cable 201. Being done.

  The pulse oximeter main body 21 is provided with an operation unit 211, a display unit 212 including an LCD, a belt locking unit 213, and the like, and an electric circuit for performing measurement control and arithmetic processing functions is provided therein. It is stored. The operation unit 211 is used by the user to input various instruction information such as measurement start, end, and display instruction. The display unit 212 is for displaying a measurement menu, measurement results, and the like. The probe 22 includes a light emitting element and a light receiving element in order to acquire photoelectric pulse wave information at the fingertip portion.

  FIG. 2 is a block diagram showing the electrical configuration of the pulse oximeter 1. In addition to the aforementioned operation unit 211 and display unit 212, the main body unit 21 includes a light emitting circuit 23, an I / V conversion unit 24 as a light receiving circuit, an A / D conversion unit 25, a control processing unit (CPU) 26, and a memory unit. 27 and a power supply unit 28 are provided.

  The probe 22 includes a light emitting unit 221 and a light receiving unit 222. In the present embodiment, two LEDs, a red LED 221A that generates red light with a wavelength λ1 in the red region and an infrared LED 221B that generates infrared light with a wavelength λ2 in the infrared region, are used as the light emitting unit 221. Is shown. On the other hand, the light receiving unit 222 includes a photoelectric conversion element that generates a current corresponding to the intensity of light received (light emitted from the light emitting unit 221). As the photoelectric conversion element, a silicon photodiode or the like having photosensitivity at least with respect to the wavelength λ1 and the wavelength λ2 is used.

  As shown in FIG. 2, the light emitting unit 221 and the light receiving unit 222 are disposed to face each other with the fingertip of the finger F (biological tissue) to be measured for measuring blood oxygen saturation. This is because the fingertips are particularly easy to optically capture arterial blood pulsations. In the present embodiment, the light emitting unit 221 is arranged on the fingertip back side (nail side), and the light receiving unit 222 is arranged on the fingertip abdomen side. As a result, the measurement light having both wavelengths λ 1 and λ 2 of the light emitting unit 221 that has passed through the living tissue of the fingertip of the finger F to be measured is received by the light receiving unit 222.

  The light emitting unit 221 and the light receiving unit 222 may be arranged so as to be arranged on the fingertip back side or the abdomen side to detect reflected light from the living tissue. In addition, the measurement site may be a site other than the fingertip if measurement is easy and data with a good S / N ratio can be obtained. For example, the earlobe or the back of the hand, the wrist, the back of the foot in the case of an infant It is good also as a measurement part.

  A light emitting circuit 23 and a light receiving circuit (I / V conversion unit 24) are connected to the light emitting unit 221 and the light receiving unit 222, respectively. The light emitting circuit 23 and the I / V conversion unit 24 are provided in the main body unit 21, and electrical connection between the light emitting unit 221 and the light receiving unit 222 is performed by a signal cable 201.

  The operation of the light emitting circuit 23 is controlled by the control processing unit 26, and a predetermined light emission control signal is given to the red LED 221A and the infrared LED 221B of the light emitting unit 221. Thereby, for example, the red LED 221A and the infrared LED 221B are driven alternately (see the light emission control signal in FIG. 2), and the red light and the infrared light are alternately emitted.

  The I / V conversion unit 24 converts the current signal generated by the photoelectric conversion in the light receiving unit 222 into a voltage signal. The current signal output from the light receiving unit 222 includes a photoelectric conversion signal (photoelectric pulse wave signal) emitted from the red LED 221A and the infrared LED 221B and transmitted through the biological tissue of the fingertip. In addition, the light (light component) obtained by photoelectric conversion of the light of the light source existing in the above is included. Furthermore, noise components due to fluorescent lamps, commercial frequencies, and the like are also included. The I / V conversion unit 24 outputs an analog photoelectric pulse wave signal corresponding to the amount of light received by the light receiving unit 222 to the A / D conversion unit 25.

  The A / D conversion unit 25 performs A / D conversion on the analog photoelectric pulse wave signal (see the light reception signal in FIG. 2) given from the I / V conversion unit 24, and sends a digital photoelectric pulse wave signal ( Data signal). The A / D converter 25 performs an A / D conversion operation at a predetermined sampling frequency based on the timing signal given from the control processor 26. This sampling frequency will be described in detail later.

