KR20090081885A - Method and apparatus for measuring heart rate - Google Patents

Abstract

An apparatus for detecting heart and controlling method thereof are provided to shorten the amount of memory and calculation time and quickly and accurately detect peak point. A method for controlling a detection apparatus of heart comprises: a step of detecting bio signal(100); a step of storing the detected bio signal in a window unit which is a regular data size(110); a step of confirming the cycle of bio signal by applying threshold in a window(170); a step of extracting peak point of bio signal(200); and a step of calculating variation of heart(220).

Description

Heart rate detection device and control method {METHOD AND APPARATUS FOR MEASURING HEART RATE}

The present invention relates to an apparatus for detecting heart rate and a control method thereof, and more particularly, to an apparatus for detecting heart rate and a method of controlling the same, capable of effectively detecting heart rate variability (HRV).

In general, the heartbeat detection apparatus acquires a biosignal through a biosignal detection sensor in contact with a human body, detects a peak value, and calculates a heart rate variability from a time interval of the peak values. This heart rate variability can be used to determine the emotional state of the examiner using a Fast Fourier Transform (FFT).

As a conventional heart rate detection device, as disclosed in Korean Patent Laid-Open Publication No. 2003-81903, after acquiring a PPG signal from a volume pulse sensor part, a signal processing method based on wavelet transform and an average of an autocorrelation function The heart rate calculation is used to detect the repetitive heart rate sequence in the PPG signal, and extract the optimal heart rate sequence or average heart rate from the candidate heart rate sequences.

The above publication uses a wavelet transform method to remove harmonics, clip a value above a certain level based on a threshold value to obtain only a signal component related to the heartbeat from a high frequency component-removed PPG signal, The method of detecting peaks by calculating correlation function values is used.

However, the wavelet transform method itself occupies a large amount of memory in signal processing, and it is difficult to perform fast signal processing, which requires a lot of computation time. In addition, clipping of a signal using a fixed threshold has a problem in that it is difficult to detect an accurate peak point because a biosignal has a different period and a peak according to a person. This lowers the probability of detecting a true heart rate variability.

Accordingly, an object of the present invention is to detect a peak point using a periodic detection method instead of a wavelet method to reduce the amount of memory and computation time and to detect the peak point more accurately, and a control method thereof. To provide.

The heart rate detection method of the present invention for achieving the above object is a step of detecting a bio-signal, storing the detected bio-signal in a window unit of a certain data size, by applying a threshold value in the window the bio-signal The method may include checking a period of time, extracting a peak point of the biosignal in the identified period, and calculating a heart rate variability using time information of the extracted peak point.

In addition, the heart rate detection apparatus of the present invention is a bio-signal detection unit for detecting a user's bio-signal, a storage unit for storing the detected bio-signal in a window unit of a certain data size, and the detected bio-signal to be the window Control to store in the storage unit, confirm the period of the bio-signal in the window, extract the peak point of the bio-signal in the identified period, and calculate the heart rate variability using the extracted peak point. It characterized in that it comprises a control unit for performing.

According to the present invention, the peak point of the biosignal is obtained using a conventional wavelet transform method by acquiring a biosignal from a user, finding a peak point of the biosignal using a periodic detection method, and calculating a reliable heart rate variability therefrom. The memory usage and computation time can be reduced, and the response to the change in the biosignal is faster and the peak point can be detected more quickly and accurately than the finding method.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings.

1 is a control block diagram of a heartbeat detection apparatus according to an embodiment of the present invention. As shown in FIG. 1, an apparatus for detecting heart rate according to an embodiment of the present invention includes an electrocardiogram sensor 10 and an optical-only pulse wave sensor 11 as biosignal detection units 10 and 11 for detecting a biosignal from a user; It includes a control unit 20, the monitoring unit 30 and the storage unit 40.

The ECG sensor 10 detects an electrocardiogram (ECG) signal representing an electrical activity of the heart that occurs during the heartbeat cycle.

The photoelectric pulse wave sensor 11 uses the principle of changing the absorption and reflection of light according to the change in the vascular proliferation according to the heartbeat, and is irradiated to the user's body from the light emitting unit and the light emitting unit that emit infrared light. It consists of a light-receiving unit that detects light, and detects a photoplethysmography (PPG) signal from a change in light blood flow with time output from the light-receiving unit.

The control unit 20 includes an A / D signal acquisition unit 21, a signal processing unit 22, and a calculation unit 23.

The A / D signal acquisition unit 21 is for A / D processing the bio signals from the ECG sensor 10 and the optical dedicated pulse wave sensor 11, and converts the bio signals in the form of analog signals into digital signals, thereby sampling frequency. The higher is, the more signals can be acquired.

The signal processor 22 processes the biosignal acquired by the A / D signal acquisition unit 21 and performs signal processing for detecting peak points according to the period.

