JP2001212100A - Equipment for measuring electrical characteristics - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide equipment for measuring electrical characteristics that calculates bioelectric impedance and estimates distributions of body fat and body fluid while controlling influence caused by the outside noise in a low-frequency region. SOLUTION: The measuring equipment 100 produces measuring signals Ia which have relatively stronger spectrum in a low-frequency comparing to that in a high-frequency depending on the frequency of measuring signals. The equipment is equipped with a generator 72 of measuring signals that input measuring signals Ia to an examinee through superficial electrodes Hc, Lc, an analog-to- digital converter 93 that measures electric current of the measuring signals that is input to the examinee, an analog-to-digital converter 83 that measures voltage generated between the body surfaces of the examinee, and a central processing unit 3 that estimates bioelectric impedance between the body surfaces of the examinee and a physical value depending on the impedance from the measured electric current and voltage.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electrical characteristic measuring apparatus useful for estimating a body fat state and a body water distribution of a subject based on a bioelectrical impedance method.

[0002]

2. Description of the Related Art The inventor of the present invention has previously applied for an apparatus using an M-sequence signal as a bioelectrical impedance measuring apparatus (Japanese Patent Laid-Open No. 10-14898). In the invention, by performing a Fourier transform on the four-terminal A / D-converted signal, bioelectric impedance at many frequencies is measured to calculate information on the water content inside and outside the cell. Although not described in the specification, this apparatus outputs an M-sequence signal many times and performs synchronous addition of each signal in order to improve the SN ratio of the signal.

[0003] The prior art will be described below. 2. Description of the Related Art In recent years, research on the electrical characteristics of living organisms has been conducted for the purpose of evaluating the body composition of humans and animals. The electrical properties of living organisms vary significantly depending on the type of tissue or organ, for example,
In the case of humans, the electrical resistivity of blood is around 150 Ω · cm, whereas the electrical resistivity of bone and fat is 1 to 5 kΩ · cm.
cm. The electrical characteristics of the living body are called bioelectric impedance, and are measured by passing a small current between a plurality of electrodes attached to the body surface of the living body. The method of estimating the body water distribution, body fat percentage, and body fat mass of the subject from the bioelectric impedance obtained in this way is called bioelectric impedance method ("Bioelectric impedance method as a method for evaluating body composition", Baumgartner , RN, etc.
Author, "Bioelectric impedance and its clinical application", Medical electronics and biotechnology, Hiroshi Kanai, 20 (3) Jun 1982, "Estimation of body limb water distribution by impedance method and its application", Medical electronics and biotechnology. See, Makoto Haeno et al., 23 (6) 1985, "Long-term measurement of urinary bladder volume by impedance method", Ergonomics, Yasuo Kuchinomachi, 28 (3) 1992, etc.).

[0004] Bioelectric impedance is derived from the resistance of a living body to the current carried by ions in the body (resistance) and the reactance associated with various types of polarization processes created by cell membranes, tissue interfaces, or non-ionized tissue. Be composed. Capacitance, which is the reciprocal of reactance, causes a time delay in current rather than voltage, and creates a phase shift (phase shift). This value is the arctangent of the ratio of reactance to resistance (arctangent). It can be determined geometrically as a phase angle.

The bioelectric impedance Z, the resistance R, the reactance X and the electric phase angle φ depend on the frequency. At very low frequencies f _L , the bioelectrical impedance Z at the cell membrane-tissue interface is too high to conduct electricity. Thus, electricity flows only through the extracellular fluid and the measured bioelectrical impedance Z is purely a resistance R.

Next, as the frequency increases, the current penetrates the cell membrane, the reactance X increases, and the phase angle φ increases. The magnitude of the bioelectrical impedance Z is equal to the value of the vector defined by Z ² = R ² + X ² . The frequency at which both the reactance X and the phase angle φ are maximized is called a critical frequency f _C , which is one electrical characteristic value of a living body that is a conductive conductor. Beyond this critical frequency f _C , the cell membrane-tissue interface loses its capacitive capacity, and the reactance X decreases accordingly. At very high frequencies f _H , the bioelectrical impedance Z is again purely equivalent to the resistance R.

FIG. 5 is an electrical equivalent circuit diagram (equivalent circuit model) of the human body. In this figure, Cm represents the cell membrane capacity, and Ri and Re represent the intracellular fluid resistance and the extracellular fluid resistance, respectively. At low frequencies f _L , the current is mainly flowing in the extracellular space, and the bioelectrical impedance Z is equal to the extracellular fluid resistance Re. At high frequencies f _H , the current passes completely through the cell membrane, and the cell membrane capacitance Cm is substantially equivalent to being short-circuited. Therefore, the bioelectrical impedance Z at a high frequency f _H is the combined resistance Ri · Re / (Ri + R
e).

