KR102568904B1 - Apparatus and method for estimating blood pressure - Google Patents

Abstract

An apparatus for non-invasively estimating blood pressure is disclosed. According to an embodiment, the blood pressure estimating apparatus extracts a cardiovascular feature based on an acquisition unit that acquires a bio-signal of a subject and the bio-signal, and calculates the blood pressure in consideration of a change tendency of the extracted cardiovascular feature compared to a reference cardiovascular feature. It may include a processor for estimating.

Description

Apparatus and method for estimating blood pressure {APPARATUS AND METHOD FOR ESTIMATING BLOOD PRESSURE}

The present invention relates to a blood pressure estimating device and method, and more particularly, to a technique for independently estimating systolic blood pressure and diastolic blood pressure in a non-invasive manner.

Due to the recent aging population structure, rapidly increasing medical expenses, and lack of professional medical service personnel, active research is being conducted on IT-medical convergence technology that combines IT and medical technology. In particular, the monitoring of the human body's health status is not limited to hospitals, but is expanding to the mobile healthcare field, which monitors the health status of users moving in daily life such as at home and offices anytime, anywhere. There are representative types of biosignals that indicate the health status of an individual, such as ECG (Electrocardiography), PPG (Photoplethysmogram), and EMG (Electromyography) signals. Signal sensors are being developed. In particular, in the case of the PPG sensor, it is possible to estimate the blood pressure of the human body by analyzing the pulse wave shape reflecting the state of the cardiovascular system.

A typical biosignal-based indirect blood pressure estimation method extracts various features correlated with blood pressure from a biosignal and estimates a blood pressure value based on these features. Blood pressure can be changed by various mechanisms during daily life, and in some cases, the change tendency of systolic blood pressure (SBP) and diastolic blood pressure (DBP) may be different. Most of the indirect blood pressure estimation methods have a problem in that it is difficult to independently and accurately estimate the systolic blood pressure and the diastolic blood pressure when a decoupling phenomenon in which the systolic blood pressure and the diastolic blood pressure change in different tendencies occurs. In the case of an indirect blood pressure estimation method in which independent estimation of systolic blood pressure and diastolic blood pressure is difficult, incorrect blood pressure information may be provided to the user.

Republic of Korea Patent Publication No. 10-2017-0048970 (2017.05.10)

An apparatus and method for independently estimating systolic blood pressure and diastolic blood pressure in consideration of changes in cardiovascular characteristics by various mechanisms are presented.

According to one aspect, the apparatus for estimating blood pressure extracts a cardiovascular feature based on an acquisition unit that acquires a biosignal of a subject and the biosignal, and calculates blood pressure in consideration of a change tendency of the extracted cardiovascular feature compared to a reference cardiovascular feature. It may include a processor for estimating.

In this case, the biosignal may include one or more of photoplethysmogram (PPG), electrocardiography (ECG), electromyography (EMG), and ballistocardiogram (BCG).

The acquisition unit may include one or more sensors that measure at least some of the bio-signals.

In addition, the blood pressure estimating apparatus may further include a communication unit for receiving a biosignal from one or more external sensors that measure at least a part of the biosignal and transmitting the received biosignal to the acquirer.

The processor may include a feature extraction unit that extracts cardiovascular features including at least one of cardiac output and total periphral resistance based on characteristic points extracted from the biosignal.

The processor provides heart rate information from the biosignal, the shape of the waveform of the biosignal, the time and amplitude of the maximum point of the biosignal, the time and amplitude of the minimum point of the biosignal, the area of the biosignal waveform, and the pulse waveforms constituting the biosignal. It may further include a feature point extraction unit that extracts a feature point including at least one of amplitude and time information.

The processor may include a blood pressure estimating unit that independently estimates the systolic blood pressure (SBP) and the diastolic blood pressure (DBP) based on the change trend.

The processor further includes a scaling unit that scales the extracted cardiovascular feature into a first cardiovascular feature and a second cardiovascular feature, and the blood pressure estimator estimates the systolic blood pressure based on the first cardiovascular feature and the diastolic blood pressure based on the second cardiovascular feature. can be estimated.

In this case, the cardiovascular characteristics may include one or more of cardiac output and total periphral resistance.

The scaling unit scales the cardiac output into the first cardiac output and the second cardiac output, scales the total vascular resistance into the first total vascular resistance and the second total vascular resistance, and the blood pressure estimation unit scales the first cardiac output and the first total vascular resistance. The systolic blood pressure may be estimated by linearly combining , and the diastolic blood pressure may be estimated by linearly combining the second cardiac output and the second total vascular resistance.

The blood pressure estimator assigns a first cardiac output weight and a first total vascular resistance weight to the first cardiac output and the first total vascular resistance, respectively, and then linearly combines the second cardiac output and the second total vascular resistance with the second cardiac output weight, respectively. After assigning weights and second total vascular resistance weights, linear combination may be performed.

The blood pressure estimator estimates the systolic blood pressure by applying a predetermined first scaling factor to the result of linearly combining the first cardiac output and the first total vascular resistance, and obtains a predetermined value from the result of linearly combining the second cardiac output and the second total vascular resistance. The diastolic blood pressure may be estimated by applying the second scaling factor.

The scaling unit may determine scaling degrees of the first cardiovascular feature and the second cardiovascular feature according to the type and change tendency of the extracted cardiovascular feature.

The scaling unit may equalize or decrease an increase rate of the second cardiac output compared to an increase rate of the first cardiac output according to an increase in cardiac output when the type of cardiovascular characteristic is cardiac output and the change tendency is increasing.

The scaling unit may equalize or increase a reduction rate of the second total vascular resistance compared to a reduction rate of the first total vascular resistance according to a decrease in total vascular resistance when the type of cardiovascular characteristic is total vascular resistance and the change tendency is reduced.

Also, the blood pressure estimating device may further include an output unit that outputs a blood pressure estimation result of the processor.

According to one aspect, a method for estimating blood pressure includes acquiring a biosignal of a subject, extracting one or more cardiovascular features based on the biosignal, and considering a change trend of the extracted cardiovascular feature compared to a reference cardiovascular feature. It may include estimating blood pressure.

In this case, the biosignal may include one or more of photoplethysmogram (PPG), electrocardiography (ECG), electromyography (EMG), and ballistocardiogram (BCG).