The calculation unit 23 calculates the heart rate variability from the peak point detected by the signal processing unit 22.

The control unit 20 having the above components acquires bio signals such as an ECG signal and a PPG signal through the A / D signal acquisition unit 21.

The control unit 20 stores the data in the storage unit 40 until the data size of each biosignal becomes a predetermined size (window).

Thereafter, the control unit 20 performs an operation for removing noise of the biosignal through the signal processing unit 22, a signal processing for checking a period, and a signal processing for finding peak points within the identified period. .

In this way, if the peak points are found within the period, the control unit 20 calculates the heart rate variability from the time information at which the peak points occurred through the calculation unit 23 in itself.

The control unit 20 displays a series of monitoring results for the bio-signal through the monitoring unit 30.

Looking at the operation of the control unit 20 in more detail, the control unit 20 detects the bio-signal, stores until the detected bio-signal is a window of a certain data size, and confirms the period of the bio-signal in the window . At this time, the average value of the peak points extracted from the previous window is calculated, a value of 1 / N of the calculated average value is set as a threshold value, and a zero crossing point of the biosignal is detected and detected based on the set threshold value. The time index of the zero crossing points is stored, and the time interval between neighboring time indices among the stored time indices is determined as a period of the biosignal.

After checking the period, the controller 20 extracts the peak point of the biosignal in the identified period, and calculates the heart rate variability using the time information of the extracted peak point. At this time, if the period value is equal to or greater than the average period value by comparing the checked period value with the average period value of the previous window, it is determined that the checked period is a normal period, otherwise it is determined to be an abnormal period to check the period again. In addition, when the identified period is a normal period, the peak point of the biosignal is extracted from the identified period, and the heart rate variability is calculated using the time information of the extracted peak point.

Figure 2 shows a control flow chart for the heart rate detection method according to an embodiment of the present invention. Referring to FIG. 2, first, the controller 20 detects a biosignal every predetermined time (100). In the case of the ECG, the biosignal is represented by a waveform as shown in FIG. 3.

The controller 20 stores the data in the storage unit 40 until the data size of the detected biosignal becomes a predetermined size (110). If the predetermined size is defined as a window, signal processing is performed when the data size of the detected biosignal is stored as much as the window (120). As illustrated in FIG. 4, the controller 20 acquires one piece of data of a bio signal for each channel every 4 ms. If one data is acquired every 4ms and the number of data becomes 512, for example, signal processing is performed. As will be described later, after the signal processing, the peak points are extracted within one window, and when the total extracted peak points is 32, for example, FFT is performed to calculate the heart rate variability.

As a method for signal processing the acquired data, a method of detecting a period of a biosignal and extracting peak points within the period is used. In order to do so, a period of the biosignal must be identified within a window. As shown in FIG. 5, the confirmation of the period determines the point of time when the sign of the biosignal is changed from (−) to (+) as a zero crossing point as a zero crossing point. In particular, when the sign of the biological signal changes from (-) to (+), it is determined that it is a rising edge. Based on the time index at this time, it is determined that it is one cycle until (+) is detected in the next (-). One period at this time means the rising edge to the next rising edge.

However, as shown in FIG. 5, when the threshold is set low, a plurality of zero crossing points are generated within one period. Therefore, a method of excluding these points so that the remaining zero crossing points Δ are not detected except the zero crossing point X in the normal period. Accordingly, it is possible to exclude the invalid zero crossing points Δ with small peak points, and extract only the valid zero crossing points X.

In other words, it is necessary to pass only a value greater than or equal to a certain value and actually detect only the maximum value within the period. To this end, the threshold uses the data of the previous window, and averages the magnitude values of the actual peak points extracted from the previous window (130), and sets the value of 1 / N of the average value as the threshold (140). At this time, the above average value cannot be obtained at the time of the first control, so the preset value is set as the threshold. Where N is a preset constant.

After setting the threshold, the threshold is applied to the biosignal stored in the storage 40 (150). When the threshold is applied to the biosignal to subtract the threshold from the biosignal, the threshold value becomes the "0" level of the biosignal, and the zero crossing points Δ that are not valid are "0" as shown in FIG. The value is below the level. Such a threshold setting is more effective in extracting a valid zero crossing point X even though the size of the peak point of the biosignal differs from user to user, because it is not a fixed threshold but a floating threshold according to a previous peak point.

After applying the threshold to the biosignal, a zero crossing point of the biosignal is extracted based on the threshold, and the period of the biosignal is checked using various zero crossing points extracted in the current window (170). That is, the difference between time indexes of neighboring zero crossing points is calculated and determined as a period.

If the cycle is confirmed, it is determined whether the checked cycle is a normal cycle (180). That is, the average period value in the previous window is calculated, and the calculated period value is compared with the calculated period value in the current window, and it is determined that the identified period value is a period of only 1 / M or more of the average period value. do. Where M is a preset constant. Then, if the checked period is a normal period, the zero crossing point in the period is extracted, and the time index of the extracted zero crossing point is stored in the storage 40 (190).

