JP2002224059A - Electronic sphygmomanometer - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately measure blood pressure at walking, at moving, and at the other body movement without particularly complicating the device constitution. SOLUTION: Taking-in of a photoelectric pulse wave from a photoelectric sensor 6 and taking-in of a signal (noise caused by a body movement, etc.), from an acceleration sensor 5 are started in a pressure reducing process after pressurizing a cuff (ST51 and ST52). When obtaining data capable of high speed Fourier transform (ST60), the high speed Fourier transform is executed on the photoelectric pulse wave and an acceleration signal to be converted into frequency series data from time series data (ST61 and ST62). The noise is removed by removing the whole except for a frequency component of the pulse wave of the frequency series data on the photoelectric pulse wave by using both data (ST63). The photoelectric pulse wave after removing the noise is restored to the time series data by inverse Fourier transform (ST64). The photoelectric pulse wave after restoration is used for calculating a blood pressure value (ST67).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic sphygmomanometer which can be identified even during body movement such as walking or exercising.

[0002]

2. Description of the Related Art Sphygmomanometers which are hardly affected by noise and artifacts caused by body movements and the like, and which are conventionally developed for measuring blood pressure during body movements, include a sphygmomanometer for an exercise load test, There is a time portable sphygmomanometer and the like. These sphygmomanometers measure blood pressure using the Korotkoff sound method, the oscillometric method for measuring pulse waves, or a combination of both methods.

However, in such a conventional sphygmomanometer, in the oscillometric method, pressure noise due to body motion or the like largely overlaps with a pressure pulse wave. In the Korotkoff sound method, there is a possibility that noise caused by body movement and the like may be mistaken for Korotkoff sound. Therefore, the conventional method has a problem in that the detection accuracy of the pulse wave and the Korotkoff sound decreases, and it becomes difficult to determine the blood pressure value, and the accuracy of the blood pressure value decreases.

Therefore, the following measures have been taken to solve this problem. First, in the oscillometric sphygmomanometer, the cuff pressure pulse wave is monitored by a predetermined logic,
The presence / absence of noise due to body motion or the like is determined, and when it is determined that there is noise, a measurement error occurs. Alternatively, information indicating that noise has occurred during measurement is added.

On the other hand, in the Korotkoff sound type blood pressure monitor, an electrocardiographic electrode is attached in addition to the Korotkoff sound sensor, and the Korotkoff sound is used as a trigger signal in order to increase the resistance to noise caused by body movement and the like. The blood pressure value is determined using only the signal of the Korotkoff sound sensor in only a short section (Korotkoff sound evaluation period) in which is likely to appear. With this method, it is possible to prevent the Korotkoff sound sensor output from being erroneously recognized as Korotkoff sound due to noise caused by body motion etc. during the period when the Korotkoff sound is unlikely to appear, so it is effective when exercise intensity is relatively weak. It is.

There are the following techniques for reducing noise caused by body movement and the like. There are various noise sources such as noises caused by body movements, such as those that transmit air from the outside and directly enter the sensor, those that transmit inside the body, and those that transmit from a cuff holding the sensor. Therefore, in order to reduce noise for each individual factor, a large number of sensors are prepared for each noise factor and signal processing is performed.

[0007]

The body movement countermeasure of the conventional oscillometric sphygmomanometer described above is a method of eliminating or invalidating the blood pressure measurement during the body movement, and makes it possible to measure the blood pressure during the body movement. It was not something.

In the conventional Korotkoff sphygmomanometer body movement countermeasures, in a situation where noise due to body movement or the like is continuously generated, for example, when exercise intensity is high, the cuff pressure is high and Korotkoff sounds originally appear. Even during the non-periodic period, noise caused by body movement etc. is included during the Korotkoff sound evaluation period with the electrocardiogram signal as a trigger signal, and the Korotkoff sound appearance time, which is an important element in determining systolic blood pressure and diastolic blood pressure, disappearance It is difficult to determine the timing, and it is difficult to determine the blood pressure value.

