KR20120108575A - Apparatus and method improving accuracy of wrist blood pressure by using multiple bio-signal - Google Patents

Abstract

PURPOSE: A method and device for improving the accuracy of wrist blood pressure by measuring a multiple bio-signal are provided to improve the accuracy of measuring the wrist blood pressure by simultaneously measuring and analyzing representative factors such as BMI or Korotkoff sound effected on the blood pressure. CONSTITUTION: A cuff applies pressure when the cuff is wound around a wrist. An oscillation signal detector(105) includes a pressure sensor connected to the cuff. The oscillation signal detector detects a pressure signal of the cuff and an oscillation signal generated by the pressure variation of the pressure signal. A Korotkoff sound detector(205) detects a Korotkoff sound signal by connecting a microphone to the connection part between the cuff and the pressure sensor. A calculation processing unit(630) obtains oscillometric blood pressure from the pressure signal and the oscillation signal. The calculation processing unit corrects the oscillometric blood pressure using the Korotkoff sound signal. [Reference numerals] (100) Pressure sensor; (110) Pressure signal preprocessing unit; (200) Microphone; (210) Korotkoff sound preprocessing unit; (220) Pre-amplifier; (300) ECG electrode unit; (310) ECG preprocessing unit; (320) Differential amplifier; (500) PPG sensor; (510) PPG preprocessing unit; (520) Current voltage converter; (600) Data collecting unit; (630) Calculation processing unit; (650) Output unit; (670) Key input unit; (690) Memory

Description

Apparatus and method improving accuracy of wrist blood pressure by using multiple bio-signal}

The present invention relates to a method and apparatus for improving the accuracy of wrist blood pressure measurement in an E-health terminal using multiple bio-signal measurement, and more particularly, to the oscillation component of blood pressure in order to improve the convenience and accuracy of measurement. Simultaneous measurement and analysis of representative factors affecting blood pressure (PTT, Korotkoff Sound, BMI, etc.) in the E-health terminal using multi-body signal measurement, which improves the accuracy of blood pressure measurement on the wrist. A method and apparatus for improving the accuracy of wrist blood pressure measurement.

 Conventional blood pressure measurement has been mainly focused on hospital patients, but the importance of periodic blood pressure measurement at home is being emphasized as the general social interest in health and medical care and the increase in hypertension patients.

 The commonly used sphygmomanometer is an oscillometric electronic sphygmomanometer, and the electronic blood pressure detection system generally uses an oscillometric method. This method uses a CUFF to apply pressure higher than the systolic blood pressure to the upper arm, wrist, and finger of the subject to occlude the artery, and then detect the width of oscillation, which is the vibration characteristic of the cuff while decompressing. The blood pressure value is calculated by taking a certain characteristic ratio.

In other words, the oscillometric blood pressure measurement method does not measure blood pressure directly, but assumes the average arterial pressure at the maximum vibration width generated by the cuff, and empirically determines the characteristic ratio having a constant ratio with the vibration width. Because it is calculated by the systolic and diastolic blood pressure, there are various factors of error occurrence.

In particular, in the determination of the maximum oscillation width, the maximum point is often represented as a flat plate instead of one point. Therefore, the definition of the maximum point is unclear, and the determination of the blood pressure by the maximum amplitude algorithm uses only one factor of blood pressure oscillation to detect the blood pressure. Therefore, not only the average arterial pressure, but also the various factors that affect the blood pressure, such as the anatomy of the artery, the shape of the blood pressure, the pressure of the pulse, the characteristics of the arm and cuff, the size of the cuff, the heart rate, etc. Errors, especially for wrist oscillometric electronic sphygmomanometers, which have poor performance below the standards of the British Hypertension Society (BHS) or the Association for the Advancement of Medical Instrumentation (AAMI). Mostly. That is, the oscillometric blood pressure measurement method has a relatively high error rate depending on the vasculature characteristics such as cuff size and artery stiffness.

Electronic blood pressure measurements that do not use an oscillometric method include the Cortkov method, the pulse wave delivery rate method, and the Finapress method.

The Kortgof method uses a microphone to convert a Cortkov sound into an electrical signal and analyzes it. A large number of devices developed in the past detected blood pressure by analyzing simple amplitude changes. . This method has problems with ambient noise, microphone position, and weak sound in hypertensive patients, and requires a new approach to spectral analysis of the Kortkov sound.

The pulse wave delivery rate method is a method based on the increase of the pulse wave rate with the increase of arterial pressure, but it is possible to measure the blood pressure continuously, but there is a limit to the correction because it defines and approaches the vasculature as a single factor of time or velocity. The device for specifying blood pressure continuously and noninvasively by a dedicated finger cuff, and the pinapress method also has a considerable error. In other words, electronic sphygmomanometers that do not use the oscillometric method also detect blood pressure by measuring a single factor, and estimate the change in blood pressure using only one factor without considering various factors affecting the blood pressure. It is a state to do.

Therefore, there is a need for a blood pressure measuring apparatus and method for improving measurement accuracy while reducing errors caused by blood pressure detection by measuring a single element.

To this end, the present invention provides an electrocardiogram (ECG), an oxygen saturation (PPG), a body impedance (Body) for consideration of factors representing the state of the cardiovascular system in order to reduce the error caused by the detection of blood pressure by measuring a single element. Implement the four hardware blocks of Impedance and NiBP and implement it as a system to provide a measurement system for multiple biosignals, and also measure and analyze pulse transit time (PTT) ( Factors such as pulse wave velocity (PWV), Korotkoff Sound Power Spectrum Density, Oscillometric Blood Pressure, body mass index (BMI), etc. In addition, by applying a multi-linear regression equation to the individual body information to improve the accuracy of blood pressure measurement on the wrist by measuring the convenience of the measurer Derived, and improved the accuracy of the measurement.

The problem to be solved by the present invention is to measure the blood pressure by mounting a cuff on the wrist for convenience of measurement, to the representative factors (PTT, Korotkoff Sound, BMI, etc.) affecting the blood pressure with the oscillation component of the blood pressure The present invention provides a method and apparatus for improving the accuracy of wrist blood pressure measurement using multiple bio-signal measurements, which results in improved accuracy in blood pressure measurement on the wrist by performing simultaneous measurements and analyzing them.

Another problem to be solved by the present invention, ECG, oxygen saturation (PPG) for consideration of factors representing the state of the cardiovascular system in order to reduce the error caused by the detection of blood pressure by measuring a single element ), Four hardware blocks of body impedance and blood pressure (NiBP) are integrated and implemented as one system to provide a measurement system for multiple bio signals.