According to the method described above, the intracellular fluid resistance R
i and the extracellular fluid resistance Re can be determined, and based on these, the body fat status and body water distribution (intracellular fluid volume, extracellular fluid volume, etc.) of the subject such as body fat percentage, fat weight, lean body mass, etc. Volume and the total body water content).
The change in the body water distribution can be estimated from the change in the resistances Re and Ri. As a bioelectrical impedance measuring device that performs measurement / estimation of each of these parameters by injecting a small sine wave current of a plurality of frequencies arbitrarily selected into a living body and digitally processing the obtained signal, there is a special table. Hei 6
The thing described in 506854 is known.

[0009]

In a conventional bioelectrical impedance measuring apparatus, an impedance locus is obtained from the calculated bioelectrical impedance by using a curve fitting method such as a least square method. Next, from the obtained impedance locus, the bioelectric impedance R0 of the subject's body at the frequency 0 and the bioelectric impedance R∞ at the infinite frequency are calculated, and from the calculation result,
The intracellular fluid resistance and the extracellular fluid resistance of the subject's body are calculated.

FIG. 6A is a diagram exemplifying a frequency spectrum of a measurement signal (M-sequence signal) applied to a subject's body in a conventional bioelectric impedance measuring apparatus. As shown in FIG. 6A, the spectral density of the M-sequence signal is constant over the entire frequency band, and the spectral intensity of each frequency is equal.

FIG. 6B is a diagram illustrating an example of an impedance locus of bioelectric impedance calculated for each frequency. Since the frequency spectrum of the measurement signal has a constant density and the same intensity over the entire frequency band as described above, the frequency spectrum has the same error (likelihood) range a for each frequency. Therefore, when curve fitting is performed as an impedance locus using a curve fitting method such as the least squares method, the high frequency side plot is dense and the curve fitting is easy, but the low frequency side plot is sparse and the curve is accurately curved. It is difficult to fit, and as a result, the low frequency region is more susceptible to external noise. SUMMARY OF THE INVENTION An object of the present invention is to provide an electrical characteristic measuring apparatus which avoids the influence of external noise on the low frequency side and is suitable for measuring bioelectric impedance, estimating the state of body fat and body water distribution, and the like.

[0012]

SUMMARY OF THE INVENTION An electrical characteristic measuring apparatus according to the present invention generates a signal whose spectrum intensity on a low frequency side is relatively larger than that on a high frequency side in accordance with the frequency of a measurement signal. , A predetermined 2
Measurement signal supply means for applying the signal to the subject via first and second electrodes conductively attached to the surface of the location, and a current for measuring a current value of the signal applied to the subject Measuring means, voltage measuring means for measuring a voltage value generated between two predetermined surface portions separated from each other of the subject, by the current value and the voltage value respectively measured by the current measuring means and the voltage measuring means, Calculating means for calculating a bioelectric impedance between the surface portions of the subject and calculating a bioelectric impedance to be obtained or a physical quantity based on the bioelectric impedance.

Further, the measurement signal supply means has a filter means for generating the signal in accordance with the frequency of the measurement signal, so that the spectrum intensity on the low frequency side can be increased by using the attenuation characteristic of the filter. It is possible to generate a measurement signal that is relatively large compared to the side spectral intensity. Also,
The absolute value of the transfer function of the filter means is a constant in a low frequency region and decreases in inverse proportion to the frequency on the high frequency side, so that the spectrum intensity on the frequency side higher than the reference frequency can be attenuated. The effect of external noise on the frequency side can be further suppressed.

Further, the absolute value of the transfer function of the filter means is proportional to (1 / (1+ (f / fo) ² )) ^1/2 (f: frequency, fo: reference frequency), so that a simple operation is possible. The filter means can be designed based on a design method. Further, since the filter means is realized by a primary CR circuit or LR circuit, the filter means can be realized with a simple configuration.

The reference frequency is 10 kHz to 3 kHz.
By setting the frequency in the range of 00 kHz, the frequency range in which the spectrum intensity is attenuated can be freely set in consideration of the critical frequency (generally about 40 kHz) unique to the subject. Also, the measurement signal is a longest linear code sequence (M sequence), an impulse signal, a chirp signal, or a sine wave whose frequency changes stepwise at equal intervals in time, so that the spectral density is constant over the entire frequency band. A signal having the same intensity and whose properties and generation principle are already well known can be used for measurement.