Further, in the method of preparing a large number of sensors for each of the factors described above and reducing noise caused by body movements, etc., this complicates the device configuration and signal processing, and is difficult to achieve with reasonable price setting. . Therefore, a method is often adopted in which the difference between the frequency components of noise and Korotkoff sound is used to collectively reduce the noise with a filter. However, of the noise, noise that overlaps the Korotkoff sound with the frequency domain cannot be completely removed.

[0010] The present invention has been made in view of the above problems, and without making the device configuration particularly complicated.
It is an object of the present invention to provide an electronic sphygmomanometer that can accurately measure during walking, exercising, and other body movements.

[0011]

An electronic sphygmomanometer according to the present invention comprises at least one cuff, pressure control means for increasing and decreasing the pressure in the cuff, and pressure detection means for detecting the pressure in the cuff. A means for detecting noise due to body movement, a pulse wave detecting means for detecting a pulse wave, and a blood pressure calculating means for calculating a blood pressure based on the detected pressure in the cuff and the pulse wave. Means for frequency-analyzing the noise and converting the time-series data into frequency-series data; means for comparing the pulse-wave and noise frequency-series data to determine the frequency of the pulse wave; Means for removing all but the frequency components of the wave, and means for converting the frequency sequence data of the pulse wave after removing all but the frequency component of the pulse wave to time-series data, and restoring the pulse wave, Calculate blood pressure based on pulse wave It is way.

In the present invention, a cuff pressure sensor, a photoelectric sensor, an impedance sensor, a strain sensor, or the like may be used as the pulse wave detecting means. In the case of a sensor that converts a movement of a living body into a physical quantity and measures it as a means for detecting a noise signal due to body movement or the like, a speed sensor, an acceleration sensor, a position sensor, a displacement sensor, an angle sensor, a direction sensor, an inclination sensor, When measuring the amount of a living body (for example, the amount of blood) that changes with the movement of a living body using a sensor or the like, a photoelectric sensor or the like may be used.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in more detail with reference to embodiments. FIG. 1 is a block diagram showing a configuration of an electronic sphygmomanometer according to one embodiment of the present invention.
The electronic sphygmomanometer according to the present embodiment is one in which the present invention is applied to an electronic sphygmomanometer that employs a volume oscillation method using a photoelectric pulse wave.

An electronic sphygmomanometer according to this embodiment includes a cuff 1 for pressurizing a measurement site of a living body, an A / D converter 2 for converting an analog signal from an amplifier into a digital signal and inputting the digital signal to a CPU, and a pressure pump. A CPU 3 for controlling a quick exhaust valve, a display, etc., a display 4 for displaying a calculated blood pressure value and the like, an acceleration sensor 5 provided on the cuff 1 and serving as a noise signal detecting means caused by body movement and the like. A photoelectric sensor 6 for detecting a photoelectric pulse wave, a pressure sensor 7 for detecting a pressure in the cuff 1, a pressurizing pump 8 for pressurizing the cuff, and a quick exhaust valve 9 for rapidly exhausting air in the cuff 1. , Cuff 1
Low-speed exhaust valve 10 for reducing the internal pressure of the vessel by blood pressure measurement
An acceleration signal amplifier 11 for amplifying the output from the acceleration sensor 5; a photoelectric signal amplifier 12 for amplifying the output from the photoelectric sensor 6; a pressure signal amplifier 13 for amplifying the output from the pressure sensor 7; , A valve opening / closing circuit 15 for opening / closing the quick exhaust valve 9, a power switch 16 for turning on / off the power of the sphygmomanometer, and a pressurizing switch for starting pressurizing the cuff 1. 17 is provided.