In the step of extracting cardiovascular features, cardiovascular features including at least one of cardiac output and total periphral resistance may be extracted based on characteristic points extracted from the biosignal.

In addition, the blood pressure estimation method includes heart rate information from the biosignal, the shape of the waveform of the biosignal, the time and amplitude of the maximum point of the biosignal, the time and amplitude of the minimum point of the biosignal, the area of the biosignal waveform, and the time elapse of the biosignal. The method may further include extracting feature points including at least one of amplitude and time information of component pulse waveforms constituting the biosignal, and a branching point between two or more extracted feature points.

In the step of estimating the blood pressure, the systolic blood pressure (SBP) and the diastolic blood pressure (DBP) may be independently estimated based on the change trend.

In addition, the method for estimating blood pressure further includes scaling the extracted cardiovascular feature into a first cardiovascular feature and a second cardiovascular feature, and the step of estimating the blood pressure estimates a systolic blood pressure based on the first cardiovascular feature and second cardiovascular features. Based on cardiovascular characteristics, diastolic blood pressure can be estimated.

In the scaling step, scaling degrees of the first cardiovascular feature and the second cardiovascular feature may be determined according to the type and change tendency of the extracted cardiovascular feature.

In the scaling step, if the type of the extracted cardiovascular feature is cardiac output and the change tendency is increasing, the increase rate of the first cardiac output compared to the increase rate of the second cardiac output may be equalized or decreased according to the increase in cardiac output.

In the scaling step, if the type of the extracted cardiovascular feature is total vascular resistance and the change tendency is reduced, the reduction rate of the second total vascular resistance compared to the reduction rate of the first total vascular resistance may be equalized or increased according to the decrease in total vascular resistance. there is.

Also, the blood pressure estimation method may further include outputting a blood pressure estimation result.

According to one aspect, the wearable device includes a main body worn on a subject, a strap connected to both ends of the main body and fixing the main body to the subject in a state wrapped around the subject, and a sensor mounted on the main body and measuring a biosignal from the subject and a processor for extracting one or more cardiovascular features based on the measured bio-signals and estimating blood pressure in consideration of a change tendency of the extracted cardiovascular features compared to a reference cardiovascular feature.

The sensor may include one or more of a photoplethysmogram (PPG) sensor, an electrocardiography (ECG) sensor, an electromyography (EMG) sensor, and a ballistocardiogram (BCG) sensor.

In this case, the cardiovascular characteristics may include one or more of cardiac output and total periphral resistance.

The processor may independently estimate systolic blood pressure (SBP) and diastolic blood pressure (DBP) based on the trend of changes in cardiovascular characteristics.

The processor may scale the extracted cardiovascular feature into a first cardiovascular feature and a second cardiovascular feature, estimate a systolic blood pressure based on the first cardiovascular feature, and estimate a diastolic blood pressure based on the second cardiovascular feature.

The wearable device may further include a display unit outputting a blood pressure estimation result of the processor.

The wearable device may further include a communication unit that transmits a blood pressure estimation result of the processor to an external device.

The accuracy of blood pressure estimation can be improved by independently estimating systolic blood pressure and diastolic blood pressure in consideration of changes in cardiovascular characteristics by various mechanisms.

1 is a block diagram of a blood pressure estimating device according to an exemplary embodiment.
2 is a block diagram of a blood pressure estimating device according to another embodiment.
3A to 3D are diagrams for explaining a correlation between a change trend of cardiovascular characteristics and blood pressure.
4 is a block diagram of a processor configuration according to the embodiment of FIGS. 1 and 2 .
5A to 5D are views for explaining extraction and scaling of cardiovascular features.
6 is a flowchart of a blood pressure estimation method according to an embodiment.
7A and 7B are diagrams for describing a wearable device according to an exemplary embodiment.

Details of other embodiments are included in the detailed description and drawings. The advantages and features of the described technology, and how to achieve them, will become clear with reference to the embodiments described below in detail in conjunction with the drawings. Like reference numbers designate like elements throughout the specification.

Terms such as first and second may be used to describe various components, but the components should not be limited by the terms. Terms are only used to distinguish one component from another. Singular expressions include plural expressions unless the context clearly dictates otherwise. In addition, when a certain component is said to "include", this means that it may further include other components without excluding other components unless otherwise stated. In addition, terms such as “… unit” and “module” described in the specification mean a unit that processes at least one function or operation, and may be implemented as hardware or software or a combination of hardware and software.

Hereinafter, embodiments of an apparatus and method for estimating blood pressure will be described in detail with reference to drawings.

1 is a block diagram of a blood pressure estimating device according to an exemplary embodiment. The apparatus 100 for estimating blood pressure according to this embodiment may be mounted on a terminal such as a smart phone, tablet PC, desktop PC, or notebook PC, or manufactured as an independent hardware device. In this case, the independent hardware device may be a wearable device such as a wristwatch type, a bracelet type, a wristband type, a ring type, a glasses type, or a hairband type that can be worn on the subject. However, it is not limited thereto.

Referring to FIG. 1 , the apparatus 100 for estimating blood pressure includes an acquisition unit 110 and a processor 120 .

The acquisition unit 110 may include one or more sensors and measure various biosignals from the subject through the sensors. In this case, the one or more sensors may be sensors that measure photoplethysmogram (PPG), electrocardiography (ECG), electromyography (EMG), ballistocardiogram (BCG), and the like. However, it is not limited thereto.

The processor 120 may estimate the blood pressure based on the biosignal measured by the acquisition unit 110 . When the biosignal is received from the acquisition unit 110, the processor 120 may extract cardiovascular features that affect blood pressure from the biosignal. In this case, the cardiovascular characteristics may include cardiac output (CO) and total periphral resistance (TPR). However, it is not limited to this.

When the cardiovascular features are extracted, the processor 120 may determine a change tendency of the extracted cardiovascular features compared to a reference cardiovascular feature, and estimate blood pressure in consideration of the change trends of the cardiovascular features. At this time, the processor 120 may independently estimate systolic blood pressure (SBP) and diastolic blood pressure (DBP).