[0016]

Embodiments of the present invention will be described below in detail with reference to the drawings. The case where the electrical characteristic measuring device of the present invention is used for a bioelectrical impedance measuring device will be described in detail.

FIG. 1 is a block diagram showing an electrical configuration of the bioelectrical impedance measuring device according to the present embodiment. The bioelectrical impedance measuring device includes a keyboard 1
The probe current Ia is transmitted as a measurement signal to the body B of the subject, and thereby the measurement processing unit 2 for digitally processing the voltage / current information obtained from the body B of the subject, and the respective units of the apparatus are controlled and the measurement process is performed. A CPU (Central Processing Unit) 3 for calculating various physical quantities relating to the bioelectrical impedance, body fat, and body water distribution of the human body based on the processing results of the unit 2, and a living body of the subject B calculated by the CPU 3 A display unit 4 for displaying electrical impedance, body fat content, body water content, etc., a ROM 6 for storing a processing program of the CPU 3, and various data (for example, height, weight, sex, extracellular fluid and intracellular (A liquid amount and the like) and a RAM 5 in which a work area of the CPU 3 is set.

The keyboard 1 is used by a measurer to input a measurement start switch for instructing the start of measurement, a human body characteristic item such as height, weight, gender, and age of the subject, a total measurement time T, a measurement interval t, and a measurement interval to be described later. The key 1 comprises various keys for setting / changing a reference frequency fo or the like according to a measurement purpose. Operation data of each key supplied from the keyboard 1 is converted into a key code by a key code generation circuit (not shown). Then, it is supplied to the CPU 3.

The measurement processing unit 2 includes a PIO (parallel interface) 71, a measurement signal generator 72, a low-pass filter (hereinafter, referred to as LPF) 73, a coupling capacitor 74, and a surface electrode Hc to be attached to a predetermined part of the body. Output processing circuit, surface electrodes Hp, Lp, Lc, which are also affixed to predetermined parts of the body, coupling capacitors 80a, 80b, 90, and a differential amplifier 8
1. I / V converter (current / voltage converter) 91, LPFs 82 and 9 comprising analog anti-aliasing filters
2. An input processing circuit including A / D converters 83 and 93 and sampling memories (ring buffers) 84 and 94.

In the measurement processing section 2, the measurement signal generator 7
2 has an output resistance of 10 kΩ or more over the entire range of the generated signal frequency, and every time a signal generation instruction signal is supplied from the CPU 3 via the PIO 71 at a predetermined cycle t during the entire measurement time T, Maximal lin
The ear code) sequence (M sequence) probe current Ia is repeatedly generated a predetermined number of times. As exemplified in FIG.
The spectrum of each frequency of the probe current Ia has a constant density and an equal intensity over the entire frequency band.

The LPF (low-pass filter) 73 makes the spectrum intensity on the low frequency side relatively larger than the spectrum intensity on the high frequency side in accordance with the frequency of the probe current Ia as the measurement signal. In particular, when a measurement signal having a frequency lower than the reference frequency fo is input, the frequency spectrum intensity is emphasized, and when a measurement signal having a high frequency is input, the frequency spectrum intensity is attenuated. In the present embodiment, the LPF 73 is a primary CR circuit or LR
It is realized by a circuit. Each element value is designed so that the amplitude characteristic is flat at a frequency lower than the vicinity of the reference frequency fo, and is attenuated at a high frequency.

FIG. 2 shows an LPF 73 according to this embodiment.
FIG. 6 is a diagram showing an amplitude characteristic of a signal, which is represented by a gain (dB) -a logarithm (logf) of a measurement frequency f. LPF73
Is used which has a substantially flat amplitude characteristic from frequency 0 to the reference frequency fo.
The absolute value of the transfer function of the LPF 73 is a constant in a frequency region lower than the reference frequency, and decreases in inverse proportion to the frequency in a higher frequency region. In the present embodiment, the absolute value of the transfer function is proportional to (1 / (1+ (f / fo) ² )) ^1/2 .

The reference frequency fo is set in the vicinity of the critical frequency f _C. In the present embodiment, the reference frequency fo is about 10 kHz shown in the frequency range b in consideration that the critical frequency f _C of a human is about 40 kHz for an adult.
It is set between 300 kHz. Also, for children,
The critical frequency f _C tends to be higher than that of an adult. However, by setting the reference frequency fo to 300 kHz, even children can measure without any problem.