Another object of the present invention is to detect factors such as pulse wave propagation time (PTT) (pulse wave velocity (PWV)), Kortokov power spectrum density, oscillometric blood pressure, body mass index (BMI), The present invention also provides a method and apparatus for improving the accuracy of blood pressure measurement on the wrist by calculating a multiple linear regression equation for systolic blood pressure, diastolic blood pressure, and mean blood pressure to which individual body information is applied.

The wrist blood pressure measuring device according to the present invention comprises a cuff made to wind the wrist and configured to apply pressure; An oscillation signal detection unit having a pressure sensor connected to the cuff, the oscillation signal detecting unit detecting a pressure signal of the cuff and an oscillation signal which is a vibration characteristic generated according to a pressure change in the pressure signal; Cortkop sound detection unit for connecting the microphone to the portion where the cuff and the pressure sensor is connected, and detects the Cortkop sound signal; Cort received by A / D conversion from the output of the oscillation signal detector to obtain the oscillometric blood pressure from the received oscillation signal and the oscillometric blood pressure from the output of the Cortkov sound detector. It is characterized in that it comprises a ;; processing unit for correcting by using the sound signal.

In addition, the wrist blood pressure measurement device of the present invention is made to wind the wrist cuff made to apply pressure; An oscillation signal detection unit having a pressure sensor connected to the cuff, the oscillation signal detecting unit detecting a pressure signal of the cuff and an oscillation signal which is a vibration characteristic generated according to a pressure change in the pressure signal; A memory unit for storing age, gender, height, arm circumference, and weight as information of a subject; The body mass index (BMI) is obtained by dividing the subject's height by weight, and A / D conversion is performed from the output of the oscillation signal detection unit to obtain an oscillometric blood pressure from the input pressure signal and the oscillation signal, and the oscillometric blood pressure is measured by the body mass index (BMI). It is characterized by comprising; arithmetic processing unit to correct using.

The present invention includes an electrocardiogram detector including two electrocardiogram electrodes and mounted on both wrists or mounted on both hands to detect an electrocardiogram; A PPG detector having an LED and an optical sensor positioned to be mounted on a wrist or a hand to detect an optical volume pulse wave (PPG); And a memory unit for storing age, gender, height, arm circumference, and weight as information of the examinee.

The present invention further includes a memory unit for storing age, gender, height, arm circumference, and weight as information of a subject.

The calculation processing unit obtains an oscillometric systolic blood pressure, an oscillometric diastolic blood pressure, an oscillometric mean blood pressure as an oscillometric blood pressure, and the diastolic blood pressure DBPcomp correcting the oscillometric diastolic blood pressure DBPcomp.

(However, DBP is the oscillometric diastolic blood pressure, Sex is the gender of the user information, Kmax-rip is the maximum increase (K_rip) of the Cortkov sound power spectrum density (K_max) Is the time between).

The calculation processing unit obtains oscillometric systolic blood pressure, oscillometric diastolic blood pressure, and oscillometric mean blood pressure as oscillometric blood pressure, and systolic blood pressure SBPcomp correcting the oscillometric systolic blood pressure SBP is

(SBP is oscillometric systolic blood pressure, Age is age in user information, Arm _c is arm circumference in user information, and BMI is body mass index).

The calculation processing unit obtains an oscillometric systolic blood pressure, an oscillometric diastolic blood pressure, an oscillometric mean blood pressure as an oscillometric blood pressure, and an average blood pressure (MAPcomp) correcting the oscillometric mean blood pressure (MAP) is

(However, MAP is the oscillometric mean blood pressure, Sex is the gender of the user information, Kmax-rip is the maximum increase (K_rip) of the Cortkov sound power spectral density (K_max) Time between, PTT is the pulse propagation time).

The arithmetic processing section calculates the Kmax-rip for each period to obtain Kmax-rip, which is the time between the maximum increase point (K_rip) of the Cortkov sound power spectral density and the highest point (K_max) of the Cortkov sound power spectral density. Detecting the highest point (K_max) of the Cortkov sound power spectral density, which is the point of the maximum Korpkov sound power spectral density among the spectral densities (PSD), and each period prior to the point of the maximum power spectral density (K_max) The maximum increase point (K_rip) of the Cortkov sound power spectral density, the point at which a sudden increase in power spectral density (PSD) occurs among data having a magnitude of 70% or less of the maximum PSD amplitude in the array of star power spectral density (PSD). ) And the time difference between the maximum increase point (K_rip) of the Krtkov sound power spectral density and the highest point (K_max) of the Krotkov sound power spectral density. Find the Kmax-rip.

The operation processor detects the point (the point having the maximum value) of the peak of the ECG signal in each period, detects the point (the point having the maximum value) of the PPG signal in each period, and detects the point of the ECG signal in each period. The time difference between the peak point and the peak point of the PPG signal is obtained, the average is calculated, and the PTT which is the pulse wave propagation time is obtained.

In addition, the wrist blood pressure measurement method of the present invention is to apply a pressure to the cuff wound around the wrist, and equipped with a pressure sensor to detect the pressure signal of the cuff and the oscillation signal which is a vibration characteristic generated by the pressure change, the pressure signal and the oscillation In the wrist blood pressure measuring method for obtaining an oscillometric blood pressure from a rate signal, an oscillation signal for detecting a peak (maximum value) in each period of the oscillation signal and detecting a valley (minimum value) for each period of the oscillation signal. Peak and valley detection step of; An oscillation maximum amplitude detection step of detecting a maximum amplitude that is a difference between a valley (minimum) located next to the next peak (maximum) in the oscillation signal after the peak and valley detection step of the oscillation signal; An amplitude signal interpolation and fitting step of interpolating using a spline interpolation method after the maximum amplitude detection step and adjusting a peak value (peak value) according to a fitting line; The maximum value of the fitting line obtained in the amplitude signal interpolation and fitting step is detected, the maximum amplitude point having the maximum value is obtained, and the maximum value of detecting the pressure value at the maximum amplitude point as an MAP (oscillometric mean blood pressure). Amplitude point detection step; From the maximum amplitude point obtained in the maximum amplitude point detection step, the pressure value of the point having the maximum value x 0.55 forward (forward direction) is obtained as SBP (oscillometric systolic blood pressure), and the maximum value obtained in the maximum amplitude point detection step. Characteristic ratio application step of obtaining the pressure value of the point having the value of the maximum value x 0.82 backward (in the reverse direction) from the amplitude point as DBP (oscillometric diastolic blood pressure).

The present invention connects a microphone to a portion where a cuff and a pressure sensor are connected, detects a Cortkop sound signal, detects an ECG signal of a wrist from two ECG electrodes, and includes an LED and an optical sensor to light a wrist or a hand. Volumetric pulse wave (PPG) is detected and corrected for oscillometric systolic blood pressure, oscillometric diastolic blood pressure, oscillometric mean blood pressure using a Cortkov sound signal, an electrocardiogram signal, and a optical volume pulse wave (PPG) signal.