For example, the processor 120 considers the change tendency of each cardiovascular feature, scales each cardiovascular feature into a cardiovascular feature for systolic blood pressure and a cardiovascular feature for diastolic blood pressure, and scales the cardiovascular feature for systolic blood pressure and the cardiovascular feature for diastolic blood pressure. Systolic blood pressure and diastolic blood pressure can be independently estimated using the features. For example, the processor 120 may assign weights to each of the cardiovascular features scaled for systolic blood pressure and diastolic blood pressure, respectively, and then linearly combine them, and estimate the systolic blood pressure and the diastolic blood pressure, respectively, based on the linear results. In this case, a scaling factor for adjusting the scaling of the entire cardiovascular feature may be applied to the result of the linear combination, and the systolic blood pressure and the diastolic blood pressure may be estimated using the result.

2 is a block diagram of a blood pressure estimating device according to another embodiment.

Referring to FIG. 2 , the blood pressure estimating apparatus 200 according to the present embodiment may include an acquisition unit 110, a processor 120, a communication unit 210, an output unit 220, and a storage unit 230. . Since the acquisition unit 110 and the processor 120 have been described with reference to FIG. 1 , the rest of the components will be mainly described below.

The communication unit 210 may collaborate in blood pressure estimation by communicating with the external device 250 under the control of the processor 120 . For example, the communication unit 210 may receive a biosignal from the external device 250 and transmit it to the acquisition unit 110 . In this case, the external device 250 may include a sensor for measuring photoplethysmogram (PPG), electrocardiography (ECG), electromyography (EMG), ballistocardiogram (BCG), and the like. Through this, the size of the main body can be reduced by mounting only a minimum sensor, for example, a PPG sensor for measuring a photoplethysmogram (PPD) signal, in the acquisition unit 110 .

In addition, the communication unit 210 transmits bio-signal measurement results and/or blood pressure estimation results to the external device 250 under the control of the processor 120, so that the external device 250 manages blood pressure history or monitors health conditions. can be made to do In this case, the external device 250 may include, but is not limited to, a smart phone, a tablet PC, a desktop PC, a notebook PC, and a device of a medical institution.

The communication unit 210 includes Bluetooth communication, Bluetooth Low Energy (BLE) communication, Near Field Communication (NFC), WLAN communication, Zigbee communication, Infrared Data Association (IrDA) communication, and WFD. It is possible to communicate with an external device using (Wi-Fi Direct) communication, UWB (ultra-wideband) communication, Ant+ communication, WIFI communication, RFID (Radio Frequency Identification) communication, 3G communication, 4G communication, and 5G communication. However, this is only an example and is not limited thereto.

The output unit 220 may output additional information such as an acquired bio-signal, a blood pressure estimation result, and a warning according to the blood pressure estimation result. For example, the output unit 220 may visually provide various types of information to the user through a display module. For example, when the blood pressure estimation result is displayed, warning information may be displayed to the user by emphasizing the estimated blood pressure outside the normal range by displaying it in red. As another example, various types of information may be provided to the user in a non-visual manner such as voice, vibration, or touch through a speaker module or a haptic module. For example, systolic blood pressure and diastolic blood pressure may be guided by voice. In this case, when the estimated blood pressure is out of the normal range, the user may be notified of an abnormal state of health through vibration or tactile sensation.

The storage unit 230 may store reference information, obtained bio-signals or blood pressure estimation results, and the like. In this case, the reference information may include user information such as the user's age, gender, and health condition, or an estimation model such as a blood pressure estimation formula.

At this time, the storage unit 230 is a flash memory type, a hard disk type, a multimedia card micro type, or a card type memory (eg, SD or XD memory). etc.), RAM (Random Access Memory: RAM) SRAM (Static Random Access Memory), ROM (Read-Only Memory: ROM), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic It includes, but is not limited to, storage media such as a memory, a magnetic disk, and an optical disk.

3A to 3D are diagrams for explaining a correlation between a change trend of cardiovascular characteristics and blood pressure.

In general, mean blood pressure (MBP) is proportional to cardiac output and total vascular resistance as shown in Equation 1 below.

Here, ΔMBP represents the difference in mean blood pressure between the left ventricle and the right atrium, and generally does not exceed 3 to 5 mmHg in the case of mean blood pressure in the right atrium, and thus has a similar value to mean mean blood pressure in the left ventricle or brachial mean blood pressure.

In general, systolic blood pressure and diastolic blood pressure use values within the range of 0.5 to 0.7 left and right of the average blood pressure thus calculated. However, the systolic blood pressure and the diastolic blood pressure may have a decoupled phenomenon that does not follow the change trend of the mean blood pressure according to a blood pressure change mechanism. Therefore, in order to improve the accuracy of blood pressure estimation, it is necessary to estimate the blood pressure by considering the effect of the blood pressure change mechanism.

Referring to FIGS. 3A and 3B , changes in cardiac output, total vascular resistance, systolic blood pressure, and diastolic blood pressure may be estimated according to various mechanisms affecting changes in blood pressure.

For example, when the mechanism of change in blood pressure is anaerobic exercise or isometric exercise, cardiac output, total blood resistance, systolic blood pressure, and diastolic blood pressure tend to all increase compared to the rest period. Therefore, when blood pressure is estimated by combining the cardiac output feature (f1) and the total vascular resistance feature (f2), when the cardiac output feature (f1) increases and the total vascular resistance feature (f2) tends to remain the same , the systolic blood pressure and the diastolic blood pressure are all proportional to the combined result.

As another example, when the mechanism of change in blood pressure is aerobic exercise, cardiac output and systolic blood pressure tend to increase, but total vascular resistance and diastolic blood pressure tend to decrease. Therefore, when blood pressure is estimated by combining the cardiac output feature (f1) and the total vascular resistance feature (f2), the cardiac output feature (f1) increases very significantly and the total vascular resistance feature (f2) tends to decrease. When shown, the systolic blood pressure increases somewhat greatly, but the degree of increase in the diastolic blood pressure becomes very small relatively compared to the increase in the cardiac output characteristic (f1).

As another example, when the mechanism of change in blood pressure is alcohol, cardiac output increases compared to rest phase, but total vascular resistance, systolic blood pressure, and diastolic blood pressure all tend to decrease. Therefore, when blood pressure is estimated by combining the cardiac output feature (f1) and the total vascular resistance feature (f2), the cardiac output when the degree of increase in the feature (f1) is greater than the degree of decrease in the total vascular resistance feature (f2) As a result of combining the feature (f1) and the total vascular resistance feature (f2), the systolic blood pressure and the diastolic blood pressure may show a decreasing tendency even though they show an increasing tendency. In this case, the degree of reduction of the diastolic blood pressure may be relatively greater than the systolic blood pressure.