The amplitude characteristic of the LPF 73 shown in FIG. 2 shows an example in which the reference frequency fo is set to about 40 kHz which is the same as the critical frequency f _{C. In} this case, the absolute value of the transfer function of the LPF 73 is the critical frequency f _C It is a constant in the lower frequency range and decreases in inverse proportion to the frequency in the frequency range higher than the critical frequency f _C. The LPF used in the related art only removes high-frequency noise, and its amplitude characteristic (shown by a dotted line) is substantially flat in a range of a measurement frequency f of about 2 kHz to 600 kHz. Therefore, the measurement signal Ia is output as a frequency spectrum having a constant density and the same intensity without depending on the frequency (see FIG. 6A). On the other hand, L of the present embodiment
The PF 73 attenuates a spectrum having a frequency higher than the reference frequency fo by using a transition region in which the amplitude decreases from the vicinity of the reference frequency fo.

FIG. 3A shows the LPF 73 of the present embodiment.
FIG. 5 is a diagram illustrating the spectrum intensity of each frequency of the measurement signal Ia output by the following. The spectral intensity at frequencies lower than the reference frequency fo (critical frequency f _C ) is emphasized, and the spectral intensity at frequencies higher than the reference frequency fo is attenuated. For this reason, the curve fitting on the low frequency side (see FIG. 3B) is more accurate than in the past. The coupling capacitor 74 includes an LPF 73
The probe current Ia is input to remove the DC component so that the DC current is not applied to the body B of the subject, and is sent to the surface electrode Hc.

The supply period of the signal generation instruction signal coincides with the measurement interval t set by the measurer using the keyboard 1. In this example, the probe current (measurement signal) I
The number of repetitions of “a” is 1 to 2 per signal generation instruction signal.
56 times. The number of repetitions may be arbitrarily set by the measurer using the keyboard 1. The number of repetitions increases as the number of repetitions increases. However, although it is a minute current, it may adversely affect the human body when continuously applied to the human body for a long time. Therefore, the number of repetitions is preferably 1 to 256.

Here, the M-sequence signal will be described. The M-sequence signal is a code signal generally used in a spread-spectrum communication system or a spread-spectrum ranging system, and is the longest code sequence generated by a shift register or a delay element having a certain length. Say. A binary M-sequence generator that generates an M-sequence signal having a length of (2 ⁿ -1) bits (n is a positive integer) is composed of an n-stage shift register and a logical combination of the n-stage states. And a logic circuit (exclusive logic circuit) that feeds back to the input of the shift register. M at a certain sample time (clock time)
The output of the sequence generator and the state of each stage is a function of the output of the feedback stage at the immediately preceding sample time. In this embodiment, an M-sequence generator having eight stages (n = 8) of shift registers is used. Further, the frequency of the shift clock of the shift register is set to 2 MHz.

When the impulse signal is used, energy concentrates in a short time interval (0.1 μsec), whereas the probe current Ia using the M-sequence signal includes many frequency components. The energy is dispersed for about 1 msec without damaging the living body.
Since it occurs at a time interval sufficiently shorter than the pulse or respiratory cycle, these effects are not affected by averaging the measured values over time. Furthermore, for example, in the case of a rectangular wave signal with a duty of 50%, the amplitude of the frequency spectrum is large at low frequencies and small at high frequencies, so that the frequency characteristic of the SN ratio deteriorates in the high frequency region, whereas the M-sequence signal is Since the amplitude of the frequency spectrum is substantially flat over the entire frequency range, the frequency characteristic of the SN ratio is also substantially flat. For details of the M-sequence signal, see “Spread Spectrum Communication System” by RCDixon (p.56 to p.89).

FIG. 4 is a diagram schematically showing a state of use of the bioelectrical impedance measuring device according to the present embodiment. The surface electrode Hc (first electrode) is conductively affixed to the right back part H of the subject by a suction method at the time of measurement,
The surface electrode Lc (second electrode) is conductively attached to the right instep L by an adsorption method. Therefore, the measurement signal (probe current) Ia is transmitted from the right hand of the subject to the body B
to go into.

The surface electrode (high potential output terminal) Hp
Is electrically conductively attached to the right back part H of the subject by suction, and the surface electrode Lp is electrically conductively attached to the right instep L by suction. At this time, the surface electrodes Hc and Lc are attached to portions farther from the center of the human body than the surface electrodes Hp and Lp. Each of the above surface electrodes Hp, L
p, Hc, and Lc are connected to the bioelectrical impedance measuring device 100 by a measuring cable 10.

