TW201519862A - Method and apparatus for monitoring heartbeat signal based on empirical mode decomposition - Google Patents

Abstract

A method and an apparatus for monitoring a heartbeat signal are provided based on an Empirical Mode Decomposition method. The method for monitoring the heartbeat signal includes: receiving the heartbeat signal, acquiring a characteristic wave from the heartbeat signal; performing an adaptive signal decomposition to the acquired characteristic wave, so as to extract a plurality of characteristic wave components; selecting the characteristic wave component which has the highest frequency component to be a test component; calculating a ratio between a root-mean-square (RMS) value of the test component and a RMS value of the characteristic wave; and determining whether the heartbeat signal is an abnormal heartbeat signal according to the evaluation parameter.

Description

Heartbeat signal detection method and detection device based on empirical mode decomposition method

The invention relates to a signal detecting method and a detecting system, and in particular to a heartbeat signal detecting method and detecting system.

With the improvement of living standards, modern people pay more and more attention to their own health concerns and the prevention of major diseases. Heart disease is one of the most common major diseases in modern society and is a common cause of sudden death (Sudden Cardiac Death). In detail, the common cause of sudden cardiac death is acute Myocardial Infarction with Malignant Ventricular Arrhymias, while patients with Ventricular Arrhythmias are at high risk of myocardial infarction. group. Therefore, if the symptoms of ventricular arrhythmia are correctly diagnosed, the patient can be alerted in advance and appropriate treatment can be provided to reduce the risk of sudden cardiac death.

Due to advances in technology, today's physicians often diagnose various heart diseases by electrocardiogram. Traditional ECG provides doctors by recording changes in the potential of cardiomyocytes A reliable basis for judgment to effectively diagnose the patient. Figure 1 is a schematic diagram of an ideal full heartbeat waveform. In more detail, a complete heartbeat waveform represents the change in potential of cardiomyocytes during a myocardial action cycle. As can be seen from Fig. 1, the ideal complete heartbeat waveform includes P wave, QRS wave (wave group formed by Q wave, R wave, and S wave) and T wave, which correspond to atrial depolarization period, ventricular depolarization period, and ventricle, respectively. Repolarization period. In general, the physician can judge whether the subject has certain heart diseases by observing the heartbeat waveform of the subject, but the abnormal potential signal that can be used to determine ventricular arrhythmia usually has a relatively high frequency and a relatively small frequency. Because of the amplitude, it is impossible to correctly judge whether the subject has a sign of ventricular arrhythmia.

The prior art further proposes to correctly detect and determine whether the heartbeat signal is an abnormal heartbeat signal and whether the subject has a ventricular arrhythmia symptom by analyzing the potential changes of the P wave, the QRS wave, and the T wave. How to specifically detect abnormal heartbeat signals is still one of the goals of those skilled in the art.

In view of the above, the present invention provides a heartbeat signal detecting method and a heartbeat signal detecting device, which can detect whether a heartbeat signal has an abnormal potential signal, thereby effectively determining whether the heartbeat signal is an abnormal heartbeat signal.

An exemplary embodiment of the present invention provides a heartbeat signal detecting method, including: receiving a heartbeat signal from a high-resolution electrocardiograph or an average signal electrocardiograph; intercepting a characteristic wave in a heartbeat signal; and adapting the intercepted characteristic wave News Number decomposition to obtain a plurality of characteristic wave components; selecting a characteristic wave component having the highest frequency component from the plurality of characteristic wave components as a test component; calculating a ratio of a root mean square value of the test component to a root mean square value of the characteristic wave to obtain Evaluating parameters; and determining whether the heartbeat signal is an abnormal heartbeat signal according to the evaluation parameter.

An exemplary embodiment of the present invention provides another heartbeat signal detecting apparatus, including a receiving unit, an intercepting unit, an adaptive signal decomposition unit, a selecting unit, a calculating unit, and a determining unit. The receiving unit receives the heartbeat signal from the high-resolution electrocardiograph or the average signal electrocardiograph, and the intercepting unit is coupled to the receiving unit to intercept the characteristic wave in the heartbeat signal. The adaptive signal decomposition unit is coupled to the capture unit to perform adaptive signal decomposition on the characteristic wave to obtain a plurality of characteristic wave components. The selection unit is coupled to the adaptive signal decomposition unit, and selects the characteristic wave component having the highest frequency component from the plurality of characteristic wave components as the test component. The calculating unit is coupled to the intercepting unit to calculate a ratio of the root mean square value of the test component to the root mean square value of the characteristic wave to obtain the evaluation parameter. The determining unit is coupled to the calculating unit, and determines whether the heartbeat signal is an abnormal heartbeat signal according to the evaluation parameter.