The present invention comprises the steps of: importing Cortkov sound array for reading Cortkov sound data; In the Cortkov sound data read in the Cortkov sound array importing step, the Cortkov sound synchronized with the peak of the oscillation signal and the peak of the oscillation signal of each period (each bit) obtained in the valley detection step. A Cortkov sound bit detection step of reading a value (an amplitude value of the Cortkov sound); The nose Cartesian co-in of each period the peak obtained in the Corp. negative bit detection step Stuttgart Corp. negative number of values of data determines whether the 2 ^n, the data can be the number of data is not a 2 ⁿ fill to zero until the 2 ⁿ adjustment step; A total PSD detection step of obtaining a total power spectral density by FFTing data having a data number of 2 ⁿ Kortkopf as a result of the data number adjustment step; A PSD detection step per bit that obtains a power spectral density (PSD) for each period (each bit) from the Cortkov sound data read in the step of importing the Cortkov sound array; A maximum PSD point detection step of detecting a point K_max of the maximum PSD among power spectral density PSDs obtained in the PSD detection step per bit; A maximum PSD point reference array separation step of dividing a power spectral density (PSD) by each period (each bit) obtained in the PSD detection step per bit, before and after the maximum PSD point obtained in the maximum PSD point detection step; Detecting a point where a sudden increase in power spectral density (PSD) occurs among data having a magnitude less than or equal to 70% of the maximum power spectral density (PSD) amplitude in the array before the maximum PSD point obtained in the maximum PSD point reference array separation step; Detecting a maximum PSD increase point; A K-rip time detection step of setting the point at which the PSD increase obtained in the maximum PSD increase point detection step is the fastest increasing point (K_rip); A maximum PSD point time detection step of detecting a time of a point (k-max) of the maximum PSD in the maximum PSD point detection step; A maximum PSD reduction point detection step of detecting a point at which a sudden decrease in PSD appears among data having a size of 70% or less of the maximum PSD amplitude in an array after the maximum PSD point obtained in the maximum PSD point reference array separation step; A K-rdp time detection step of setting the point at which the PSD decrease obtained in the maximum PSD increase point detection step is a rapidly decreasing point (K_rdp); And a K_max-rip detection step of obtaining a K_max-rip which is a time from a point (K_max) of the maximum PSD obtained in the maximum PSD point detection step to a K_rip (rapidly increasing point) of the K-rip time detection step. .

The present invention detects peak points (points having a maximum value) in each period of an ECG signal, and peak detection steps of ECG and PPG detecting peak points (points having a maximum value) in each period of the PPG signal. ; And a PTT detection step of detecting a PTT by obtaining a time difference between a peak point of the ECG signal and a peak point of the PPG signal in each period obtained in the peak detection step of the ECG and PPG.

The diastolic blood pressure (DBPcomp) of correcting the oscillometric diastolic blood pressure (DBP) is

(However, DBP is the oscillometric diastolic blood pressure, Sex is the gender of the user information, Kmax-rip is the maximum increase (K_rip) of the Cortkov sound power spectrum density (K_max) Time between)

Obtained by

Systolic blood pressure (SBPcomp) correcting the oscillometric systolic blood pressure (SBP) is

(However, SBP is oscillometric systolic blood pressure, Age is age in user information, Arm _c is arm circumference in user information, and BMI is body mass index)

Obtained by

The mean blood pressure (MAPcomp) corrected for the oscillometric mean blood pressure (MAP) is

(However, MAP is the oscillometric mean blood pressure, Sex is the gender of the user information, Kmax-rip is the maximum increase (K_rip) of the Cortkov sound power spectral density (K_max) Time between, PTT is the pulse propagation time)

Obtained by

According to the method and apparatus for improving the accuracy of wrist blood pressure measurement using the multi-body signal measurement of the present invention, while measuring the blood pressure at the wrist, representative factors affecting blood pressure along with the oscillation component of blood pressure (PTT, Korotkoff) Simultaneous measurement and analysis of the sound, BMI, etc., improves the convenience of measurement and accuracy in wrist blood pressure measurement.

The present invention provides for electrocardiogram (ECG), oxygen saturation (PPG), and body impedance for consideration of factors that represent the state of the cardiovascular system in order to reduce errors resulting from the detection of blood pressure by measurement of a single factor. It integrates four hardware blocks of blood pressure (NiBP) and implements them into one system to provide a measurement system for multiple bio signals.

The present invention detects factors such as pulse wave propagation time (PTT) (pulse wave velocity (PWV)), Kortokoppe power spectrum density, oscillometric blood pressure, body mass index (BMI), and the systolic system to which individual body information is applied together with these factors. Multilinear regression equations for blood pressure, diastolic blood pressure and mean blood pressure are calculated to improve the accuracy of blood pressure measurement on the wrist.

Since the present invention measures blood pressure by mounting a cuff on the wrist, it eliminates the inconvenience of removing clothes when measuring the upper arm blood pressure, and improves the long-term monitoring of blood pressure and the accuracy of early diagnosis of hypertension to prevent cardiovascular diseases. Will be able to contribute.

In other words, the upper arm oscillometric electronic sphygmomanometer, which is conventionally used for blood pressure measurement of home and hospital patients, not only has the inconvenience of taking off clothes when measuring, but also often decreases the accuracy of blood pressure measurement. There was a problem of inaccuracy in the measurement of those who need, and the proposed method for measuring wrist blood pressure through the measurement of the multi-body signal proposed by the present invention is based on the physical information of the individual with consideration of factors representing the condition of blood vessels. Consideration has been made, and it has the advantage of improving the accuracy as well as providing convenience for the measurer through the blood pressure measurement on the wrist, it can be applied to the easy and accurate measurement of blood pressure in household equipment such as E-Health terminal.

1 is a schematic block diagram of a device for improving the accuracy of wrist blood pressure using multiple bio-signal measurement according to an embodiment of the present invention.
2A and 2B are flowcharts of detecting systolic blood pressure, diastolic blood pressure, and average blood pressure in the processing unit of FIG. 1.
3 is an example of a multi-bio signal obtained by the data collector of FIG. 1.
4 is an explanatory diagram for explaining the detection of the Cortkov sound PSD and the blood pressure correction element.
5 is an explanatory diagram illustrating a PTT detection method as a blood pressure correction element using ECG and PPG signals.