Next, the measurement signal processing will be described. FIG.
As shown in the figure, the surface electrode (high-potential output terminal) Hp is conductively attached to the right back part H of the subject by a suction method, while the front electrode Lp is sucked to the right foot part L. It is stuck conductively by the method. The differential amplifier 81 shown in FIG. 1 detects a potential (potential difference) between two surface electrodes Hp and Lp. That is, when the probe current Ia is applied to the body B of the subject, the differential amplifier 81 detects the voltage Vp between the right limbs of the subject and inputs the same to the LPF. This voltage Vp is a voltage drop between the surface electrode Hp and the surface electrode Lp due to the bioelectric impedance of the body B of the subject.

The LPF 82 removes high frequency noise from the voltage Vp and supplies it to the A / D converter 83. LPF
The cutoff frequency 82 is lower than half the sampling frequency of the A / D converter 83. Thereby, aliasing noise generated in the A / D conversion processing by the A / D converter 83 is removed. The A / D converter 83 converts the noise-removed voltage Vp into a digital signal at a predetermined sampling cycle every time the digital conversion signal Sd is supplied from the CPU 3, and converts the digitized voltage Vp to the sampling cycle. The data is supplied to the sampling memory 84 every time.

Next, as shown in FIG. 4, the surface electrode Lc is attached to the right instep L of the subject by the suction method. The surface electrode Lc and the coupling capacitor 90 (see FIG. 1) are connected by a coaxial cable (not shown), and the shield of the coaxial cable is grounded. The I / V converter 91 detects a current flowing between the two surface electrodes Hc and Lc and converts the current into a voltage. That is, I
When the probe current Ia is applied to the body B of the subject, the / V converter 91 detects the probe current Ia flowing between the right limbs of the subject, converts the probe current Ia to a voltage Vc, and supplies the voltage Vc to the LPF 92.

The LPF 92 removes high-frequency noise from the input voltage Vc and supplies it to the A / D converter 93.
The cutoff frequency of the LPF 92 is lower than half the sampling frequency of the A / D converter 93. In this case, A /
The aliasing noise generated in the A / D conversion processing by the D converter 93 is removed. The A / D converter 93 converts the noise-removed voltage Vc into a digital signal at a predetermined sampling cycle every time the digital conversion signal Sd is supplied from the CPU 3, and converts the digitized voltage Vc into a sampling cycle. The data is supplied to the sampling memory 94 every time. CP
U3 starts the measurement by the above-described measurement processing unit 2 according to the processing program stored in the ROM 6, and performs control to stop the measurement after sampling the detection voltages Vp and Vc a predetermined number of times at a predetermined sampling period. In addition, the following processing is performed.

First, the voltages Vp and Vc, which are functions of time, stored in the sampling memories 84 and 94 are sequentially read out and subjected to Fourier transform processing, respectively, and the voltages Vp (f) and Vc (f) (functions of frequency) are obtained. After converting to f), averaging is performed to calculate bioelectric impedance Z (f) [= Vp (f) / Vc (f)] for each frequency. Next, based on the obtained bioelectrical impedance Z (f) for each frequency, the CPU 3 obtains an impedance locus D as shown in FIG. 3 by making full use of curve fitting by a least squares calculation method.

FIG. 3B is a diagram showing a display example of an impedance locus D measured by the bioelectrical impedance measuring device according to the present embodiment. The spectrum intensity in the low frequency region of the frequency spectrum shown in FIG.
The relative difference generated between the spectrum intensity on the high frequency side and the spectrum intensity affects the error range of each plot shown in FIG. That is, on the frequency side higher than the reference frequency fo (critical frequency f _C ), the error range (likelihood) is reduced because each spectrum intensity is smaller than in the conventional case (see FIG. 6B).
Is large, but the effect of external noise after curve fitting is small because the plot is dense. On the other hand, on the frequency side lower than the reference frequency fo, the error range becomes smaller because the spectrum intensity is larger than in the conventional case, the curve fitting becomes more accurate, and the influence of external noise is suppressed.