The noise removing means of this electronic sphygmomanometer compares the frequency analysis result of the photoelectric pulse wave signal obtained by the photoelectric sensor 6 with the frequency analysis result of the noise signal obtained by the acceleration sensor 5 so that the body movement is reduced. And so on. Noise caused by body movement or the like superimposed on the cuff pressure of the cuff 1 is removed using a digital filter (a function of the CPU 3). The photoplethysmographic amplitude is obtained by calculating the amplitude for each fixed section to remove noise.

The blood pressure determining means applies a predetermined algorithm to the photoelectric pulse wave signal and the cuff pressure signal taken in the CPU 3 to determine the systolic blood pressure and the diastolic blood pressure. For this algorithm, you can use a well-known algorithm.
It is not unique to the present invention.

Next, the operation of blood pressure measurement including noise removal of the electronic blood pressure monitor of the above embodiment will be described with reference to flowcharts shown in FIGS. 2, 3 and 4.

When the power switch 16 is pressed (step ST10), the RAM, ports and the like in the CPU 3 are initialized (step ST11). Next, taking in of a pressure signal through the pressure sensor 7, the pressure signal amplifier 13, and the A / D converter 2 is started (step ST12).

Next, the process proceeds to step ST20, where it is determined whether or not the power switch 16 is turned on (pressed). If the power switch 16 is not pressed, the pressure switch 17
Is determined to be ON (step ST21). If the power switch 16 is pressed in step ST20 (the power supply OF
F), the pressure capture is stopped (step ST30), and the operation ends. If the pressure switch 17 is not turned on in step ST21, the process returns to step ST20, and steps ST20 and ST are performed until either the power switch 16 or the pressure switch 17 is turned on (pressed).
Stand by at 21.

When the press switch 17 is pressed, the quick exhaust valve 9 is closed (step ST40), the pressurizing pump 8 is started, and pressurization of the cuff 1 is started (step ST41). CP
The valve opening / closing signal from U3 is signal-converted by the valve opening / closing circuit 15 and transmitted to the quick exhaust valve 9. Similarly, a pump drive signal from the CPU 3 is also converted by the pump drive circuit 14 and transmitted to the pressurizing pump 8. Next, step ST4
The process proceeds to 2 to monitor whether or not the cuff pressure has reached the pressurization set pressure.

When the cuff pressure reaches the pressurization set value, the pressurizing pump 8 is stopped to start the slow exhaust process (step ST5).
0), the capture of the photoelectric pulse wave signal from the photoelectric sensor 6 and the capture of the acceleration signal from the acceleration sensor 5 are started (steps ST51 and ST52). The photoelectric pulse wave signal output from the photoelectric sensor 6 is amplified by the photoelectric signal amplifier 12, digitized by the A / D converter converter 2, and
Captured by U3. The acceleration signal output from the acceleration sensor 5 is also amplified by the acceleration signal amplifier 11, digitized by the A / D converter 2, and captured by the CPU 3.

Photoplethysmographic data and acceleration data are collected until a fast Fourier transform becomes possible (step ST6).
0). When data necessary for the fast Fourier transform is collected, the collected photoplethysmographic data is subjected to the fast Fourier transform to be converted into frequency sequence data (step ST61). Further, the collected acceleration data is subjected to a fast Fourier transform to be converted into frequency series data (step ST62). Next, a noise component caused by body movement is removed from the frequency-sequence photoelectric pulse wave data using the frequency-sequence acceleration data as an index (step ST63). The frequency-sequence photoelectric pulse wave data from which noise has been removed is subjected to inverse fast Fourier transform to return to time-series data (step ST64).

Next, the cuff pressure data on which the noise is superimposed is input to the digital filter, and the cuff pressure base line data is extracted (step ST65). Subsequently, the time-series photoplethysmographic data in which noise is detected is divided into fixed time sections, and the photoplethysmographic wave amplitude in each section is calculated (step S).
T66). Then, the blood pressure value is calculated from the photoelectric pulse wave amplitude and the cuff pressure by a predetermined algorithm (step ST6).
7).