According to the present invention, an oscillometric systolic blood pressure, by detecting the width of oscillation (oscillation), which is a vibration characteristic generated in the cuff while decompressing the arteries by applying a cuff to the wrist pressure of the subject using a cuff, as in an oscillometric electronic blood pressure monitor, Obtain the oscillometric diastolic blood pressure, the oscillometric mean blood pressure, and calculate the oscillometric systolic blood pressure, the oscillometric diastolic blood pressure, and the oscillometric mean blood pressure as the pulse wave propagation time (PTT), the time between the maximum increase and peak of the Korptoff power. (kmax-rip), body mass index (BMI), and arm circumference (Armc) are used to detect corrected systolic blood pressure, corrected diastolic blood pressure, and corrected mean blood pressure. Here, the pulse wave propagation time (PTT), the time between the maximum increase point and the highest point of the Cortkoppe power (kmax-rip) detects the width of oscillation (oscillation), which is a vibration characteristic occurring in the cuff, The change in the Kortkov sound is measured electrically, and at the same time, electrocardiogram (ECG) and PPG are detected and details obtained from these parameters will be described later.

1 is a schematic configuration diagram of an apparatus for improving the accuracy of wrist blood pressure using multiple bio-signal measurements according to an exemplary embodiment of the present invention, including an oscillation signal detector 105, a Cortkop sound detector 205, The electrocardiogram detector 305, the PPG detector 505, the data collector 600, the calculation processor 630, the memory 650, the output unit 670, and the key input unit 690 are included.

The oscillation signal detector 105 detects the width of the oscillation which is a vibration characteristic generated according to the pressure change of the cuff, that is, the oscillation signal among the pressure signals detected by the cuff. The oscillation signal detector 105 includes a pressure sensor unit 100 and a pressure signal preprocessor 110.

The pressure sensor unit 100 electrically detects the pressure signal from the cuff. The pressure sensor unit 100 may include a voltage-to-current converter for outputting the output current pressure signal as a voltage pressure signal when the pressure signal detected by the pressure sensor is output as a current.

The pressure signal preprocessor 110 detects a change in the DC component of the pressure signal through the low pass filter (LPF) 120 through the pressure signal input from the pressure sensor, and the pressure output from the low pass filter (LPF) 120 The signal is passed through a high pass filter (HPF) 140 for detecting a vibration component (blood pressure oscillation), and a blood pressure oscillation signal (hereinafter referred to as an oscillation signal) is detected through an amplifier (AMP) 150. Amplified and transmitted to the data collector 600.

In addition, the pressure signal preprocessor 110 transmits the pressure signal amplified by the output of the low frequency filter (LPF) 120 through the amplifier (AMP) 130 to the data collector 600.

A 10 Hz secondary low pass filter can be used as the low pass filter (LPF) 120, and a 0.5 Hz first order high pass filter can be used as the high pass filter (HPF) 140.

In other words, the change of DC component of the pressure signal was detected through 10Hz secondary lowpass filtering on the signal obtained by the pressure sensor, and the small vibration component (blood pressure oscillation) included in the DC component through 0.5Hz primary highpass filtering and amplification. Detection).

The cortkop sound detection unit 205 detects blood vessel sounds through a microphone simultaneously with the detection of the oscillation signal, as in the case of a cork Kottkoff sound in a stethoscope of a general mercury sphygmomanometer. Cortkop sound detection unit 205 includes a microphone unit 200, Cortkop sound pre-processing unit 210.

In a conventional mercury sphygmomanometer, the arm is covered with a cuff (CUFF) of the sphygmomanometer, the stethoscope (microphone) is placed on the inner pulse region of the cuff, the pressure of the cuff starts to increase, and the sound of blood vessels, that is, the cortkop ( Korotkoff) is designed to measure blood pressure while listening to sounds. In the present invention, the arm of the blood pressure cuff (CUFF) is sufficiently wrapped, the microphone is placed inside the multi-body signal measuring system composed of one system in the present invention, and when the pressure of the cuff starts to increase, It is made to measure. In other words, the Cortkov sound detection unit 205 of the present invention is also configured to detect blood vessel sounds through the microphone at the same time as the oscillation signal is detected.

The microphone unit 200 is integrally connected to a portion where the cuff and the pressure sensor are connected to detect a Cortkop sound (vascular sound). A condenser microphone may be used to detect the Cortkov sound in the microphone unit 200. Since the cuff is mounted on the wrist, the present invention detects a change in copecoff sound in the radial artery.

Cortkop sound pre-processing unit 210 pre-amplifies the Cortkop sound received from the microphone unit 200 through the preamplifier 220, and passes through a band pass filter (BPF) 230 for noise reduction Then, amplified by the amplifier (AMP) 240 and transmitted to the data collector 600. The band pass filter (BPF) 230 may use a 20-200 Hz band pass filter, and detects a sharp form of signal by removing noise from a first amplified Cortkov sound signal through the preamplifier 220. Let's do it.

Here, the oscillation signal detector 105 and the Cortkop sound detector 205 may be referred to as a blood pressure NIBP detector.

The ECG detector 305 attaches an ECG electrode to both wrists or both hands, that is, detects an ECG by two ECG electrodes. The electrocardiogram detector 305 includes an electrocardiogram electrode 300 and an electrocardiogram preprocessor 310.

In the present invention is configured to enable the measurement of ECG Lead I signal. In general, ECG Lead I attaches the electrodes to the right wrist (RA) and left wrist (LA), and attaches the ground (usually the right ankle (RL)) electrode, then the right wrist (RA) and left wrist ( ECG signal is detected by the potential difference of LA).

The electrocardiogram electrode unit 300 detects an electrocardiogram by mounting two electrocardiogram electrodes on both wrists or both hands. The electrocardiogram electrode uses a bivalent chromium plating electrode having good conductivity and excellent biocompatibility.

The ECG preprocessor 310 amplifies the ECG signal received from the ECG electrode unit 300 and removes noise. The ECG preprocessing unit 310 extracts the ECG signal output from the two ECG electrodes of the ECG electrode unit 300 through the differential amplifier 320, and extracts the ECG signal through the band pass filter (BPF) 330. Elimination and elimination of baseline variations are amplified through an amplifier (AMP) and transmitted to the data collection unit (600). The bandpass filter (BPF) 330 may use a secondary bandpass filter to have a pass band of 0.1 to 130Hz.

The PPG detector 505 is a means for detecting a photoplethysmography (PPG) (hereinafter referred to as a pulse wave) and includes a PPG sensor unit 500 and a PPG preprocessor 510. In the present invention, the PPG detector 505 is designed with reference to Webster's Pulse Oximeter System to measure the PPG signal required for the detection of PTT (PWV).

The PPG sensor unit 500 is composed of an LED and an optical sensor, and detects the light from which the light generated by the LED is transmitted or reflected from the tissue through the optical sensor. As the LED, an infrared LED having a wavelength of 890 nm may be used, and a photo detector having a center response frequency of 900 nm may be used as the optical sensor. The PPG sensor unit 500 further includes an LED drive for driving the LED, and the current supplied to the LED is made constant through the LED drive.