Next, the CPU 3 calculates the bioelectric impedance R0 of the subject B at the frequency 0 and the bioelectric impedance R∞ at the infinite frequency of the subject B from the obtained impedance trajectory D. Then, the intracellular fluid resistance and the extracellular fluid resistance of the body B of the subject are calculated. In the section of the prior art, cells in a tissue of a human body are expressed by a simple electric equivalent circuit (FIG. 5: Re is extracellular fluid resistance, Rik is intracellular fluid resistance of each cell, and Cmk is cell membrane capacity of each cell). However, in an actual human body tissue, cells of various sizes are irregularly arranged. Therefore, an electric equivalent circuit close to the actual one has a time constant τ = Cmk · Rik as shown in FIG. Is represented by a distributed constant circuit in which elements connected in series with a capacitor and a resistor having the following are distributed. Therefore, in the present embodiment, since the intracellular fluid resistance and the extracellular fluid resistance are determined by using an electrical equivalent circuit that is close to actual, the impedance trajectory D of the human body is shown in FIG. As shown in FIG. 6 (B), the center is an arc that is higher than the real axis.

Next, based on the calculated intracellular fluid resistance and extracellular fluid resistance, and the human body characteristic data such as the height, weight, sex, and age of the subject input from the keyboard 1, a processing program is executed in advance. The body fat percentage, fat weight, lean body mass, intracellular fluid volume, extracellular fluid volume, and the sum of these amounts of body water (body fluid volume) ) Is calculated. Then, the calculated data is displayed on the display unit 4 including a display controller and a display (for example, an LCD).

Next, the operation of the bioelectrical impedance measuring apparatus 100 according to the present embodiment will be described. First, prior to the measurement, as shown in FIG.
c, Hp on the right back part H of the subject, two surface electrodes L
p and Lc are attached to the right instep L of the subject by the suction method. At this time, the surface electrodes Hc and Lc are attached to a portion farther from the center of the human body than the surface electrodes Hp and Lp. Next, the measurer (or the subject himself) uses the keyboard 1 of the bioelectrical impedance measuring device 100 to input the human body characteristic items such as the height, weight, sex, and age of the subject, and from the start of measurement to the end of measurement. The measurement time T, the measurement interval t, the reference frequency fo, and the like are set. The data and setting values input from the keyboard 1 are
Stored in the RAM 5. The reference frequency fo may be fixedly set in the ROM 6 or the like.

Next, when the measurer (or the subject himself) turns on the measurement start switch of the keyboard 1, the CPU 3
Sends a signal generation instruction signal to the measurement signal generator 72 of the measurement processing unit 2 after performing a predetermined initial setting. This initial setting includes a process of calculating the total number of samples in the above-described impedance locus, the sampling period of the A / D converters 83 and 93, the generation timing of the digital conversion signal Sd, and the like.

As a result, the measurement signal generator 72 repeatedly generates the probe current Ia a predetermined number of times, and outputs the probe current Ia as a measurement signal via the LPF 73, the coupling capacitor 74, and the measurement cable 10 which is a double shielded wire to the hand of the subject. Since the measurement signal Ia is transmitted to the surface electrode Hc (see FIG. 2) attached to the portion H, a measurement signal Ia of about 100 μA flows from the surface electrode Hc through the body B of the subject, and the first measurement is started.

When the measurement signal Ia is applied to the body B of the subject, the voltage Vp generated between the right limbs to which the surface electrodes Hp and Lp are attached in the differential amplifier 81 of the measurement processing unit 2
Is supplied to the A / D converter 83 via the LPF 82. On the other hand, in the I / V converter 91, the surface electrode H
After the probe current Ia flowing between the right limb to which c and Lc are attached is detected and converted to the voltage Vc, the LPF 9
The signal is supplied to the A / D converter 93 through the second line. At this time, C
The PU 3 outputs an A / D converter 8 every sampling period.
The digital conversion signal Sd is supplied to 3,93.

The A / D converter 83 converts the voltage Vp into a digital signal each time the digital conversion signal Sd is supplied, and supplies the digital signal to the sampling memory 84. The sampling memory 84 sequentially stores the digitized voltage Vp. On the other hand, the A / D converter 93 converts the voltage Vc into a digital signal each time the digital converted signal Sd is supplied, and supplies the digital signal to the sampling memory 94. The sampling memory 94 sequentially stores the digitized voltage Vc.

The CPU 3 calculates the probe current (measurement signal) I
When the number of repetitions of “a” reaches a preset number, after performing control to stop the measurement, first, the voltages Vp, which are stored in the sampling memories 84 and 94 and are a function of time, as a function of time.
The voltages Vp (f) and Vc (f), which are functions of frequency, are sequentially read out and subjected to Fourier transform processing.
(F is a frequency), and after averaging, bioelectric impedance Z (f) (= Vp (f) /
Vc (f)) is calculated.