  FIG. 3 shows an example of the photoelectric pulse wave waveform. The ◆ marks in the figure are sampling points (sampling frequency = 30 Hz). The pulse rate of the human body is generally in the range of 20 to 240 times / minute. When this is converted into frequency, it is about 0.3 Hz to 4 Hz. The example illustrated in FIG. 3 is a signal of about 0.7 Hz. That is, the signal component derived from the living body in the pulse oximeter 1 is such a low-frequency signal component. Since the noise of 50 Hz (first commercial frequency) or 60 Hz (second commercial frequency) is superimposed on the photoelectric pulse wave waveform in Japan, the commercial frequency of the usage environment of the pulse oximeter 1 is either 50 Hz or 60 Hz. After determining whether or not there is a noise, noise is removed by performing a filtering process suitable for each.

The blood oxygen saturation (SpO ₂ ) of arterial blood can be derived using such a photoelectric pulse waveform. Oxygen is transported by oxidation / reduction of hemoglobin in the blood. When this hemoglobin is oxidized, the absorption of red light decreases and the absorption of infrared light increases. Conversely, when it is reduced, the absorption of red light increases and the absorption of infrared light decreases. It has characteristics. Using this characteristic, the value of SpO ₂ can be obtained by measuring the variation in the amount of transmitted light of red light and infrared light.

Here, the principle of measuring SpO ₂ using light will be described. As is well known, oxygen is transported to each cell of the living body by hemoglobin (Hb). Hemoglobin combines with oxygen in the lung to become oxygenated hemoglobin (HbO ₂ ), and hemoglobin is consumed when oxygen is consumed by the cells of the living body. Return to. SpO ₂ indicates the ratio of oxygenated hemoglobin in the blood. When the hemoglobin concentration is represented by C _Hb and the oxygenated hemoglobin concentration is represented by C _HbO 2, it is defined as the following equation (1).

On the other hand, the absorbance of hemoglobin and the absorbance of oxyhemoglobin have wavelength dependence, and each extinction coefficient α (λ) has an extinction characteristic as shown in FIG. 4, for example. The horizontal axis in FIG. 4 is the wavelength of light shown in nm, and the vertical axis is the extinction coefficient shown in x10 ⁻⁹ cm ² / mole. Hemoglobin (Hb) and oxygenated hemoglobin (HbO ₂ ) have different light absorption characteristics as shown in FIG. Hemoglobin absorbs more light in the red light R in the red region than oxyhemoglobin, but absorbs less light in the infrared light IR in the infrared region than oxyhemoglobin.

That is, for example, the wavelength λ1 of the red light R is set to 660 nm where the difference between the absorption coefficients of oxygenated hemoglobin and hemoglobin is the largest, for example, and the wavelength λ2 of the infrared light IR is set to 815 nm where the difference between the absorption coefficients of oxygenated hemoglobin and hemoglobin is equal. In this case, the amount of transmitted light of red light R increases as the amount of hemoglobin increases. On the other hand, the amount of transmitted light of infrared light IR does not change even if the ratio of oxygenated hemoglobin to hemoglobin changes. Thus, SpO ₂ can be obtained by taking the ratio of the amount of transmitted light between the red light R and the infrared light IR. The pulse oximeter 1 obtains SpO ₂ by using the difference in absorption characteristics of hemoglobin and oxyhemoglobin with respect to red light R and infrared light IR.

  When a living body is irradiated with light, part of the light is absorbed and part is transmitted. The living body is composed of an arterial blood layer, a venous blood layer, and a tissue other than the arterial blood layer and the venous blood layer (tissue other than blood). As shown in FIG. 5A, the light absorption in the living body includes absorption by the arterial blood layer, absorption by the venous blood layer, and absorption by tissues other than the arterial blood layer and the venous blood layer. Since tissues other than the arterial blood layer and the venous blood layer and the venous blood layer do not change with time, the absorption of light in this portion is substantially constant. On the other hand, since arterial blood changes in blood vessel diameter due to heartbeat, the absorption of light by the arterial blood layer, that is, the intensity of transmitted light (transmitted light shown in FIG. 5A), depends on the pulse as shown in FIG. Changes over time. The amount of change in the transmitted light intensity is based on information on arterial blood alone, and includes almost no influence from venous blood or living tissue other than arterial blood and venous blood. In FIG. 5B, the horizontal axis represents time, and the vertical axis represents transmitted light intensity.

By the way, when comparing the light quantity change between the red light R and the infrared light IR, it is necessary to cancel (correct) the difference in the incident light quantity. FIGS. 6A to 6C are schematic views showing the relationship between incident light and transmitted light on a living body. As shown in FIG. 6 (a), to the same quantity of incident light I ₀ to the living body by the red light R and the infrared light IR is substantially difficult, even if the if the same, tissue and vein Since the absorbance due to blood differs between red light R and infrared light IR, it is not possible to make a comparison based only on the change.