Since the PPG signal output from the PPG sensor unit 500 is output as a current signal, the PPG preprocessing unit 510 converts the PPG signal into a PPG signal of a voltage through the current voltage converter 520 and generates a low pass filter (LPF) ( The DC component of the PPG signal is detected through 530, and the AC signal is detected through the high pass filter (HPF) 550 from the PPG signal output from the low pass filter (LPF) 530. Amplified through 560 and transmitted to the data collector 600.

In addition, the PPG preprocessor 510 transmits the PPG signal amplified by the output of the low frequency filter (LPF) 520 through the amplifier (AMP) 540 to the data collector 600.

The low pass filter (LPF) 530 may use a 20 Hz second order low pass filter, and the high pass filter (HPF) 550 may use a 0.5 Hz second order high pass filter.

Here, the pressure sensor unit 100, the microphone unit 200, the electrocardiogram electrode unit 300, and the PPG sensor unit 500 may be referred to as a sensor device unit, and the pressure signal preprocessor 110 and the Cortkop sound preprocessor. The electrocardiogram preprocessor 310 and the PPG preprocessor 510 may be referred to as a signal preprocessor.

The data collection unit 600 receives each signal detected by the oscillation signal detector 105, the Cortkop sound detector 205, the electrocardiogram detector 305, and the PPG detector 505, converts the signals into digital signals, and calculates the data. To 630. The sampling rate of the data collector 600 may be 1 kHz. That is, the data collection unit 600 converts the output signals of the oscillation signal detector 105, the cortkov sound detector 205, the electrocardiogram detector 305, and the PPG detector 505 to the arithmetic processing unit. do. An example of the multiple bio signals obtained by the data collector is shown in FIG. 3.

The operation processor 630 obtains an oscillometric systolic blood pressure, an oscillometric diastolic blood pressure, an oscillometric average blood pressure from the oscillation signal received from the oscillation signal detector 105 through the data collector, and the Cortkop sound detector 205. Obtains the time between the maximum increase and peak of the Kortkov power (kmax-rip) from the Cortkov sound signal received from the data collector from the oscillator, and obtains the oscillometric systolic blood pressure and oscillometric diastolic blood pressure obtained from the oscillation signal. The oscillometric mean blood pressure is corrected to detect corrected systolic blood pressure, corrected diastolic blood pressure, and corrected mean blood pressure.

The memory unit 650 stores information of the examinee, that is, age, gender, height, arm circumference, and the like. In addition, the operation processor 630 stores the output.

The output unit 670 outputs the corrected systolic blood pressure, the corrected diastolic blood pressure, and the corrected average blood pressure output from the operation processor 630 to the screen.

The key input unit 690 inputs the examinee's information, that is, age, gender, height, arm circumference, etc., and sets a test mode and the like.

2A and 2B are flowcharts of detecting systolic blood pressure, diastolic blood pressure, and average blood pressure in the processing unit of FIG. 1.

In the data input step, output signals of the oscillation signal detector 105, the Cortkop sound detector 205, the electrocardiogram detector 305, and the PPG detector 505, that is, the pressure signal, the oscillation signal, and the Cortkop sound signal The ECG signals and the PPG signals are input from the data collector (S110).

In the zero phase filtering step, the output signals of the oscillation signal detector 105, the Cortkop sound detector 205, the electrocardiogram detector 305, and the PPG detector 505, that is, the pressure signal, the oscillation signal, and the Cortkov Zero phase filtering is performed to match the starting point of the sound signal, the electrocardiogram signal, and the PPG signal (S120).

This is to receive the data in real time, but in order to eliminate the slight time difference because the A / D conversion in the data acquisition unit 600 sequentially, in time, too fast delayed back, too lagging behind to start by pulling forward Make it the same.

After the zero phase filtering step (S120), the blood pressure is calculated using the output signal of the oscillation signal detector 105 (S130 to S195), and at the same time, the maximum increase of the Cortkoppe power is output by the output signal of the Cortkov sound detector 205. Obtain the time (kmax-rip) between the point and the highest point (S210 to S295), and simultaneously obtain the pulse wave propagation time (PTT) from the output signal of the ECG detector 305 and the output signal of the PPG detector 505 (S310). ~ S350).

The flow of calculating the blood pressure using the output signal of the oscillation signal detector 105 is as follows.

In the oscillation signal reading step, after the zero phase filtering step S120, the output signal of the oscillation signal detecting unit 105, that is, the pressure signal and the oscillation signal, is read (S130).

In the oscillation signal peak detection step, the inflection points are obtained at each period (each bit) of the oscillation signal, and the inflection points representing the peak are detected, that is, the peak (maximum value) at each period of the oscillation signal. ) Is detected (S140).

In the oscillation signal valley detection step, the inflection points are obtained at each period (each bit) of the oscillation signal, and the inflection points representing the valleys are detected, that is, the valley (minimum value) for each period of the oscillation signal. It is detected (S150).

In the oscillation maximum amplitude detection step, the difference between the previous peak (maximum) and the next valley (minimum) in the oscillation signal, that is, the maximum amplitude is detected at each period (S160).

In the amplitude signal interpolation and fitting step, interpolation is performed using the spline interpolation method, and the peak value (peak value) is adjusted according to the fitting line (S170). That is, the amplitude signal extracted in the oscillation maximum amplitude detection step S160 is interpolated using a spline interpolation method, and the interpolated signal extracts an envelope of the signal through Gaussian peak fitting.

In the maximum amplitude point detection step, the maximum value of the fitting lines obtained in the amplitude signal interpolation and fitting step (S170) is detected, and the pressure value at the point having the maximum value, that is, the maximum amplitude point is MAP (oscillometric average blood pressure). (S180).

As the characteristic ratio application step, the characteristic ratio is applied to calculate the SBP (oscillometric systolic blood pressure) and DBP (oscillometric diastolic blood pressure) (S190), where the characteristic ratio is 55% for SBP and 82% for DBP. From the maximum amplitude point obtained in the maximum amplitude point detection step (S180), a pressure value of 55% of the maximum value (that is, in the forward direction), that is, a point having a maximum value x 0.55, is calculated as SBP, and the maximum amplitude point. The pressure value at the point having a value of 82% of the maximum value, i.e., the maximum value of 0.82, from the maximum amplitude point obtained in the detection step S180 (backward) is obtained by DBP (S190). The characteristic ratio is the ratio obtained empirically in the experiment of Geddes.

In the SBP and DBP detection step, the SBP and DBP calculated in the feature ratio application step (S190) are stored for the next operation (S195).

The following describes the flow of obtaining the time (kmax-rip) between the maximum increase point and the highest point of the Cortkoppe power as the output signal of the Cortkov sound detector 205.