Next, the CPU 3 performs curve fitting based on the calculated bioelectrical impedance Z (f) for each frequency by a calculation method of the least squares method, and obtains an impedance locus D as shown in FIG. From the obtained impedance locus D, the body B of the subject
Bioelectric impedance R0 at frequency 0 and bioelectric impedance R∞ at infinite frequency (corresponding to the X-axis coordinate value of the point where the arc of impedance locus D intersects the X-axis)
Is calculated, and the intracellular fluid resistance and the extracellular fluid resistance of the body B of the subject are calculated from the calculation result.

The CPU 3 executes a processing program in advance based on the calculated intracellular fluid resistance and extracellular fluid resistance, and human body characteristic data such as the height, weight, sex and age of the subject input from the keyboard 1. The body fat percentage, fat weight, lean body mass, intracellular fluid volume, extracellular fluid volume, and the sum of these body water ( Volume). Then, the calculated data are stored in the RAM 5 and displayed on the display unit 4. Next, the CPU 3 determines whether or not the total measurement time T has elapsed. If it is concluded that the measurement time T has elapsed, the CPU 3 terminates the subsequent measurement processing. If not, the CPU 3 determines the time corresponding to the measurement interval. After waiting for t to elapse, the same measurement processing is started again. Then, the above process is repeated until the entire measurement time T has elapsed.

As described above, according to the configuration of this example, the probe current Ia disperses in energy of about 1 msec despite including many frequency components.
Since the amplitude of the frequency spectrum uses an M-sequence signal that is almost flat over the entire frequency range, it does not damage the living body and removes the effects of respiration and pulse in the measurement of body fat status and body water distribution. It is possible,
Good measurement of the SN ratio is possible over the entire frequency range. Further, the measurement signal can be generated only from the shift register and the plurality of logic circuits, so that the configuration becomes very simple.

Since the bioelectrical impedance at the infinite frequency can be obtained by using the curve fitting method by the least squares method, the influence of stray capacitance and external noise can be avoided. The extracellular fluid resistance and the intracellular fluid resistance can be determined. In particular, the LPF 73 having an amplitude characteristic as illustrated in FIG.
Is provided, the spectrum intensity in the low frequency region of the measurement signal can be relatively increased compared to the spectrum intensity in the high frequency region, and as a result, the influence of external noise in the low frequency region can be suppressed. .

Although the embodiments of the present invention have been described in detail with reference to the drawings, the specific configuration is not limited to these embodiments, and a design within a range not departing from the gist of the present invention is provided. May be changed. For example, the above-described embodiment (FIG. 3A) is based on the conventional technology (FIG. 6).
(A), the total energy in the entire frequency band is equalized. In this case, the reference frequency f
o By emphasizing the spectrum at lower frequencies and attenuating the spectrum at higher frequencies, the influence of external noise in the low frequency region (see FIG. 3B) can be suppressed.

On the other hand, in the amplitude characteristic illustrated in FIG.
The gain from the frequency 0 to the critical frequency f _C may be set to 1 to maintain the spectral intensity at a frequency lower than the critical frequency f _C and attenuate only the spectral intensity at the higher frequency. As a result, it is possible to suppress the influence of external noise in the low frequency region and save power. In any case, any configuration may be used as long as the spectrum intensity in the low frequency region is relatively larger than that in the high frequency region with respect to the reference frequency fo.

The measurement signal is not limited to the M-sequence signal, but may be an impulse signal, a chirp signal, or a signal obtained by temporally changing the frequency of a sine wave. Further, in the above embodiment, the current value applied to the subject via the surface electrode Hc and the surface electrode Lc is measured, and between the predetermined two surface portions via the surface electrode Hp and the surface electrode Lp. The resulting voltage value is measured.
As another embodiment, for example, the current value and the voltage value may be measured by the surface electrode Hc and the surface electrode Lc (so-called two-terminal method), and the number of the surface electrodes limits the present invention. Not something.

The calculated bioelectric parameters are not limited to bioelectric impedance, impedance locus, extracellular fluid resistance and intracellular fluid resistance, but also bioelectric admittance, admittance locus, the bioelectric impedance or bioelectric admittance, extracellular The amount of temporal change such as fluid resistance and intracellular fluid resistance and a part thereof may be used. In this case, not only measurement of body fat percentage and the like but also various medical systems (for example, dialysis state measurement) ) Can be expected. Also, the location where the electrode is attached is not limited to the hand or foot.