As another example, when the mechanism of change in blood pressure is breath holding, when blood pressure is estimated by combining the cardiac output feature (f1) and the total vascular resistance feature (f2), the total vascular resistance feature (f2) increases and the cardiac output feature When (f1) remains the same, both systolic and diastolic blood pressures are proportional to their combined results.

3C and 3D illustrate changes in systolic blood pressure and diastolic blood pressure according to changes in cardiac output and total vascular resistance.

Referring to FIG. 3C , when cardiac output increases or decreases, both systolic and diastolic blood pressures show the same change trend. However, when cardiac output gradually increases compared to rest, the increase in diastolic blood pressure tends to be smaller than the increase in systolic blood pressure. That is, when cardiac output (x-axis) gradually increases, the degree of increase in diastolic blood pressure (y-axis) gradually decreases, and when cardiac output (y-axis) gradually increases, the degree of increase in systolic blood pressure (x-axis) gradually increases. indicate a trend.

Referring to FIG. 3D , when total vascular resistance increases or decreases, both systolic and diastolic blood pressure show the same change trend. However, when total vascular resistance gradually decreases compared to the rest period, the degree of decrease in diastolic blood pressure tends to gradually increase more than the degree of decrease in systolic blood pressure. That is, when cardiac output (x-axis) gradually decreases, the degree of decrease in diastolic blood pressure (y-axis) gradually increases, and when cardiac output (y-axis) gradually decreases, the degree of decrease in systolic blood pressure (x-axis) gradually decreases. indicate a trend.

As described above, cardiovascular characteristics such as cardiac output and total vascular resistance, and changes in systolic blood pressure and diastolic blood pressure vary according to the mechanism of blood pressure change. In particular, systolic blood pressure is sensitive to changes in cardiac output, and diastolic blood pressure is more sensitive to changes in total vascular resistance. Therefore, according to the present embodiment, cardiovascular features such as cardiac output and/or total vascular resistance are extracted from various biosignals, the extracted cardiovascular features are compared with the reference cardiovascular features in the rest period to determine the change trend, and the change in cardiovascular features The accuracy of blood pressure estimation can be improved by independently estimating the systolic blood pressure and the diastolic blood pressure in consideration of the trend.

4 is a block diagram of a processor configuration according to the embodiment of FIGS. 1 and 2 . 5A to 5D are views for explaining extraction and scaling of cardiovascular features. An embodiment in which the processor 400 estimates blood pressure based on biosignals will be described with reference to FIGS. 4 to 5D.

Referring to FIG. 4 , the processor 400 may include a feature point extraction unit 410, a feature extraction unit 420, a scaling unit 430, and a blood pressure estimating unit 440.

The feature point extraction unit 410 may extract feature points for blood pressure estimation from various bio-signals. At this time, the feature points constitute the heartbeat information, the shape of the bio-signal waveform, the time and amplitude of the maximum point of the bio-signal, the time and amplitude of the minimum point of the bio-signal, the area of the bio-signal waveform, the time elapse of the bio-signal, and the bio-signal. Amplitude and time information of a component pulse waveform to be configured, and at least one of a division point between two or more feature points, but is not limited thereto.

5A illustrates a pulse wave signal among biosignals acquired from a subject. Referring to FIG. 5A , an example in which the feature point extraction unit 410 extracts a feature point from a pulse wave signal will be described.

In general, a pulse wave signal may be composed of a superimposition of a propagation wave departing from the heart and heading toward an extremity of the body and a reflection wave returning from the extremity. 5A illustrates that the waveform of the pulse wave signal PS obtained from the subject is composed of five constituent pulses, eg, a traveling wave fw and a reflected wave rw1, rw2, rw3, rw4 overlapped.

The feature point extraction unit 410 may extract feature points by analyzing the component pulse waveforms fw, rw1, fw2, rw3, and rw4 in the pulse wave signal PS. In general, component pulses up to the third are mainly used to estimate blood pressure. Subsequent pulses may not be observed depending on the person, and are usually difficult to find due to noise or have low correlation with blood pressure estimation.

For example, the feature point extraction unit 410 determines the time (T ₁ , T ₂ , T ₃ ) and amplitude (P ₁ , P ₂ , P ₃ ) can be extracted as a feature point. At this time, when the pulse wave signal PS is acquired, the feature point extractor 410 secondarily differentiates the obtained pulse wave signal PS and uses the second differential signal to determine the maximum point of the component pulse waveforms fw, rw1, and rw2. Time (T ₁ , T ₂ , T ₃ ) and amplitude (P ₁ , P ₂ , P ₃ ) can be extracted. For example, by searching for local minimum points in the second differential signal, times (T ₁ , T ₂ , T ₃ ) corresponding to the first to third local minimum points are extracted, and the pulse wave signal (PS) ), the amplitudes (P ₁ , P ₂ , P ₃ ) corresponding to the time (T ₁ , T ₂ , T ₃ ) can be extracted. Here, the local minimum point means a point in which the signal decreases and then increases around a specific point when a partial section of the second order differential signal is observed, that is, has a downward convex shape.

As another example, the feature point extractor 410 may extract the time (T _max ) and amplitude (P _max ) of the point where the amplitude is maximum in a predetermined section of the pulse wave signal PS as the feature point. In this case, the predetermined section may indicate a point from the beginning of the pulse wave signal (PS) indicating a systolic section of blood pressure to a point where a dicrotic notch (DN) occurs.

As another example, the feature point extractor 310 may extract the time elapsed (PPG _dur ) or the area of the bio signal waveform (PPG _area ), which means the entire measurement time of the bio signal, as the feature point. In this case, the area of the bio-signal waveform may mean the entire area of the bio-signal waveform or the area of the bio-signal waveform corresponding to a predetermined ratio (eg, 70%) of the entire time elapsed (PPG _dur ).

As another example, the feature point extraction unit 410 may extract a branch point between two or more extracted feature points as an additional feature point. An unstable waveform may be generated in the pulse wave signal due to a non-ideal environment such as motion noise or sleep, and feature points may be extracted at an incorrect position. In this way, the blood pressure measurement can be supplemented by utilizing the internal division between the erroneously extracted feature points.