In a step of importing a k-sound array, after the zero phase filtering step S120, the Korkkop sound data, which is the output of the Korptkov sound preprocessing unit 210, is read (S210).

In the Cortkov sound bit detection step, the Cortkov sound value (amplitude value of the Cortkov sound) at the peak of each period (each bit) obtained in the oscillation signal peak detection step S140 is obtained (S220).

A data number 2 ⁿ state determining step, it is determined Cartesian co Corp. negative bit detection step (S220) the nose Stuttgart Corp. negative number of values that the data 2 ⁿ in each period obtained by the peak (S230).

In the data number 2 ⁿ state determining step (S230) co Stuttgart Corp. data number of negative value is 2 ⁿ ahniramyeo, to zero peding step, the number of data is not a 2 ⁿ and filled with zero until the 2 ⁿ (S235).

In the total PSD detection step, the total power spectral density is FFT by FFTing data of a Korkopkov sound value having a data number of 2 ⁿ through the Cortkov sound bit detection step (S220) and the zero fedding step (S235). Obtain (S240). 4 is an explanatory diagram of PSD and blood pressure correction element detection of Cortkov sound. In FIG. 4, PSD is an example of total power spectral density.

In the PSD detection step per bit, the power spectrum density (PSD) is determined for each period (each bit) in the Kortkop sound data input in the step of importing a k-sound array (S210). Obtain (S250). Each period (each bit) in the Cortkov sound data is between a minimum value and an adjacent minimum value, or between a maximum value and an adjacent maximum value. In FIG. 4, a plurality of dots denoted as K-Sound power Spectral Density are obtained by periods.

In the maximum PSD point detection step, the point K_max of the maximum PSD is detected among the power spectral densities PSD of each cycle obtained in the PSD detection step S250 per bit (S260). In FIG. 4, the maximum PSD point is K_MAX.

In the PSD PSD reference array separation step, the PSD is divided by before and after each cycle (each bit) obtained in the PSD detection step per bit (S250) based on the maximum PSD point obtained in the maximum PSD point detection step (S260). S270). That is, it is divided into an array before the maximum PSD point and an array after the maximum PSD point based on the maximum PSD point.

As a maximum PSD increase point detection step, a point having the largest PSD increase is detected in the previous PSD point array arranged in the maximum PSD point reference array separation step (S270) (S280). The maximum PSD increase point detection step optionally detects a point where a sudden increase in PSD appears in data having a size of 70% or less of the maximum PSD amplitude in an array before K_max.

In the K-rip time detection step, the point at which the PSD increase obtained in the maximum PSD increase point detection step S280 is the largest is K_rip (rapidly increasing point) (S290).

In the maximum PSD point time detection step, the time of the point (k-max) of the maximum PSD is detected in the maximum PSD point detection step (S260) (S275).

As a maximum PSD reduction point detection step, a point having the largest PSD reduction in the array after the maximum PSD point obtained in the maximum PSD point reference array separation step (S270) is detected (S280). That is, the time until the point where the difference between the previous value and the next value of the PSD indicates the largest value is obtained. The maximum PSD reduction point detection step detects a point where a sudden decrease in PSD appears in data having a size of 70% or less of the maximum PSD amplitude in an array after K_max in some cases (S285).

In the K-rdp time detection step, the point at which the PSD decrease obtained in the maximum PSD increase point detection step S280 is the largest is K_rdp (rapidly decreasing point) (S295).

Here, the time from K_max to K_rip is denoted by K_max-rip, the time from K-rdp to K_max is denoted by K_rdp-max, and the time from K-rdp to K_rip is denoted by K_rdp-rip.

The flow of calculating the average blood pressure pulse wave propagation time (PTT) using the output signal of the ECG detector 305 and the output signal of the PPG detector 505 is as follows.

In the ECG and PPG array import step, after the zero phase filtering step S120, the output signals of the ECG detector 305 and the PPG detector 505, that is, the ECG signals and the PPG signals, are read (S310).

In the ECG median filtering step, median filtering is performed to stabilize the baseline of the ECG signal (S320).

In the PPG first derivative step, in order to detect the maximum point of the PPG signal, the PPG signal is firstly differentiated (S330).

As a peak detection step of ECG and PPG, a point (a point having a maximum value) that is a peak in each period of the ECG signal output in the ECG median filtering step (S320) is detected, and is output in the PPG first derivative step (S330). Using the first derivative PPG signal, a point (a point having a maximum value) that is a peak in each period of the PPG signal is detected (S340).

In the PTT detection step, the PTT is detected by obtaining the time difference between the peak point of the ECG signal and the peak point of the PPG signal in each period obtained in the peak detection step (S340) of the ECG and PPG (S350). 5 is an explanatory diagram illustrating a PTT detection method as a blood pressure correction element using ECG and PPG signals. For PTT detection, a time difference between a peak point of the ECG signal and a peak point of the PPG signal is obtained.

Next, the blood pressure detection step, the oscillometric systolic blood pressure, oscillometric diastolic blood pressure, oscillometric mean blood pressure (MAP) calculated in S130 ~ S195, between the maximum increase and peak of the Cortkov power calculated in S210 ~ S295 Corrected systolic blood pressure by using the time (kmax-rip), pulse wave transfer time (PTT) calculated in S310 to S350, and body mass index (BMI) and arm circumference (Armc) obtained from the input user information. The corrected diastolic blood pressure and the corrected average blood pressure are detected by the following Equations 2 to 4 (S400).

In the present invention, the user information is input through the key input unit 690 before the test, and the input information is a name, height, age, weight, gender, arm circumference (upper arm), and the operation processor 630 has a body mass index. (BMI) is calculated using Eq. (1) using height and weight.

The corrected mean blood pressure (MAPcomp) is obtained as shown in Equation 2.

Where MAP is the oscillometric mean blood pressure, Sex is the gender of the user information, and Kmax-rip is between the maximum increase (K_rip) of the Krotkov sound power spectral density (K_max). PTT is the pulse wave delivery time.

Corrected systolic blood pressure (SBPcomp) is obtained as shown in Equation 3.

Where SBP is oscillometric systolic blood pressure, Age is age in user information, Arm _c is arm circumference in user information, and BMI is body mass index.

Corrected diastolic blood pressure (DBPcomp) is obtained as shown in Equation 4.

Where DBP is the oscillometric diastolic blood pressure, Sex is the gender of the user information, and Kmax-rip is between the maximum increase of the Krotkoff power spectral density (K_rip) and the highest point of the Korkkopk power spectral density (K_max). Say time.