When the calculation of the blood pressure value is completed, the quick exhaust valve 9
To quickly reduce the pressure of the cuff 1 (step S
At the same time as T70), the capture of the photoelectric pulse wave signal is stopped (step ST71). Further, the acquisition of the acceleration signal is stopped (step ST72). The blood pressure value is displayed on the display 4 (step ST73), and waits until the cuff pressure reaches the atmospheric pressure (step ST74). When the pressure reaches the atmospheric pressure, the process proceeds to step ST20, that is, the power switch 16 and the pressure switch 1
It returns to the state of waiting for 7 pushes.

Next, a method for removing noise caused by body motion superimposed on a photoelectric pulse wave will be described. The measured photoelectric pulse wave signal and acceleration signal are taken into the CPU 3 as time-series data. The photoelectric pulse wave time series data and the acceleration time series data are subjected to frequency analysis and converted into frequency series data. As a means for frequency analysis, for example, a method using a fast Fourier transform, an autoregressive model, or the like is used.

FIG. 5 shows time series data and frequency series data (power spectrum) of the photoelectric pulse wave before noise removal, and FIG. 6 shows frequency series data and time series data after noise removal. In FIG. 5, Fourier transform is performed from time series data to frequency series data, and in FIG. 6, inverse Fourier transform is performed from frequency series data to time series data.

When there is no noise due to body motion or the like, only the frequency component of the photoelectric pulse wave exists in the power spectrum of the photoelectric pulse wave, and no characteristic spectrum exists in the power spectrum of the acceleration signal. If there is noise due to body motion, etc., both the frequency component of the photoelectric pulse wave and the frequency component of the body motion are present in the power spectrum of the photoplethysmogram, and the body spectrum is included in the power spectrum of the acceleration signal. There exists a spectrum of the frequency component of Therefore, the frequency component of the photoelectric pulse wave is extracted by comparing the two.

The frequency series data from which only the frequency component of the pulse wave is extracted is converted into time series data. At this time, an inverse fast Fourier transform or the like is used. The data converted into the time-series data in this way is photoelectric pulse-wave time-series data in which only the photoelectric pulse wave component exists and noise due to body motion or the like has been removed.

Further, for both the photoplethysmogram and the noise, noise removal is performed by utilizing the fact that the harmonic frequency exists at a position of an integral multiple of the fundamental frequency.

Here, the frequency component of the photoelectric pulse wave is extracted,
When the waveform is reproduced by removing the frequency band other than the frequency range, it is determined whether or not the original waveform can be faithfully reproduced depending on how many harmonic components are reproduced. However, in determining the blood pressure, it is possible to determine the blood pressure value only by the amplitude information of the original waveform, and if a result proportional to the amplitude is obtained, the blood pressure value can be determined even if the waveform information is somewhat lost. Has no effect. Therefore, when reproducing the waveform, even if the waveform other than the fundamental frequency of the frequency component of the photoelectric pulse wave is removed and the waveform is reproduced, the accuracy of the final blood pressure value is not affected. However, here, an example is shown in which the photoplethysmogram includes all the fundamental waves and harmonics.

The noise removal of the cuff pressure is performed by inputting the cuff pressure data to a finite impulse response low-pass filter and extracting the cuff pressure baseline data. FIG. 7 shows an example of cuff pressure data before noise removal and cuff pressure baseline data after noise removal.

Another method is to divide the cuff pressure data into fixed time sections, interpolate the cuff pressure data within the section, and use the arrangement of the interpolated data as cuff pressure baseline data. A method of averaging the cuff pressure data within the section and taking the arrangement of the average values as the cuff pressure baseline data, and extracting the cuff pressure data after the noise removal using frequency analysis in the same manner as the noise removal of the photoelectric pulse wave. There are methods.