  Here, the transmitted light amount when the artery is the thinnest (when the transmitted light amount is the largest) is I, and the transmitted light amount when the artery is the thickest (when the transmitted light amount is the smallest) is I−ΔI (symbol “−”). Indicates subtraction). As shown in FIG. 6 (a) or 6 (b), it is considered that when the arterial blood having a thickness (width) ΔD is irradiated with light I, transmitted light of I−ΔI is obtained. The state of change in the transmitted light amount according to ΔD arterial blood is shown in the graph of FIG. 6C (the horizontal axis represents time, and the vertical axis represents the transmitted light amount).

Accordingly, as shown in FIG. 7, the transmitted light amount I _R of the red light R shown in the graph of the code 51, the transmitted light amount I _IR of the infrared light IR shown in the graph of the code 52 is normalized to have the same , i.e. as shown in the graph of the code 53, by normalizing such a transmitted light amount I _IR '= transmitted light quantity I _R corresponding to the transmitted light amount I _IR, which is the ratio of quantity of light caused by the arterial blood ([Delta] I _R / SpO ₂ can be calculated by obtaining I _R ) / (ΔI _IR / I _IR ) (the symbol “/” indicates division).

  By the way, the relationship between incident light and transmitted light can be expressed by the following equation (2) according to Lambert Beer's law. However, in the following formula (2), E represents the extinction coefficient of the light-absorbing material, and C represents the concentration of the light-absorbing material.

The two wavelengths of the red light R and the infrared light IR are respectively applied to the above formula (2) (I in the formula (2) is replaced with I _R or I _IR ), and by taking the ratio thereof, the following (3 ) Formula can be obtained. However, in the above (3), _{I R} represents a transmitted light amount of the red light _{R, I IR} represents the amount of transmitted infrared light IR, _{E R} represents an extinction coefficient of the red light _{R, E IR} red It represents the extinction coefficient of external light IR.

FIG. 8 is a graph showing the relationship between the extinction coefficient ratio and SpO ₂ . The horizontal axis in FIG. 8 is SpO ₂ , and the vertical axis is the extinction coefficient ratio. As shown in the figure, for example, when the wavelengths of the red light R and the infrared light IR are about 660 nm and about 815 nm, respectively, the relationship between the extinction coefficient ratio (E _R / E _IR ) and SpO ₂ decreases downward. Represented by a straight line. Based on such a relationship, SpO ₂ can be calculated from the ratio of extinction coefficients. As the method of calculating the SpO _2, is not limited to this method, various methods can be adopted.

The control processing unit 26 includes a ROM that stores a predetermined control program, a RAM that temporarily stores data, a central processing unit (CPU) that reads and executes the control program from the ROM, and a DSP (Digital Signal Processor). Thus, it controls the operation of the entire pulse oximeter 1. The control processing unit 26 controls the operation of the light emitting unit 221 and the light receiving unit 222 based on the data and programs stored in these RAMs and ROMs, as described above from the acquired two-wavelength photoelectric pulse wave data. An arithmetic process or the like for calculating the SpO ₂ value is performed based on the calculation method.

The memory unit 27 includes an EEPROM (Electrically Erasable Programmable ROM) that temporarily stores data such as arithmetic processing and control processing, and a nonvolatile memory such as a flash memory. The memory unit 27 stores the two-wavelength photoelectric pulse wave data or the instantaneous SpO ₂ value for each sampling period calculated by the control processing unit 26 based on the two-wavelength photoelectric pulse wave data in association with the time information. By associating with time information in this manner, post-mortem data analysis or live display of SpO ₂ values can be easily performed.

  The power supply unit 28 includes a predetermined power supply circuit, a power supply battery such as a button battery, and the like, and supplies a drive voltage to each part of the pulse oximeter 1.

  FIG. 9 is a functional block diagram showing a functional configuration of the control processing unit 26 (arithmetic processing unit). The control processing unit 26 includes a noise extraction unit 261, a band pass filter (BPF) 262, a biological information calculation unit 263, a display control unit 264, a measurement control unit 265, and a determination unit 266.

  The noise extraction unit 261 functions to extract a noise signal not derived from a living body from the data signal provided from the A / D conversion unit 25. Specifically, the noise extraction unit 261 performs a process of subtracting the output signal component (dark component) sampled during non-light emission from the output signal component sampled during light emission of the red LED 221A and the infrared LED 221B. This process is for removing the influence of external light that is superimposed on the light transmitted through the living body and incident on the light receiving unit 222.