2, the maximum amplitude algorithm (MAA) was used to analyze the blood pressure signal in the present invention. Basically, the peak of the oscillation signal (maximum value at each bit of blood pressure oscillation) is detected, and the minimum value between each peak is detected as a valley (minimum value at each bit of blood pressure oscillation), and the difference is obtained. The amplitude signal was extracted. The extracted amplitude signals were interpolated using spline interpolation. The interpolated signal was extracted with Gaussian peak fitting, and the pressure value at the maximum peak point in the extracted envelope was MAP. After detection of MAP, the SBP and DBP were calculated by applying the characteristic ratio. The characteristic ratio was 55% for SBP and 82% for DBP.

Kortkov's analysis focused on changing the PSD. Based on the oscillation peak, the Cortkov signal is classified as beat by beat, and the number of data is set to 2 ⁿ through Zero Padding for Fourier transform of the classified signal. The change was confirmed by detection. K_MAX is defined as the point where the maximum size of the PSD changes, and K_rip (rapidly increasing point), the point where the sudden increase of PSD appears in the data having the size of 70% or less of the maximum PSD amplitude in the array before K_max, K_rdp (rapidly decreasing point) is defined as the point where the sudden decrease of PSD among the data having the size of 70% or less of the maximum PSD amplitude in the array after K_max, and the time between each point is K_max-rip, K_rdp-max, K_rdp-rip.

For the detection of PTT, a method of obtaining the time between peak points of ECG and PPG was applied. For the detection of ECG peak, the method of detecting the R-Peak by highlighting the difference between the original signal and the median filtering was used. In the case of the PPG signal, the peak was detected through the derivative. In addition, a signal having a time between two peaks of 1 second or more was determined to be an error and excluded.

Although the embodiments of the present invention have been described above with reference to the accompanying drawings, those skilled in the art to which the present invention belongs may be embodied in other specific forms without changing the technical spirit or essential features of the present invention. You will understand that. Therefore, the embodiments described above are to be understood in all respects as illustrative and not restrictive.

100: pressure sensor
105: oscillation signal detector
110: pressure signal preprocessor
120, 530: LPF
130, 150, 240, 340, 540, 560: AMP
140, 550: HPF
200: microphone unit
205: Cortkop sound detector
210: Cortkov sound preprocessing unit
220: preamplifier
230, 330: BPF
300: ECG electrode portion
305: ECG detection unit
310: ECG pretreatment unit
320: differential amplifier
500: PPG sensor unit
505: PPG detector
510: PPG pretreatment unit
520: current voltage converter
600: data collector
630: arithmetic processing unit
650: memory
670: output unit
690: key input unit

Claims (16)

A cuff made to wind around the wrist and configured to apply pressure;
An oscillation signal detection unit having a pressure sensor connected to the cuff, the oscillation signal detecting unit detecting a pressure signal of the cuff and an oscillation signal which is a vibration characteristic generated according to a pressure change in the pressure signal;
Cortkop sound detection unit for connecting the microphone to the portion where the cuff and the pressure sensor is connected, and detects the Cortkop sound signal;
Cort received by A / D conversion from the output of the oscillation signal detector to obtain the oscillometric blood pressure from the received oscillation signal and the oscillometric blood pressure from the output of the Cortkov sound detector. An arithmetic processing unit for correcting using the cope sound signal;
Wrist blood pressure measuring device, characterized in that consisting of.
A cuff made to wind around the wrist and configured to apply pressure;
An oscillation signal detection unit having a pressure sensor connected to the cuff, the oscillation signal detecting unit detecting a pressure signal of the cuff and an oscillation signal which is a vibration characteristic generated according to a pressure change in the pressure signal;
A memory unit for storing age, gender, height, arm circumference, and weight as information of a subject;
The body mass index (BMI) is obtained by dividing the subject's height by weight, and A / D conversion is performed from the output of the oscillation signal detection unit to obtain an oscillometric blood pressure from the input pressure signal and the oscillation signal, and the oscillometric blood pressure is measured by the body mass index (BMI). A calculation processing unit for correcting using;
Wrist blood pressure measuring device, characterized in that consisting of.
The method of claim 1,
An electrocardiogram detector including two electrocardiogram electrodes to be mounted on both wrists or mounted on both hands to detect an electrocardiogram;
A PPG detector having an LED and an optical sensor positioned to be mounted on a wrist or a hand to detect an optical volume pulse wave (PPG);
A memory unit for storing age, gender, height, arm circumference, and weight as information of a subject;
Wrist blood pressure measuring device further comprising.
The method of claim 1,
Wrist information measuring apparatus further comprises a memory for storing the age, sex, height, arm circumference, weight as the information of the subject.
The method of claim 4
The calculation processing unit obtains oscillometric systolic blood pressure, oscillometric diastolic blood pressure, oscillometric average blood pressure as oscillometric blood pressure,
The diastolic blood pressure (DBPcomp) of correcting the oscillometric diastolic blood pressure (DBP) is

(However, DBP is the oscillometric diastolic blood pressure, Sex is the gender of the user information, Kmax-rip is the maximum increase (K_rip) of the Cortkov sound power spectrum density (K_max) Time between)
Wrist blood pressure measuring device, characterized in that obtained by.
The method according to claim 2, wherein
The calculation processing unit obtains oscillometric systolic blood pressure, oscillometric diastolic blood pressure, oscillometric average blood pressure as oscillometric blood pressure,
Systolic blood pressure (SBPcomp) correcting the oscillometric systolic blood pressure (SBP) is

(However, SBP is oscillometric systolic blood pressure, Age is age in user information, Arm _c is arm circumference in user information, and BMI is body mass index)
Wrist blood pressure measuring device, characterized in that obtained by.
The method of claim 3,
The calculation processing unit obtains oscillometric systolic blood pressure, oscillometric diastolic blood pressure, oscillometric average blood pressure as oscillometric blood pressure,
The mean blood pressure (MAPcomp) corrected for the oscillometric mean blood pressure (MAP) is