For example, when the feature point extraction unit 410 first extracts feature points (T ₁ , P ₁ ) and (T _max , P _max ) in the systolic period of blood pressure, the two feature points (T ₁ , P ₁ ) and (T _max , P ) are extracted. _max ) can _be calculated _. At this time, the feature point extractor 410 assigns weights to the time values T ₁ and T _max of the two feature points (T ₁ , P ₁ ) and (T _max , P _max ), and uses each weighted time value The time (T _sys ) of the in-between point is calculated, and the amplitude (P _sys ) corresponding to the calculated time (T _sys ) of the in-between point may be extracted. However, it is not limited thereto, and through the analysis of the acquired biosignal waveform, the feature points (T ₁ , P ₁ ), ( _{T 2} , P ₂ ), the internal branch point between the third and fourth component pulse waveforms (rw _2, rw ₃ ) in the diastolic period of blood pressure (T ₃ , P ₃ ), and the internal branch point between (T ₄ , P ₄ ) more can be calculated.

The feature extraction unit 420 may extract cardiovascular features such as cardiac output and/or total vascular resistance by combining feature points extracted from various biosignals. As an example, referring to FIG. 5B , the feature extraction unit 420 converts the amplitude P ₃ extracted from the area of the biosignal (PPG _area ) and the third component pulse waveform (rw ₂ ) to the maximum amplitude of the blood pressure systolic section ( P _max ) or the value divided by the amplitude of the inner branch (P _sys ), the value divided by the maximum amplitude of the blood pressure systolic section (P _max ) by the area of the biosignal waveform (PPG _area ), the reciprocal of the time elapsed time of the biosignal (PPG _dur ) etc. can be extracted as cardiac output characteristics.

As another example, the feature extractor 420 may extract the time (T ₁ ) from the first component pulse waveform (fw) at the time (T ₃ ) extracted from the third component pulse waveform (rw ₂ ), or the time of the inner branch point ( T _sys ) or the reciprocal of the value minus the time (T _max ) corresponding to the maximum amplitude (P _max ) of the blood pressure systolic interval, the first component pulse at the time (T ₂ ) extracted from the second component pulse waveform (rw ₁ ) The reciprocal of the time (T ₁ ) extracted from the waveform (fw), and the amplitude (P ₂ , P ₃ ) extracted from the second or third component pulse waveforms (rw ₁ , rw ₂ ) to the first component pulse waveform value divided by the amplitude (P ₁ ) extracted from (fw), the value obtained by dividing the amplitude (P ₃ ) extracted from the third component pulse waveform (rw ₂ ) by the maximum amplitude (P _max ), or the area of the biosignal (PPG _area ) can be extracted as characteristics of total vascular resistance. However, it is not limited to these examples, and cardiac output characteristics or total vascular resistance characteristics may be extracted by a combination of various other feature points.

The scaling unit 430 may scale the cardiovascular feature extracted by the feature extraction unit 420 into a first cardiovascular feature and a second cardiovascular feature in consideration of a change tendency of the cardiovascular feature. In this case, the first cardiovascular feature may be a cardiovascular feature for estimating systolic blood pressure, and the second cardiovascular feature may be a cardiovascular feature for estimating diastolic blood pressure. The scaling unit 430 may determine the degree of scaling of the first cardiovascular feature and the second cardiovascular feature according to the type (eg, cardiac output or total vascular resistance) and change tendency (eg, increase or decrease) of the extracted cardiovascular feature. . In this case, the degree of scaling may be predefined for each user through a preprocessing process.

For example, as described above, when cardiac output increases, cardiac output characteristics have a greater effect on systolic blood pressure than diastolic blood pressure. Therefore, referring to FIG. 5C , for a section (x>1) in which cardiac output increases, when the first cardiac output scaling degree is defined by the formula y=x, the second cardiac output scaling degree is the first cardiac output volume It can be defined as any one of y = x, y = 2 * sqrt(x) -1, y = ln (x) and y = sqrt (x) to be less than or equal to the increase rate of . In addition, the degree of scaling of the first cardiac output and the second cardiac output may be defined by the same equation (eg, y=x) for a period in which cardiac output decreases (0≤x≤1). However, the illustrated function formulas are only examples and are not particularly limited. Meanwhile, in FIG. 5C , the x-axis represents the cardiac output before scaling, and the y-axis represents the cardiac output after scaling.

As another example, as described above, when the total vascular resistance decreases, the total vascular resistance has a greater effect on the diastolic blood pressure than the systolic blood pressure. Therefore, referring to FIG. 5D, when the first degree of scaling of total vascular resistance is defined by the formula y=x for the period in which total vascular resistance decreases (0≤x≤1), the degree of scaling of second total vascular resistance is It may be predefined as any one of y=x, y=2*x-1, y=-1/x+2, and y=ln(x)+1 so as to be greater than or equal to the reduction rate of the first total vascular resistance. In addition, the degree of scaling of the first total vascular resistance and the second total vascular resistance for the period (x>1) in which the total vascular resistance increases may be defined by the same equation (eg, y=x). However, the illustrated function formulas are only examples and are not particularly limited. Meanwhile, in FIG. 5D, the x-axis represents the total vascular resistance before scaling, and the y-axis represents the total vascular resistance after scaling.

When the cardiovascular feature is scaled by the scaling unit 430 into the first cardiovascular feature and the second cardiovascular feature, the blood pressure estimator 440 estimates the systolic blood pressure model and the diastolic blood pressure, respectively, for the first cardiovascular feature and the second cardiovascular feature. By inputting into the model, systolic and diastolic blood pressures can be estimated independently. For example, the systolic blood pressure estimation model and the diastolic blood pressure estimation model may be in the form of function expressions such as Equations 2 to 4 below. However, it is not limited thereto.

Equation 2 shows a function expression as an example of a blood pressure estimation model. Here, SBP _est means the estimated systolic blood pressure. F _{CO_SBP} is the scaled first cardiac output, F _{TPR_SBP} is the scaled first total vascular resistance, A _{CO_SBP} is the first cardiac output weight for systolic blood pressure, and A _{TPR_SBP} is the first total vascular resistance weight for systolic blood pressure. do. B _SBP means systolic blood pressure offset. Also, DBP _est means an estimated diastolic blood pressure. F _{CO_DBP} is a scaled second cardiac output, F _{TRP_DBP} is a scaled second total vascular resistance, A _{CO_DBP} is a second cardiac output weight for diastolic blood pressure, and A _{TPR_DBP} is a second total vascular resistance weight for diastolic blood pressure. do. B _DBP means diastolic blood pressure offset.