  FIG. 10 is a schematic diagram for explaining dark component subtraction processing in the noise extraction unit 261. S1 and S3 are λ1 output signals detected when the red LED 221A emits red light of wavelength λ1, and S2 and S4 are λ2 output signals detected when the infrared LED 221B emits infrared light of wavelength λ2. To do. Further, an output signal detected during the light emission interval (when no light is emitted) of the red LED 221A and the infrared LED 221B is referred to as a dark component D (D1 to D4).

  In this case, since the dark component D is superimposed on the output signals S1 to S4, the noise extraction unit 261 performs the subtraction process of “S3-D2” on the λ1 output signal S3, for example, the λ2 output signal. For S2, the λ1 signal component and the λ2 signal component derived from the living body from which the dark component D is removed are calculated by performing the subtraction process of “S2-D1”. Alternatively, the noise extraction unit 261 calculates the λ1 signal component by performing a subtraction process of “S3− (D2 + D3) / 2” on the λ1 output signal S3, for example. The dark component D is used by the determination unit 266 to determine the commercial frequency.

  The band pass filter 262 is configured by a variable digital filter (may be a fixed digital filter), and performs a filtering process to remove a noise component from the photoelectric pulse wave signal provided from the A / D conversion unit 25. The filter characteristics of the bandpass filter 262 can be changed according to the commercial frequency discrimination result in the judgment unit 266. That is, it is possible to change the filter characteristic suitable for noise removal with commercial frequency = 50 Hz and the filter characteristic suitable for noise removal with commercial frequency = 60 Hz.

The biological information calculation unit 263 performs a calculation process for obtaining the SpO ₂ value based on the photoelectric pulse wave signal after the filtering process. This calculation process is a process of first obtaining an instantaneous SpO ₂ value for each sampling period (about 30 Hz to 60 Hz) and performing an averaging process and then obtaining an SpO ₂ value for each unit time.

As the instantaneous SpO ₂ value, two-wavelength photoelectric pulse wave data composed of the λ1 signal component and λ2 signal component (each of which has been subjected to dark component subtraction processing and bandpass filter processing) derived from the above-described living body is used, and each sampling period is used. It is calculated by taking the ratio of the two. The averaging process here is intended to remove noise, and excludes data that is clearly determined to be abnormal from the instantaneous SpO ₂ values (unnatural data caused by missing measurements, body movements, etc.). This is a process of taking a moving average of a plurality of instantaneous SpO ₂ values. Using such an instantaneous SpO ₂ value after the moving average processing, for example, an SpO ₂ value for every _second is calculated.

The display control unit 264 generates a control signal for causing the display unit to display the calculation result of the biological information calculation unit 263 in an appropriate display form. For example, the display control unit 264 displays an average value of SpO ₂ values obtained every _second , displays an SpO ₂ curve in which the SpO ₂ value is developed on the time axis, and displays SpO ₂ based on the SpO ₂ curve. A control signal is generated for detecting the presence of Dip, which is a peak of decrease in value, and displaying the result.

  The measurement control unit 265 controls the light emission timing of the light emission unit 221 and the A / D conversion timing according to a predetermined measurement program. Specifically, the measurement control unit 265 gives a timing pulse to the light emitting circuit 23 and the A / D conversion unit 25 to cause the light emitting unit 221 to emit light at every sampling period, and to synchronize with the light emission timing from the light receiving unit 222 to the A. The photoelectric pulse wave signal (data signal) is acquired through the / D conversion unit 25.

  Note that the sampling frequency can be changed, and the measurement control unit 265 changes the sampling frequency according to the situation. However, in this embodiment, at least during the measurement operation, the Nyquist frequency is smaller than the commercial frequency (50 Hz / 60 Hz) due to the relationship of detecting aliasing noise (in other words, the commercial frequency can be determined during the measurement operation). The sampling frequency is selected in the range having The measurement control unit 265 changes the sampling frequency according to, for example, the commercial frequency determination result in the determination unit 266.

  The determination unit 266 is based on the dark component D detected when the red LED 221A and the infrared LED 221B are not emitting light, and the power supply system in the environment where the pulse oximeter 1 is placed is 50 Hz (first commercial frequency). Or 60 Hz (second commercial frequency).

  In the case of discriminating between 50 Hz noise and 60 Hz noise, in order to directly detect each noise waveform, the noise signal needs to be sampled at least at 120 Hz, and preferably at 10 times 1200 Hz. If sampling is performed at such a high frequency, data twice as many as the sampling points shown in FIG. 20 can be obtained, so that it is possible to accurately determine whether the frequency is 50 Hz or 60 Hz.