(However, MAP is the oscillometric mean blood pressure, Sex is the gender of the user information, Kmax-rip is the maximum increase (K_rip) of the Cortkov sound power spectral density (K_max) Time between, PTT is the pulse propagation time)
Wrist blood pressure measuring device, characterized in that obtained by.
The method according to any one of claims 5 to 7,
The calculation processing unit,
To find the Kmax-rip, which is the time between the maximum increase (K_rip) of the Krotkov sound power spectral density (K_max),
Detecting the highest point (K_max) of the Cortkop sound power spectral density, which is the point of the maximum Cortkop sound power spectral density, among the Cortkov sound power spectral density (PSD) for each period,
A sudden increase in power spectral density (PSD) among data having a magnitude less than or equal to 70% of the maximum PSD amplitude in the previous arrangement of power spectral density (PSD) for each period based on the point of the maximum power spectral density (K_max) Detects the maximum increase point (K_rip) of the Cortkov sound power spectral density, where
A wrist blood pressure measuring device, characterized by obtaining a Kmax-rip by detecting a time difference between a maximum increase point (K_rip) of a Rotkov sound power spectral density and a maximum point (K_max) of a Kortkov sound power spectral density.
The method of claim 7, wherein
The calculation processing unit,
Detects the point (the point with the maximum value) that is the peak of the ECG signal in each period,
In each period, the point (the point having the maximum value) of the peak of the PPG signal is detected.
Wrist blood pressure measuring device, characterized in that the time difference between the peak point of the electrocardiogram signal and the peak point of the PPG signal in each cycle is obtained, and the average is obtained to obtain the PTT which is the pulse wave propagation time.
A wrist that applies pressure to a cuff wound around the wrist, and is equipped with a pressure sensor to detect the pressure signal of the cuff and an oscillation signal that is a vibration characteristic generated by the pressure change, and obtain an oscillometric blood pressure from the pressure signal and the oscillation signal. In the blood pressure measurement method,
A peak and valley detection step of the oscillation signal, detecting a peak (maximum value) in each period of the oscillation signal, and detecting a valley (minimum value) for each period of the oscillation signal;
An oscillation maximum amplitude detection step of detecting a maximum amplitude that is a difference between a valley (minimum) located next to the next peak (maximum) in the oscillation signal after the peak and valley detection step of the oscillation signal;
An amplitude signal interpolation and fitting step of interpolating using a spline interpolation method after the maximum amplitude detection step and adjusting a peak value (peak value) according to a fitting line;
The maximum value of the fitting line obtained in the amplitude signal interpolation and fitting step is detected, the maximum amplitude point having the maximum value is obtained, and the maximum value of detecting the pressure value at the maximum amplitude point as an MAP (oscillometric mean blood pressure). Amplitude point detection step;
From the maximum amplitude point obtained in the maximum amplitude point detection step, the pressure value of the point having the maximum value x 0.55 forward (forward direction) is obtained as SBP (oscillometric systolic blood pressure), and the maximum value obtained in the maximum amplitude point detection step. A characteristic ratio application step of obtaining a pressure value at a point having a value of the maximum value x 0.82 backward from the amplitude point as an oscillometric diastolic blood pressure (DBP);
Wrist blood pressure measuring method comprising a.
The method of claim 10,
Connect the microphone to the part where the cuff and the pressure sensor are connected to detect the Cortkov sound signal, detect the ECG signal of the wrist from the two ECG electrodes, and have the LED and light sensor And detecting and correcting oscillometric systolic blood pressure, oscillometric diastolic blood pressure, and oscillometric mean blood pressure by using a Cortkov sound signal, an electrocardiogram signal, and a optical volume pulse wave (PPG) signal. How to measure blood pressure.
The method of claim 11,
Importing a Cortkov sound array for reading Cortkov sound data;
In the Cortkov sound data read in the Cortkov sound array importing step, the Cortkov sound synchronized with the peak of the oscillation signal and the peak of the oscillation signal of each period (each bit) obtained in the valley detection step. A Cortkov sound bit detection step of reading a value (an amplitude value of the Cortkov sound);
The nose Cartesian co-in of each period the peak obtained in the Corp. negative bit detection step Stuttgart Corp. negative number of values of data determines whether the 2 ^n, the data can be the number of data is not a 2 ⁿ fill to zero until the 2 ⁿ adjustment step;
A total PSD detection step of obtaining a total power spectral density by FFTing data having a data number of 2 ⁿ Kortkopf as a result of the data number adjustment step;
A PSD detection step per bit that obtains a power spectral density (PSD) for each period (each bit) from the Cortkov sound data read in the step of importing the Cortkov sound array;
A maximum PSD point detection step of detecting a point K_max of the maximum PSD among power spectral density PSDs obtained in the PSD detection step per bit;
A maximum PSD point reference array separation step of dividing a power spectral density (PSD) by each period (each bit) obtained in the PSD detection step per bit, before and after the maximum PSD point obtained in the maximum PSD point detection step;
Detecting a point where a sudden increase in power spectral density (PSD) occurs among data having a magnitude less than or equal to 70% of the maximum power spectral density (PSD) amplitude in the array before the maximum PSD point obtained in the maximum PSD point reference array separation step; Detecting a maximum PSD increase point;
A K-rip time detection step of setting the point at which the PSD increase obtained in the maximum PSD increase point detection step is the fastest increasing point (K_rip);
A maximum PSD point time detection step of detecting a time of a point (k-max) of the maximum PSD in the maximum PSD point detection step;
A maximum PSD reduction point detection step of detecting a point at which a sudden decrease in PSD appears among data having a size of 70% or less of the maximum PSD amplitude in an array after the maximum PSD point obtained in the maximum PSD point reference array separation step;
A K-rdp time detection step of setting the point at which the PSD decrease obtained in the maximum PSD increase point detection step is a rapidly decreasing point (K_rdp);
A K_max-rip detection step of obtaining a K_max-rip which is a time from a point (K_max) of the maximum PSD obtained in the maximum PSD point detection step to a rapidly increasing point (K_rip) of the K-rip time detection step;
Blood pressure measuring method comprising a.
The method of claim 12,
Detecting a peak point (a point having a maximum value) in each period of the ECG signal and detecting a peak point (a point having a maximum value) in each period of the PPG signal;
A PTT detection step of detecting a PTT by obtaining a time difference between a peak point of the ECG signal and a peak point of the PPG signal in each period obtained in the peak detection step of the ECG and PPG;
Blood pressure measuring method comprising a.
The method of claim 13,
The diastolic blood pressure (DBPcomp) of correcting the oscillometric diastolic blood pressure (DBP) is

(However, DBP is the oscillometric diastolic blood pressure, Sex is the gender of the user information, Kmax-rip is the maximum increase (K_rip) of the Cortkov sound power spectrum density (K_max) Time between)
Wrist blood pressure measurement method characterized in that obtained by.
The method of claim 13,
Systolic blood pressure (SBPcomp) correcting the oscillometric systolic blood pressure (SBP) is

(However, SBP is oscillometric systolic blood pressure, Age is age in user information, Arm _c is arm circumference in user information, and BMI is body mass index)
Wrist blood pressure measurement method characterized in that obtained by.
The method of claim 13,
The mean blood pressure (MAPcomp) corrected for the oscillometric mean blood pressure (MAP) is

(However, MAP is the oscillometric mean blood pressure, Sex is the gender of the user information, Kmax-rip is the maximum increase (K_rip) of the Cortkov sound power spectral density (K_max) Time between, PTT is the pulse propagation time)
Wrist blood pressure measurement method characterized in that obtained by.

Images