In this state, the maximum value of the signal between the time index [n + 1] and the time index [n] is extracted as the peak point within a period by using the time index of the zero crossing point stored in this state. Also, from the last zero crossing point to the end of the window, the largest value of the biosignal is stored, compared to the first zero crossing point of the next window, and the largest value in the period is extracted as the peak point. That is, when the window is changed from the previous window to the current window, since the maximum point from the start point of the window can be determined as the peak point, there is a fear that the peak point is incorrect even though it is not the actual peak point.

To prevent this, the time index between the last zero crossing point of the biosignal in the previous window and the starting zero crossing point of the biosignal in the current window is stored, and the maximum value of the corresponding interval is greater than the average value using the average value of the previous peak points. Ensure that only points are extracted as peak points.

In other cases, the maximum value is extracted as a peak point by comparing data between time exponents of zero crossing points.

Then, the time index of the peak points obtained from the above-described signal processing is stored in the storage 40 (210), and the heart rate variability is calculated using these differences (220).

Thereafter, as shown in FIG. 7, the monitoring result of the biosignal is displayed through the monitoring unit 30.

1 is a control block diagram of a heart rate detection apparatus according to an embodiment of the present invention.

2 is a control flowchart of a heart rate detection method according to an embodiment of the present invention.

3 is a diagram illustrating a biosignal detected in FIG. 2.

FIG. 4 is a diagram illustrating that the biosignal detected in FIG. 2 is stored in units of windows.

FIG. 5 is a diagram for describing checking a period of a biosignal using a zero crossing point in FIG. 2.

FIG. 6 is a diagram for describing peak point detection using a fluid threshold in FIG. 2.

* Description of the symbols for the main functions of the drawings *

10: ECG sensor 11: PPG sensor

20: control unit 21: A / D signal acquisition unit

22: signal processor 23: calculator

30: monitoring unit 40: storage unit

Claims (15)

Detecting a biosignal; Storing the detected biosignals in units of windows having a predetermined data size; Confirming a period of the biosignal by applying a threshold in the window; Extracting peak points of the bio-signals in the identified period; And Computing heart rate variability using the time information of the extracted peak point. The method of claim 1, The determining of the period may include setting the threshold, extracting zero crossing points of the biosignal based on the set threshold in the window, and based on a time index of the extracted zero crossing points. Heartbeat detection method comprising the step of checking the cycle of the bio-signals. The method of claim 2, The setting of the threshold value may include calculating an average value of the peak points detected in the previous window, and setting a value of 1 / N of the calculated average value as the threshold value. The method of claim 2, And the zero crossing point is a point that becomes zero level at the threshold value. The method of claim 2, And determining whether the checked period is a normal period. The method of claim 5, The determining of whether the period is normal includes comparing the checked period value with an average period value of a previous window, and if the period value is greater than or equal to the average period value, determining that the period is a normal period. Characteristic heart rate detection method. The method of claim 6, The extracting of the peak point may include extracting a maximum value of the bio signal as a peak point in the identified period when the identified period is a normal period. The method of claim 7, wherein The extracting of the peak point may include: when the previous window and the current window are changed, the maximum value between the last zero crossing point of the previous window and the starting zero crossing point of the current window is equal to or greater than the average of the maximum values in the previous window. And extracting the maximum value as a peak point. The method of claim 1, And displaying a monitoring result of the detected biosignal after the heart rate variance calculation step. A biosignal detection unit detecting a biosignal of a user; A storage unit for storing the detected biosignal in units of windows having a predetermined data size; And The detected biosignal is stored in the storage unit until it becomes the window, the period of the biosignal is checked in the window, the peak point of the biosignal is extracted in the identified period, and the extracted peak. And a control unit for performing a control to calculate a heart rate variability using points. The method of claim 10, The controller calculates an average value of the peak points extracted from the previous window, sets a value of 1 / N of the calculated average value as a threshold value, extracts a zero crossing point of the biosignal based on the set threshold value, and And storing time indexes of the extracted zero crossing points, and determining a time interval between neighboring time indexes among the stored time indexes as the period of the biosignal. The method of claim 10, The controller may compare the checked period value with an average period value of a previous window, and if the period value is greater than or equal to an average period value, determining the determined period as a normal period. The method of claim 12, The controller may extract the maximum value of the bio-signal as a peak point in the checked period when the checked period is a normal period. The method of claim 10, If the maximum value of the biosignal between the last zero crossing point of the previous window and the start zero crossing point of the current window is equal to or greater than an average value of peak points of the previous window when the previous window and the current window are changed time points, A heart rate detection device, characterized in that for extracting the maximum value as a peak point. The method of claim 10, Heart rate detection device further comprises a monitoring unit for displaying the monitoring result of the detected bio-signals.

Images