For this reason, in the conventional general pulse oximeter, the A / D conversion output is taken out at a high speed of 1200 Hz without turning on the light emitting part before the SpO ₂ measurement operation after the power is turned on. Processing to determine the commercial frequency has been performed. The discrimination processing method is a method of specifying a noise frequency by performing frequency conversion processing (Fourier transform processing) on a noise data signal after A / D conversion, for example.

  FIG. 11 is an example of frequency conversion data obtained when a noise waveform with a sampling frequency of 1200 Hz and a commercial frequency of 50 Hz is detected. As shown in FIG. 11, since the frequency peak clearly appears in the frequency conversion data, the frequency of the noise source can be specified. In addition, the noise frequency can also be specified by a method for obtaining the peak interval of the waveform of the noise data signal.

  However, such commercial frequency discrimination that requires data sampling at a high frequency is not performed during the measurement operation because of power consumption or the like, but is performed before the start of measurement. For this reason, conventional pulse oximeters cannot cope with changes in the commercial frequency environment during measurement. Therefore, in this embodiment, a noise data signal (dark component) is detected at a sampling frequency having a Nyquist frequency smaller than the commercial frequency of the noise source, and an additional electric circuit such as a low-pass filter is not required, and the measurement operation is performed. Is also able to determine the commercial frequency. For this reason, the determination unit 266 determines the commercial frequency using aliasing noise that appears when data sampling is performed with a period longer than the period of the noise source.

  FIG. 12 is a schematic diagram showing a state in which a 60 Hz sine wave is sampled (A / D converted) at a sampling frequency of 45 Hz. In this case, the 60 Hz sine wave is sampled only three times in four periods. As a result, in spite of trying to detect a sine wave of 60 Hz, a noise of 15 Hz that does not exist originally is detected. This is aliasing noise.

  FIG. 13 is an example of frequency conversion data obtained when a noise waveform with a sampling frequency = 45 Hz and a commercial frequency = 60 Hz is detected. Due to the cause shown in FIG. 12, aliasing noise appears at 15 Hz, which is a frequency obtained by folding 60 Hz, with the Nyquist frequency (22.5 Hz in this case) as the axis of symmetry on the frequency axis.

  On the other hand, when a sine wave of 50 Hz is sampled at a sampling frequency of 45 Hz, the aliasing noise of 5 Hz is detected as a result of sampling 9 times in 10 cycles. FIG. 14 is an example of frequency conversion data obtained when a noise waveform with a sampling frequency = 45 Hz and a commercial frequency = 50 Hz is detected. As shown in the figure, aliasing noise appears at 5 Hz, which is a frequency obtained by turning 50 Hz around the Nyquist frequency.

  Thus, when the sampling frequency is the same, aliasing noise of different frequency components is detected in the digital signal after A / D conversion when 50 Hz noise is mixed in the analog signal and when it is 60 Hz. Become so. Therefore, by detecting aliasing noise, it becomes possible to specify 50 Hz / 60 Hz without sampling at a high frequency.

  In view of the above, the determination unit 266 uses the dark component D (FIG. 10) used in the subtraction process in the noise extraction unit 261 to perform an operation for obtaining the aliasing noise of the commercial frequency noise component. When the 50 Hz / 60 Hz distinction is specified, the determination result is output to the measurement control unit 265. In response to this, the measurement control unit 265 appropriately changes the sampling frequency for the A / D conversion unit 25 and the filter characteristics of the band pass filter 262. This point will be described with an example based on FIGS.

  As shown in FIG. 15, when the data signal is sampled from the A / D converter 25 under the environment where the sampling frequency is 45 Hz and the commercial frequency is 50 Hz, the noise waveform N1 of 50 Hz and the Nyquist frequency of 22.5 Hz are turned back. As a result, aliasing noise N2 of 5 Hz is detected. As described above, the signal component a derived from the living body in the pulse oximeter 1 is about 0.3 Hz to 4 Hz corresponding to the pulsation. In this case, since the frequencies of the signal component a and the aliasing noise N2 are close to each other, it is difficult to filter the aliasing noise N2. On the other hand, when the commercial frequency is 60 Hz, as shown in FIG. 13, aliasing noise is 15 Hz, and the frequency is separated from the signal component a, so noise removal is relatively easy.