The noise removal of the photoplethysmogram is performed by taking out the photoplethysmogram from which the noise has been removed for each predetermined time interval, and calculating the difference between the maximum value and the minimum value of the photoplethysmogram in the predetermined time interval. This is performed by setting the amplitude and further performing a moving average. FIG. 8 shows an example of a photoelectric pulse wave after noise removal, a photoelectric pulse wave amplitude envelope, and a moving averaged photoelectric pulse wave amplitude envelope.

In the present embodiment, the means for detecting a noise signal caused by body movement or the like is an acceleration sensor. It may be a sensor that converts and measures, or a photoelectric sensor that measures a biomass (for example, blood volume) that changes by movement.

Further, even if there is no means for detecting a noise signal,
By recognizing the appearance pattern of the pulse wave, it is also possible to remove noise. For example, if there is a fundamental frequency of a pulse wave on the frequency axis, the fundamental frequency and the second harmonic of noise due to body motion and the like exist on both sides of the fundamental wave (see FIG. 9).
Noise can be removed by utilizing this relationship and the relationship in which the harmonic frequency exists at a position that is an integral multiple of the fundamental frequency for both the photoelectric pulse wave and the noise.

Further, in this embodiment, the pulse wave signal for determining the blood pressure is a photoelectric pulse wave signal extracted from the photoelectric sensor.
The pulse wave signal may be detected using a pressure sensor, an impedance sensor, or a strain sensor.

[0037]

According to the present invention, pulse wave and noise are frequency-analyzed, and means for converting time series data to frequency series data are compared with pulse wave and noise frequency series data.
Means for determining the frequency of the pulse wave, means for removing all but the frequency component of the pulse wave from the frequency sequence data of the pulse wave, and time-series data for the frequency sequence data of the pulse wave after removing all but the frequency component of the pulse wave And a means for restoring the pulse wave, and the blood pressure is calculated based on the restored pulse wave, so that the blood pressure can be determined with high accuracy even under the condition that noise such as body motion is mixed. Further, noise due to the internal condition of the subject such as respiratory noise can be removed, and blood pressure can be determined with high accuracy.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an electronic sphygmomanometer according to an embodiment of the present invention.

FIG. 2 is a flowchart illustrating a processing operation of the electronic sphygmomanometer according to the embodiment.

FIG. 3 is a flowchart for explaining a processing operation of the electronic sphygmomanometer according to the embodiment, together with FIG. 2;

FIG. 4 is a flowchart for explaining a processing operation of the electronic sphygmomanometer according to the embodiment, together with FIGS. 2 and 3;

FIG. 5 is a diagram showing an example of conversion from a time-series waveform of a photoplethysmogram to a frequency-series waveform before noise removal in the electronic blood pressure monitor of the embodiment.

FIG. 6 is a diagram showing an example of conversion from a frequency series (power spectrum) waveform of a photoplethysmogram after noise removal to a time series waveform in the electronic blood pressure monitor of the embodiment.

FIG. 7 is a diagram for explaining noise removal of cuff pressure in the electronic blood pressure monitor of the embodiment.

FIG. 8 is a diagram illustrating a case where a pulse wave amplitude envelope is obtained in the electronic blood pressure monitor of the embodiment.

FIG. 9 is a diagram illustrating a relationship between a pulse wave fundamental wave, a noise fundamental wave, and a noise second harmonic.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Cuff 2 A / D converter 3 CPU 4 Display 5 Acceleration sensor 6 Photoelectric sensor 7 Pressure sensor 8 Pressurizing pump 9 Quick exhaust valve 10 Slow exhaust valve 11 Acceleration signal amplifier 12 Photoelectric signal amplifier 13 Pressure signal amplifier 14 Pump drive circuit 15 Valve opening / closing circuit 16 Power switch 17 Pressure switch

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Yukiya Sawanoi Shimogyo-ku, Shimogyo-ku, Kyoto 801 Minami-Fudo-do, Higashikawa-Higashi-ri, Omron Life Science Research Institute, Inc. 801 Horikawa Higashiiri Minami Fudodoucho OMRON Life Science Laboratories, Inc. F-term (reference) 4C017 AA08 AA09 AC03 AC05 AC26 AD08 BC11 BC14 BC18 CC01 DE06 FF05