As illustrated in Equation 2, the blood pressure estimator 440 may estimate the systolic blood pressure after assigning a first cardiac output weight and a second cardiac output weight to the scaled first cardiac output and first total vascular resistance, respectively. there is. Similarly, the blood pressure estimator 440 may estimate the systolic blood pressure after assigning a first cardiac output weight and a second cardiac output weight to the scaled first cardiac output and first total vascular resistance, respectively.

Equation 3 shows a function expression as another example of a blood pressure estimation model. ρ _SBP means a scaling factor for systolic blood pressure to control the overall scaling of cardiovascular features, and ρ _DBP means a scaling factor for diastolic blood pressure to control the overall scaling of cardiovascular features.

Equation 4 shows a function expression of another example of a blood pressure estimation model. Here, A _SBP is a coefficient for systolic blood pressure, and A _DBP is a coefficient for diastolic blood pressure.

6 is a flowchart of a blood pressure estimation method according to an embodiment.

FIG. 6 is an example of the method for estimating blood pressure according to the embodiment of FIG. 1 or FIG. 2 . Since it has been described in detail above, it will be briefly described below to avoid redundant description.

First, the blood pressure estimating device may receive a blood pressure estimation request (610). The blood pressure estimating apparatus may provide an interface to a user and receive a blood pressure estimation request input by the user through the interface. Alternatively, the blood pressure estimating device may communicate with an external device and receive a blood pressure estimation request from the external device. In this case, the external device may be a smartphone or a tablet PC carried by the user, and the user may control the operation of the blood pressure estimating device through a device having superior interface performance or computing performance.

Next, the blood pressure estimating apparatus may acquire a biosignal from the subject by controlling an internally mounted sensor or receive a biosignal from an external sensor for estimating blood pressure (620). At this time, the sensor installed in the blood pressure estimating device and the external sensor may acquire various biosignals such as PPG signal, ECG signal, EMG signal, and BCG signal from various parts (eg, wrist, chest, finger, etc.) of the subject. .

Next, feature points may be extracted by analyzing the acquired bio-signals (630). At this time, the feature points constitute the heartbeat information, the shape of the bio-signal waveform, the time and amplitude of the maximum point of the bio-signal, the time and amplitude of the minimum point of the bio-signal, the area of the bio-signal waveform, the time elapse of the bio-signal, and the bio-signal. Amplitude and time information of a component pulse waveform to be configured, internal division points between two or more feature points, and the like may be included.

Next, cardiovascular features may be extracted based on the extracted feature points (640). In this case, the cardiovascular characteristics may include cardiac output characteristics and total vascular resistance characteristics. For example, the apparatus for estimating blood pressure may extract cardiovascular features by using the extracted feature points as they are or by combining two or more feature points as shown in FIG. 5B.

Next, the apparatus for estimating blood pressure may scale the extracted cardiovascular feature as a first cardiovascular feature for estimating systolic blood pressure (650). In this case, scaling may be performed to the first cardiac output feature and the first total vascular resistance feature based on a predefined scaling degree in consideration of the correlation between the change tendency of each of the cardiac output feature and the total vascular resistance feature on the systolic blood pressure.

Also, the apparatus for estimating blood pressure may scale the extracted cardiovascular feature as a second cardiovascular feature for estimating the diastolic blood pressure (660). In this case, scaling may be performed to the second cardiac output feature and the second total vascular resistance feature based on a predefined scaling degree in consideration of the correlation between the change tendency of each of the cardiac output feature and the total vascular resistance feature on the diastolic blood pressure.

For example, when cardiac output increases, the cardiac output feature has a greater influence on systolic blood pressure than on diastolic blood pressure, so when the cardiac output feature tends to increase, the second cardiac output feature is dependent on the first cardiac output feature. It can be scaled so that the increase rate is the same or smaller than that. In addition, when the total vascular resistance decreases, the total vascular resistance characteristic has a greater effect on the diastolic blood pressure than the systolic blood pressure, so when the total vascular resistance characteristic tends to decrease, the second total vascular resistance is the first total vascular resistance characteristic. It can be scaled so that the reduction rate is greater than or equal to .

Next, the blood pressure estimating apparatus may estimate the systolic blood pressure using the first cardiovascular feature (670) and the diastolic blood pressure using the second cardiovascular feature (680). For example, the systolic blood pressure and the diastolic blood pressure may be independently estimated by inputting the first cardiovascular feature and the second cardiovascular feature into the systolic blood pressure estimation model and the diastolic blood pressure estimation model as shown in Equations 2 to 4 above.

Then, the blood pressure estimating apparatus may output and provide additional information such as a warning according to the obtained bio-signal, blood pressure estimation result, and blood pressure estimation result to the user (690). For example, the blood pressure estimation result is visually output through the display module, and when the estimated blood pressure is displayed, if the estimated blood pressure is out of the normal range, it is emphasized by displaying it in red to display warning information to the user. can Alternatively, the blood pressure estimation result may be output as a voice through the speaker module. At this time, if the estimated blood pressure is out of the normal range, the user may be notified of a health condition abnormality through vibration or tactile sensation through the haptic module.

7A and 7B are diagrams for describing a wearable device according to an exemplary embodiment. As illustrated, various embodiments of the above-described blood pressure estimating device may be mounted on a smart watch worn on a wrist or a smart band-type wearable device. However, since this is only one example for convenience of description, the shape of the wearable device is not particularly limited thereto.

Referring to FIGS. 7A and 7B , a wearable device 700 may include a device body 710 and a strap 720 .

The strap 720 may be configured to be flexible, and may be bent in a form wrapped around the user's wrist or bent in a form separated from the user's wrist. Alternatively, the strap 720 may be configured in the form of a non-separable band. In this case, the strap 720 may have air injected therein to have elasticity according to a change in pressure applied to the wrist, or may include an air bag, and may transmit the change in pressure of the wrist to the main body 710 .

A battery for supplying power to the wearable device may be embedded inside the body 710 or the strap 720 .