  Therefore, when the determination unit 266 determines that the commercial frequency = 50 Hz, the measurement control unit 265 changes the sampling frequency so that the aliasing noise N2 appears at a frequency separated from the frequency band of the signal component a. For example, as shown in FIG. 16, if the sampling frequency is 37.5 Hz, the aliasing noise N2 of 12.5 Hz is detected with the Nyquist frequency of 18.75 Hz as the turning axis. As a result, the aliasing noise N2 and the signal component a are separated in frequency, and noise removal becomes easy. Accordingly, the measurement control unit 265 changes the filter characteristics of the bandpass filter 262 to a target that targets 12.5 Hz, and effectively removes noise components in the digital signal.

Here, when the measurement control unit 265 changes the sampling frequency, it is desirable to select the sampling frequency so that aliasing noise appears in an integer of the sampling frequency. This contributes to simplification of the averaging process in the biological information calculation unit 263. When obtaining the instantaneous SpO ₂ value, the biological information calculation unit 263 performs, for example, a process of taking a moving average of the λ1 signal component and the λ2 signal component after the bandpass filter processing.

In the moving average process of the λ1 signal component and the λ2 signal component, in the case of an environment where the sampling frequency is 45 Hz and the commercial frequency is 60 Hz, as shown in FIG. The phase matches the commercial frequency. Therefore, noise can be canceled by taking the moving average of the values of the λ1 signal component and the λ2 signal component at three points continuous in the time axis direction. In this way, it is possible to reduce noise by moving average calculation with a small number of sampling points.
Sampling frequency (45 Hz): aliasing noise (15 Hz) = 3: 1
This is because they are in an integer multiple relationship.

  On the other hand, in the case of an environment where the sampling frequency is 45 Hz and the commercial frequency is 50 Hz, as shown in FIG. 17, 10 cycles of the commercial frequency (t11 to t20) and 9 cycles of the sampling frequency (s11 to s19). And are consistent. In other words, in terms of the sampling period, the commercial frequency and the phase are matched once every nine periods. Accordingly, by obtaining the moving average value DA1 of the values of the λ1 signal component and the λ2 signal component (d1 to d9) in units of nine points that are continuous in the time axis direction, noise can be canceled between the moving average acquisition point group units. Can do.

Again,
Sampling frequency (45 Hz): aliasing noise (5 Hz) = 9: 1
Since the two have an integer multiple relationship, noise can be canceled, but it is desirable that the number of data points when taking a moving average is small. In this case, it suffices to change the sampling frequency so that the phases are matched with a smaller period. Here, when the commercial frequency is 50 Hz, as shown in FIG. 16, if the sampling frequency is 37.5 Hz, aliasing noise of 12.5 Hz is generated. At this time,
Sampling frequency (37.5 Hz): aliasing noise (12.5 Hz) = 3: 1
Therefore, the moving average can be obtained in units of three points as in the case of 60 Hz described above.

  That is, as shown in FIG. 18, the four cycles of commercial frequencies (t11 to t14, t15 to t18) and the three cycles of sampling frequencies (s11 to s13, s14 to s16) are matched. For this reason, by obtaining the moving average values DA1 and DA2 of the values (d1 to d3, d4 to d6) of the λ1 signal component and the λ2 signal component in units of three points that are continuous in the time axis direction, these moving average values DA1, Noise between DA2 can be canceled. Therefore, noise can be reduced while simplifying the arithmetic processing.

  The measurement control unit 265 sets the sampling frequency to about 1200 Hz in a state where the light emitting unit 221 is not turned on, for example, when the commercial frequency is determined by the same method as the conventional method at the start of measurement. The data signal at this time is directly input from the A / D conversion unit 25 to the determination unit 266, and the determination unit 266 performs a commercial frequency determination process by the same method as the conventional method. Note that this function of the measurement control unit 265 may be omitted.

  The operation of the pulse oximeter 1 according to this embodiment configured as described above will be described based on the flowchart shown in FIG. When the power of the pulse oximeter 1 is turned on, the measurement control unit 265 sets the sampling frequency of the A / D conversion unit 25 to 1200 Hz, and the noise data signal is acquired without lighting the light emitting unit 221. The noise data signal is frequency-converted by the determination unit 266, and the peak frequency is obtained to detect the commercial frequency of the environment in the initial measurement (step # 1).

  Then, in accordance with the detected commercial frequency, the sampling frequency of the A / D converter 25 and the filter characteristics of the bandpass filter 262 are initialized (step # 2). For example, when the commercial frequency of the environment is 60 Hz, the sampling frequency = 45 Hz and the filter characteristics of the bandpass filter 262 are set for removing aliasing noise of 15 Hz. Note that steps # 1 and # 2 may be omitted.