The Fourier transform processing by the CPU 3 is not limited to the present invention, and may be any method that uses an arithmetic technique that converts a function represented in a time domain into a function represented in a frequency domain. Furthermore, in the above-described embodiment, the case where the height, weight, gender, and age of the subject are input as the human body characteristic items has been described, but gender, age, and the like may be omitted as necessary, or Items such as race may be added. The calculated bioelectric parameters of the human body may be output to a printer. Further, a pulse wave sensor or a sensor capable of detecting a respiratory cycle may be attached to a human body, and the measurement timing may be set based on the output signal of each sensor.

[0054]

As described above, according to the present invention,
In consideration of the critical frequency unique to the subject, the spectrum at frequencies lower than the reference frequency is emphasized, and the spectrum at higher frequencies is attenuated, so that the influence of external noise in the low frequency region can be suppressed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an electrical configuration of a bioelectrical impedance measuring device according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating amplitude characteristics of an LPF according to the embodiment of the present invention.

FIG. 3A is a diagram exemplifying a spectrum intensity of each frequency of a measurement signal output by the LPF according to the present embodiment, and FIG. 3B is a diagram illustrating a living body according to the present embodiment; It is a figure showing the example of a display of impedance locus D measured by an electric impedance measuring device.

FIG. 4 is a diagram schematically showing a state of use of the bioelectrical impedance measuring device according to the embodiment of the present invention.

FIG. 5 is an electrical equivalent circuit diagram (equivalent circuit model) of a human body.

FIG. 6A is a diagram illustrating a frequency spectrum of a measurement signal (M-sequence signal) input to the body of a subject in a conventional bioelectric impedance measuring apparatus.
(B) is a diagram illustrating an impedance locus of bioelectric impedance calculated for each frequency.

FIG. 7 is an electrical equivalent circuit diagram of a human body which is close to the actual one.

[Explanation of symbols]

Reference Signs List 1 keyboard 2 measurement processing unit 3 CPU (calculation means) 4 display unit 5 RAM 6 ROM 10 measurement cable 72 measurement signal generator (part of measurement signal supply means) 73 LPF (part of measurement signal supply means) 81 difference Dynamic amplifier (part of voltage measuring means) 82 LPF (part of voltage measuring means) 83 A / D converter (part of voltage measuring means) 84,94 Sampling memory 91 I / V converter (of current measuring means) 92 LPF (part of current measuring means) 93 A / D converter (part of current measuring device) 100 Bioelectric impedance measuring device (electric property measuring device) Hc surface electrode (first electrode) Lc surface electrode (Second electrode) Hp surface electrode Lp surface electrode

Claims (7)

[Claims] 1. A method according to claim 1, wherein a signal having a spectrum intensity on a low frequency side which is relatively larger than a spectrum intensity on a high frequency side is generated in accordance with a frequency of a measurement signal, and two predetermined surface portions of the subject separated from each other. Measurement signal supply means for applying the signal to the subject via the first and second electrodes conductively attached to the; and current measurement means for measuring a current value of the signal applied to the subject, Voltage measuring means for measuring a voltage value generated between two predetermined surface portions separated from each other of the subject, and the current value and the voltage value measured by the current measuring means and the voltage measuring means, respectively, Calculating means for calculating a bioelectric impedance between surface portions and calculating a bioelectric impedance to be obtained or a physical quantity based on the bioelectric impedance. An electrical characteristic measuring device characterized by the above-mentioned. 2. The electrical characteristic measuring apparatus according to claim 1, wherein said measurement signal supply means has a filter means for generating said signal in accordance with a frequency of said measurement signal. 3. The electric characteristic measuring apparatus according to claim 2, wherein the absolute value of the transfer function of the filter means is a constant in a low frequency region and decreases in inverse proportion to the frequency in a high frequency region. 4. An absolute value of a transfer function of the filter means is (1 / (1+ (f / fo) ² )) ^1/2 (f: frequency,
3. An electric characteristic measuring apparatus according to claim 2, wherein the electric characteristic is proportional to (fo: reference frequency). 5. The apparatus according to claim 4, wherein said filter means is realized by a primary CR circuit or an LR circuit.
The electrical property measurement device according to the above. 6. The reference frequency is 10 kHz to 300.
The electric characteristic measuring apparatus according to claim 4, wherein the electric characteristic is set in a range of kHz. 7. The electric signal according to claim 1, wherein the measurement signal is a longest linear code sequence, an impulse signal, a chirp signal, or a sine wave whose frequency changes stepwise at equal intervals in time. Characteristic measuring device.