In addition, one or more sensors for measuring various bio-signals may be mounted inside the main body 710 . For example, the pulse wave sensor 711 may be mounted on the rear surface of the main body 710 in contact with the object OBJ, for example, a wrist, in a form exposed to the object OBJ. The pulse wave sensor 711 may include a light source 711a for radiating light to the object OBJ and a detector 711b for measuring a pulse wave signal by detecting light scattered or reflected from the object OBJ. there is. In this case, the light source 711a may include at least one of a light emitting diode (LED), a laser diode, and a phosphor, and may be formed in one or two or more arrays. Light sources formed of two or more arrays may be formed to emit light of different wavelengths. In addition, the detector may include a photodiode, an image sensor, and the like, and may be formed in one or two or more arrays.

A main body 710 of the wearable device 700 may be equipped with a pulse wave sensor 711 and/or a processor 712 for estimating blood pressure based on a biosignal received from an external sensor. The processor 712 may control the pulse wave sensor 711 by generating a control signal according to a user's request for estimating blood pressure, and, if necessary, control the communication unit 713 to receive a biosignal from an external sensor.

The communication unit 713 may be mounted inside the main body 710 and communicate with an external device under the control of the processor 712 to transmit/receive necessary information. For example, the communication unit 713 may receive a biosignal from an external sensor that measures a biosignal, such as an ECG sensor, an EMG sensor, or a BCG sensor. Also, a request for estimating blood pressure may be received from the user's portable terminal. In addition, the extracted feature point or feature information may be transmitted to an external device to estimate the blood pressure. In addition, the result of blood pressure estimation can be transmitted to an external device to be displayed to the user, or used for various purposes such as blood pressure history management and disease research. In addition, reference information such as a blood pressure estimation model or a scaling formula for each cardiovascular feature may be received from an external device.

When a biosignal is received from the pulse wave sensor 711 and/or an external sensor, the processor 712 may extract feature points from the received biosignal. For example, when a pulse wave signal is received, the component pulse waveforms constituting the pulse wave signal are analyzed as described above, and the time and amplitude of the maximum point of each component pulse waveform, the time and amplitude of the point where the amplitude is maximum in the systolic period of blood pressure, and Amplitude, time lapse of the pulse wave signal, area of the pulse wave signal waveform, etc. may be extracted as feature points.

When the feature points are extracted, the processor 712 may combine the feature points to extract cardiovascular features, such as cardiac output features and total vascular resistance features, as illustrated in FIG. 5B .

In addition, the processor 712 scales the cardiovascular feature into a cardiovascular feature for systolic blood pressure and a cardiovascular feature for diastolic blood pressure, and independently estimates the systolic blood pressure and the diastolic blood pressure using the cardiovascular feature for systolic blood pressure and the cardiovascular feature for diastolic blood pressure, respectively. can In this case, the degree of scaling of the cardiovascular characteristics for systolic blood pressure and the degree of scaling of the cardiovascular characteristics for diastolic blood pressure are determined in advance by considering the correlation between the change tendency of the cardiovascular characteristics, for example, the cardiac output characteristics and the total vascular resistance characteristics, on the systolic blood pressure and the diastolic blood pressure. can be defined

The processor 712 may manage blood pressure estimation results, blood pressure history information, bio signals used to measure blood pressure, extracted feature points, cardiovascular features before and after scaling, and the like in a storage device.

The wearable device 700 may further include a control unit 715 and a display unit 714 mounted on the main body 710 .

The control unit 715 may receive a user's control command and transmit it to the processor 712, and may include a power button for inputting a command for turning on/off the power of the wearable device 700.

The display unit 714 may provide a user with various information related to the detected blood pressure under the control of the processor 712 . For example, the display unit 714 may display additional information such as detected blood pressure, alarm, and warning to the user in various visual/non-visual ways.

Meanwhile, the present embodiments can be implemented as computer readable codes in a computer readable recording medium. The computer-readable recording medium includes all types of recording devices in which data that can be read by a computer system is stored.

Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc., and implementation in the form of a carrier wave (for example, transmission over the Internet) include In addition, the computer-readable recording medium may be distributed to computer systems connected through a network, so that computer-readable codes may be stored and executed in a distributed manner. In addition, functional programs, codes, and code segments for implementing the present embodiments can be easily inferred by programmers in the art.

Those of ordinary skill in the art to which this disclosure belongs will understand that it can be implemented in other specific forms without changing the disclosed technical idea or essential features. Therefore, the embodiments described above should be understood as illustrative in all respects and not limiting.

100, 200: blood pressure estimation device 110: acquisition unit
120, 400: processor 210: communication unit
220: output unit 230: storage unit
410: feature point extraction unit 420: feature extraction unit
430: scaling unit 440: blood pressure estimation unit
700: wearable device 710: body
711 pulse wave sensor 711a light source
711b: detector 712: processor
713: communication unit 714: display unit
715: control unit 720: strap

Claims (33)