Thereafter, the mode shifts to a photoelectric pulse wave measurement mode (step # 3). The measurement control unit 265 causes the light emitting unit 221 to execute the red light having the wavelength λ1, the infrared light having the wavelength λ2, and the light projecting operation at a predetermined period. In addition, the measurement control unit 265 causes the noise extraction unit 261 to capture a digital two-wavelength photoelectric pulse wave signal at a sampling frequency of 45 Hz from the A / D conversion unit 25. Through the subtraction of the dark component in the noise extraction unit 261 and the filtering in the band pass filter 262, the instantaneous SpO ₂ value is calculated in the biological information calculation unit 263, and the obtained instantaneous SpO ₂ value is associated with the time information in the memory unit. 27 (step # 4).

  While performing the process of step # 4, it is confirmed whether or not it is a predetermined timing when the noise confirmation operation is performed (step # 5). The noise to be confirmed here is another commercial frequency. If it is time to confirm noise (YES in step # 5), the determining unit 266 detects aliasing noise based on the dark component signal (step # 6).

  From the detection result of the aliasing noise, the determination unit 266 determines whether or not the commercial frequency of the measurement environment has changed (step # 7). If the commercial frequency has changed, for example, if it has changed from 60 Hz to 50 Hz (YES in step # 7), the measurement control unit 265 uses the sampling frequency of the A / D conversion unit 25 = 37.5 Hz, a bandpass filter A setting change is made to set the filter characteristics of 262 to remove aliasing noise of 12.5 Hz (step # 8).

  Thereafter, it is confirmed whether or not to end the measurement operation (step # 9). When the measurement is continued (NO in step # 9), the process returns to step # 4 and is repeated. When the measurement is finished (YES in step # 9), the process is finished. If it is not the timing for noise confirmation (NO in step # 5), steps # 6 to # 8 are skipped. If there is no change in the commercial frequency (NO in step # 7), step # 8 is skipped.

  According to the pulse oximeter 1 according to the present embodiment as described above, the sampling frequency of the A / D conversion unit 25 is lower than the commercial frequency. There is no significant burden on the system. Therefore, the current consumption of the pulse oximeter 1 can be suppressed. Further, since the commercial frequency can be determined using the dark component obtained at a relatively low sampling frequency for measuring the photoelectric pulse wave measurement, it is easy to perform the commercial frequency discrimination operation even during measurement of biological information. . For this reason, even when the environment of the commercial frequency changes during measurement, noise can be accurately removed according to the commercial frequency after the change.

  Although the embodiments of the present invention have been described above, the present invention is not limited to these embodiments, and for example, the following modified embodiments (1) to (4) can be taken.

(1) In the above embodiment, the frequency conversion is performed on the noise data signal, and the peak frequency is obtained to detect the aliasing noise, that is, specify the commercial frequency. Alternatively, the commercial frequency may be specified by detecting the period of noise. As described above, when the sampling frequency is selected to be 45 Hz, when the commercial frequency is 60 Hz, a cycle appears for every three sampling points, and when the commercial frequency is 50 Hz, a cycle appears for every nine sampling points. . Therefore, the commercial frequency can be specified by obtaining the noise period.

(2) In the above-described embodiment, an example in which the sampling frequency is set to 45 Hz and the sampling frequency is set to 37.5 Hz when the commercial frequency is detected to be 50 Hz is shown. These sampling frequencies are merely examples, and any sampling frequency having a Nyquist frequency lower than the commercial frequency may be used. Among these, it is desirable that the sampling frequency is a low frequency of commercial frequency (50 Hz) or less.

(3) In the said embodiment, the technique which reduces noise by taking a moving average in the biometric information calculating part 263 was illustrated. Instead, a high-order filter such as a FIR (Finite Impulse Response) filter or an IIR (Infinite Impulse Response) filter may be used.

(4) In the said embodiment, the pulse oximeter 1 which detects a two-wavelength photoelectric pulse wave and calculates | requires blood oxygen saturation as an example of a biological information measuring device was illustrated. In addition, detection of pulse and arrhythmia (atrial fibrillation / extra systole), monitoring during defibrillation based on single-wavelength photoelectric pulse wave (in this case, only one light emitting element of the light emitting unit 221 is necessary), autonomous The present invention can also be applied to a biological information measuring apparatus that diagnoses neurological disorders, blood vessel age, and the like. Furthermore, in addition to the pulse wave, the present invention can also be applied to an apparatus for measuring biological information such as electrocardiogram, mouth / nasal flow, chest / abdominal movement and the like.