Claims (9)

[Claims] 1. A cuff, a pressure control means for increasing and decreasing the pressure in the cuff, a pressure detection means for detecting a pressure in the cuff, and detecting noise caused by at least one or more body movements. Means, a pulse wave detecting means for detecting a pulse wave, and a blood pressure calculating means for calculating a blood pressure based on the detected pressure in the cuff and the pulse wave. Means for converting data to frequency sequence data, comparing the pulse wave and noise frequency sequence data, means for determining the frequency of the pulse wave,
Means for removing all but the pulse wave frequency component from the pulse wave frequency sequence data, and converting the pulse wave frequency sequence data after removing all but the pulse wave frequency component to time series data to restore the pulse wave Means for calculating blood pressure based on the restored pulse wave. 2. The electronic sphygmomanometer according to claim 1, wherein the cuff pressure is obtained by removing noise by filtering, extracting cuff pressure baseline data from which noise has been removed, and calculating blood pressure. 3. The cuff pressure is obtained by dividing a cuff pressure on which noise is superimposed into fixed time sections, performing interpolation processing for each section, arranging the data after the interpolation processing, and forming cuff pressure baseline data from which noise has been removed. The electronic sphygmomanometer according to claim 1, wherein the blood pressure is calculated. 4. The cuff pressure is obtained by dividing a cuff pressure on which noise is superimposed into fixed time intervals, obtaining an average value for each interval, arranging the data of the average value, and forming cuff pressure baseline data from which noise has been removed. The electronic sphygmomanometer according to claim 1, wherein is calculated. 5. A cuff pressure and a noise are frequency-analyzed, time-series data is converted into frequency-sequence data, the frequency-sequence data is compared, a frequency component of the cuff pressure is obtained, and the cuff-pressure frequency sequence data is Removes all but the cuff pressure frequency component, converts the cuff pressure frequency sequence data after removal of all but the cuff pressure frequency component to time-series data, restores the cuff pressure, and uses the restored cuff pressure as a basis. The electronic sphygmomanometer according to claim 1, wherein the blood pressure is calculated. 6. A pulse wave from which noise has been removed is separated at predetermined time intervals, and a representative pulse wave amplitude is calculated in the interval to realize noise removal of the pulse wave amplitude and calculate a blood pressure. The electronic sphygmomanometer according to claim 1, wherein: 7. The electronic device according to claim 1, wherein the blood pressure is calculated by removing the noise of the pulse wave by utilizing that the harmonic frequency of the pulse wave and the noise is an integral multiple of the fundamental frequency. Sphygmomanometer. 8. A method of calculating only a fundamental frequency from frequency components of a pulse wave, removing all but the fundamental frequency component of the pulse wave from the frequency sequence data of the pulse wave, and removing all but the frequency component of the pulse wave. 2. The electronic sphygmomanometer according to claim 1, wherein frequency series data of the pulse wave is converted into time series data, the pulse wave is restored, and the blood pressure is calculated based on the restored pulse wave. 9. A cuff, pressure control means for increasing and decreasing pressure in the cuff, pressure detection means for detecting pressure in the cuff, pulse wave detection means for detecting a pulse wave, and blood pressure calculation means. An electronic sphygmomanometer comprising: a means for frequency-analyzing a pulse wave and converting time-series data to frequency-series data; a means for identifying an appearance pattern of the frequency-series data to obtain a frequency of the pulse wave; and a frequency of the pulse wave. Means for removing all but the frequency component of the pulse wave from the sequence data, and means for converting the frequency sequence data of the pulse wave after removing all but the frequency component of the pulse wave to time-series data and restoring the pulse wave An electronic sphygmomanometer which calculates a blood pressure based on a restored pulse wave.