an acquisition unit that acquires bio-signals of the subject; and
A processor for extracting a cardiovascular feature based on the bio-signal and independently estimating a systolic blood pressure (SBP) and a diastolic blood pressure (DBP) in consideration of a change tendency of the extracted cardiovascular feature compared to a reference cardiovascular feature; ,
The processor
a scaling unit for scaling the extracted cardiovascular features into first cardiovascular features and second cardiovascular features; and
a blood pressure estimator configured to estimate a systolic blood pressure based on the first cardiovascular feature and a diastolic blood pressure based on the second cardiovascular feature;
The scaling part
The apparatus for estimating blood pressure for determining scaling degrees of a first cardiovascular feature and a second cardiovascular feature according to the type of the extracted cardiovascular feature and the change tendency. According to claim 1,
The biosignal is
A blood pressure estimating device including at least one of photoplethysmogram (PPG), electrocardiography (ECG), electromyography (EMG), and ballistocardiogram (BCG). According to claim 2,
the acquisition unit
A blood pressure estimating device comprising one or more sensors for measuring at least a portion of the bio-signals. According to claim 2,
The blood pressure estimating apparatus further includes a communication unit configured to receive a biosignal from one or more external sensors that measure at least a portion of the biosignal and transfer the received biosignal to the acquisition unit. According to claim 1,
The processor
A blood pressure estimating device comprising a feature extractor for extracting a cardiovascular feature including at least one of cardiac output and total periphral resistance based on a characteristic point extracted from the biosignal. According to claim 5,
The processor
Heart rate information from the bio-signal, shape of the waveform of the bio-signal, time and amplitude of the maximum point of the bio-signal, time and amplitude of the minimum point of the bio-signal, area of the bio-signal waveform, The blood pressure estimating apparatus further comprising a feature point extraction unit that extracts a feature point including at least one of amplitude and time information of a component pulse waveform. delete delete According to claim 1,
The cardiovascular feature includes at least one of cardiac output and total periphral resistance. an acquisition unit that acquires bio-signals of the subject; and
A processor for extracting a cardiovascular feature based on the bio-signal and independently estimating systolic blood pressure and diastolic blood pressure in consideration of a change tendency of the extracted cardiovascular feature compared to a reference cardiovascular feature;
The processor
a scaling unit for scaling the extracted cardiovascular features into first cardiovascular features and second cardiovascular features; and
a blood pressure estimator configured to estimate a systolic blood pressure based on the first cardiovascular feature and a diastolic blood pressure based on the second cardiovascular feature;
The cardiovascular characteristics
including cardiac output and total periphral resistance,
The scaling part
scaling the cardiac output to a first cardiac output and a second cardiac output, and scaling the total vascular resistance to a first total vascular resistance and a second total vascular resistance;
The blood pressure estimation unit
The blood pressure estimating apparatus for estimating systolic blood pressure by linearly combining the first cardiac output and the first total vascular resistance, and estimating the diastolic blood pressure by linearly combining the second cardiac output and the second total vascular resistance. According to claim 10,
The blood pressure estimation unit
After assigning a first cardiac output weight and a first total vascular resistance weight to the first cardiac output and the first total vascular resistance, respectively, linearly combining them;
A blood pressure estimating apparatus for linearly combining the second cardiac output weight and the second total vascular resistance weight after assigning a second cardiac output weight and a second total vascular resistance weight to the second cardiac output and the second total vascular resistance, respectively. According to claim 10,
The blood pressure estimation unit
estimating systolic blood pressure by applying a predetermined first scaling factor to a result of linearly combining the first cardiac output and the first total vascular resistance;
A blood pressure estimating device for estimating diastolic blood pressure by applying a predetermined second scaling factor to a result of linearly combining the second cardiac output and the second total vascular resistance. delete According to claim 1,
The scaling part
If the type of cardiovascular feature is cardiac output and the change tendency is increasing, the blood pressure estimating device equalizes or reduces the increase rate of the second cardiac output compared to the increase rate of the first cardiac output according to the increase in the cardiac output. According to claim 1,
The scaling part
When the type of cardiovascular feature is total vascular resistance and the change tendency is a decrease, the blood pressure estimating device for equalizing or increasing a reduction rate of a second total vascular resistance compared to a reduction rate of the first total vascular resistance according to a decrease in the total vascular resistance. According to claim 1,
The blood pressure estimating device further comprises an output unit outputting a blood pressure estimation result of the processor. A blood pressure estimation device,
obtaining bio-signals of the subject;
extracting one or more cardiovascular features based on the biosignal; and
Independently estimating systolic blood pressure (SBP) and diastolic blood pressure (DBP) in consideration of a change trend of the extracted cardiovascular feature compared to a reference cardiovascular feature,
The step of estimating the blood pressure is
scaling the extracted cardiovascular features into a first cardiovascular feature and a second cardiovascular feature; and
estimating a systolic blood pressure based on the first cardiovascular characteristic and estimating a diastolic blood pressure based on the second cardiovascular characteristic;
The scaling step is
A blood pressure estimation method for determining scaling degrees of a first cardiovascular feature and a second cardiovascular feature according to the type of the extracted cardiovascular feature and the change tendency. According to claim 17,
The biosignal is
A blood pressure estimation method including at least one of photoplethysmogram (PPG), electrocardiography (ECG), electromyography (EMG), and ballistocardiogram (BCG). According to claim 17,
The step of extracting the cardiovascular features
A blood pressure estimation method for extracting cardiovascular features including at least one of cardiac output and total vascular resistance based on characteristic points extracted from the biosignal. According to claim 19,
Heartbeat information from the bio-signal, shape of the waveform of the bio-signal, time and amplitude of the maximum point of the bio-signal, time and amplitude of the minimum point of the bio-signal, area of the bio-signal waveform, and time elapse of the bio-signal , extracting a feature point including at least one of amplitude and time information of a component pulse waveform constituting the bio-signal, and a branching point between two or more extracted feature points. delete delete delete According to claim 17,
The scaling step is
If the type of the extracted cardiovascular feature is cardiac output and the change tendency is increasing, equalizing or reducing the increase rate of the second cardiac output compared to the increase rate of the first cardiac output according to the increase in the cardiac output blood pressure estimation method. According to claim 17,
The scaling step is
If the type of the extracted cardiovascular feature is total vascular resistance and the change tendency is a decrease, blood pressure estimation that equalizes or increases the decrease rate of the second total vascular resistance compared to the decrease rate of the first total vascular resistance according to the decrease in the total vascular resistance method. According to claim 17,
The blood pressure estimation method further comprising outputting the blood pressure estimation result. a main body worn on the subject;
straps connected to both ends of the main body and fixing the main body to the subject while wrapping the subject; and
a sensor mounted on the main body and measuring a biosignal from the subject; and
Extracting one or more cardiovascular features based on the measured bio-signal, and estimating systolic blood pressure (SBP) and diastolic blood pressure (DBP) independently by considering the change tendency of the extracted cardiovascular feature compared to the reference cardiovascular feature contains a processor;
The processor
Scaling the extracted cardiovascular feature into a first cardiovascular feature and a second cardiovascular feature, estimating a systolic blood pressure based on the first cardiovascular feature and estimating a diastolic blood pressure based on the second cardiovascular feature, A wearable device that determines scaling degrees of a first cardiovascular feature and a second cardiovascular feature according to the type and the change tendency. The method of claim 27,
The sensor
A wearable device including at least one of a photoplethysmogram (PPG) sensor, an electrocardiography (ECG) sensor, an electromyography (EMG) sensor, and a ballistocardiogram (BCG) sensor. The method of claim 27,
The wearable device of claim 1 , wherein the cardiovascular characteristic includes at least one of cardiac output and total periphral resistance. delete delete The method of claim 27,
The wearable device further comprising a display unit outputting a blood pressure estimation result of the processor. The method of claim 27,
The wearable device further comprising a communication unit transmitting the blood pressure estimation result of the processor to an external device.

Images