It is a top view which shows the pulse oximeter which is one Embodiment of the biometric information measuring apparatus of this invention. It is a block diagram which shows the electric constitution of a pulse oximeter. It is a graph which shows an example of a photoelectric pulse wave waveform. It is a graph which shows the light absorption characteristic of hemoglobin and oxyhemoglobin. It is explanatory drawing which shows typically the absorption state of the light by a biological body. It is explanatory drawing which shows typically the relationship between the incident light and incident light which inject into a biological body. It is explanatory drawing for demonstrating normalization of the transmitted light amount by infrared light. Is a graph showing the relationship between the ratio and the SpO ₂ extinction coefficient. It is a functional block diagram which shows the function structure of a control processing part (arithmetic processing part). It is a schematic diagram for demonstrating the dark component subtraction process in a noise extraction part. It is explanatory drawing which shows an example of the frequency conversion data obtained when the noise frequency of sampling frequency = 1200Hz and commercial frequency = 50Hz is detected. It is a schematic diagram which shows the state which is sampling (A / D conversion) the sine wave of 60 Hz with the sampling frequency of 45 Hz. It is explanatory drawing which shows an example of the frequency conversion data obtained when the noise frequency of sampling frequency = 45Hz and commercial frequency = 60Hz is detected. It is explanatory drawing which shows an example of the frequency conversion data obtained when the noise frequency of sampling frequency = 45Hz and commercial frequency = 50Hz is detected. It is explanatory drawing which shows the aliasing noise which appears when sampling frequency = 45Hz and commercial frequency = 50Hz. It is explanatory drawing which shows the aliasing noise which appears when sampling frequency = 37.5Hz and commercial frequency = 50Hz. It is a schematic diagram for demonstrating a moving average process. It is a schematic diagram for demonstrating a moving average process. It is a flowchart which shows operation | movement of the pulse oximeter which concerns on embodiment of this invention. It is a schematic diagram for demonstrating the sampling of the conventional measurement data.

Explanation of symbols

1 Pulse Oximeter 21 Main Body 211 Operation Unit 212 Display Unit 22 Probe (Part of Sensor Unit)
221 Light emitting unit 222 Light receiving unit (photoelectric conversion element)
23 Light Emitting Circuit 24 I / V Converter 25 A / D Converter 26 Control Processing Unit (Calculation Processing Unit)
261 Noise extraction unit 262 Band pass filter (digital filter)
263 Biometric information calculation unit 264 Display control unit 265 Measurement control unit 266 Judgment unit 27 Memory unit 28 Power supply unit

Claims (10)

A sensor unit that measures parameters related to biological information and outputs an analog signal;
The analog signal is converted into a digital signal at a sampling frequency having a Nyquist frequency smaller than the first commercial frequency and the second commercial frequency, and an A / D converter that outputs the digital signal as a data signal; An arithmetic processing unit that performs predetermined data analysis processing based on
The arithmetic processing unit includes:
A noise extraction unit that extracts a noise signal not derived from a living body from the data signal;
A determination unit that identifies the first and second commercial frequencies by detecting aliasing noise of a periodic noise component from the noise signal;
A biological information measuring device comprising:   The biological information measuring apparatus according to claim 1, wherein the analog signal of the sensor unit is generated by a photoelectric conversion element.   The biological information measuring apparatus according to claim 1, wherein the determination unit detects aliasing noise by frequency-converting the noise signal and obtaining a peak frequency thereof.   The biological information measuring apparatus according to claim 1, further comprising a measurement control unit that controls the sampling frequency.   5. The biological information measuring apparatus according to claim 4, wherein the measurement control unit selects the sampling frequency so that a frequency of a signal component derived from a living body and a frequency of the aliasing noise are separated from each other. .   The living body according to claim 4, wherein the measurement control unit is capable of changing the sampling frequency according to the first and second commercial frequencies specified by the determination unit. Information measuring device.   The biological information measurement apparatus according to claim 6, wherein the measurement control unit selects the sampling frequency so that the aliasing noise appears in an integer of a sampling frequency. A digital filter for removing the noise signal from the data signal;
The biometric information according to any one of claims 1 to 3, wherein a filter characteristic of the digital filter is changeable according to the first and second commercial frequencies specified by the determination unit. measuring device.   The biological information according to claim 1, wherein the determination unit performs an operation of identifying the first and second commercial frequencies during a biological information measurement operation. measuring device.   The biological information measuring device according to claim 1, wherein the analog signal output from the sensor unit is a one-wavelength photoelectric pulse wave signal or a two-wavelength photoelectric pulse wave signal.

Images