Based on the above, the heartbeat signal detecting method provided by the present invention obtains a plurality of characteristic wave components by performing adaptive signal decomposition on the characteristic wave, and selects a characteristic wave component having the highest frequency component from the plurality of characteristic wave components as Test ingredients. The heartbeat signal detection method further calculates the ratio of the root mean square value of the test component to the root mean square value of the characteristic wave to obtain an evaluation parameter, thereby determining whether the heartbeat signal is an abnormal heartbeat signal. Since the abnormal potential signal for judging ventricular arrhythmia is generally relatively high in frequency for the characteristic wave, the present invention provides The heartbeat signal detecting method can correctly parse the test component including the abnormal potential signal, and accurately and effectively judge whether the subject has a symptom of ventricular arrhythmia, thereby effectively overcoming/solving the problems described in the prior art.

The above described features and advantages of the present invention will be more apparent from the following description. However, it is to be understood that the foregoing general description and the claims

P‧‧‧P wave

Q, R, S‧‧‧ QRS waves

T‧‧‧T wave

700‧‧‧ heartbeat signal detection device

710‧‧‧ receiving unit

720‧‧‧ interception unit

730‧‧‧Adaptive signal decomposition unit

740‧‧‧Selection unit

750‧‧‧Computation unit

760‧‧‧judging unit

Steps for S210~260‧‧‧ heartbeat signal detection method

S232~S236‧‧‧Get multiple characteristic wave components

S262~266‧‧‧Steps to determine whether the heartbeat signal is an abnormal heartbeat signal

The following drawings are a part of the specification of the invention, and illustrate the embodiments of the invention

Figure 1 is a schematic diagram of an ideal full heartbeat waveform.

2 is a flow chart of a method for detecting a heartbeat signal according to an exemplary embodiment of the invention.

FIG. 3 is a flow chart of a method for obtaining a plurality of characteristic wave components according to an exemplary embodiment of the invention.

FIG. 4 is a schematic diagram showing characteristic wave and characteristic wave components.

FIG. 5 is a flow chart of a method for determining whether a heartbeat signal is an abnormal heartbeat signal according to an exemplary embodiment of the invention.

FIG. 6 is a schematic diagram of a heartbeat signal detecting apparatus according to an exemplary embodiment of the invention.

DETAILED DESCRIPTION OF THE INVENTION Reference will now be made in detail to the exemplary embodiments embodiments In addition, wherever possible, the same reference numerals in the drawings

In general, determining whether the heartbeat signal is an abnormal heartbeat signal can be determined by detecting whether the heartbeat signal has a delayed potential signal or an abnormal potential signal. For example, referring to FIG. 1, a common way is to detect whether the QRS wave end of the heartbeat signal has ventricular late potentials (VLP). If the heartbeat signal of the subject has a ventricular delay potential, it means that the heartbeat signal is an abnormal heartbeat signal and has a sign of ventricular arrhythmia.

In the heartbeat signal detection method disclosed in the embodiment of the present invention, the main detection object is an abnormal potential signal in the heartbeat signal, in particular, an Abnormal Intra-QRS Potentials (AIQP). The abnormal potential signal inside the QRS wave is superimposed on the QRS wave and has a small amplitude and a high frequency with respect to the QRS wave. Therefore, it is quite difficult to directly capture and interpret the abnormal potential signal inside the QRS wave. .

2 is a flow chart of a method for detecting a heartbeat signal according to an exemplary embodiment of the invention. Referring to FIG. 2, the heartbeat signal detecting method provided in the embodiment of the present invention first receives a heartbeat signal from a high-resolution electrocardiograph or an average signal electrocardiograph (step S210), and intercepts a characteristic wave in the heartbeat signal (step S220). . The characteristic wave is the QRS wave in the heartbeat signal in this embodiment, but the present invention is not limited to the special The wave is a QRS wave. Further, the heartbeat signal detecting method provided by the present invention is suitable for detecting an abnormal potential at any position in the heartbeat signal, so the P wave, the QRS wave or the T wave in the heartbeat signal can be used as the characteristic wave. Due to the heartbeat signal provided by the traditional 12-lead ECG, it is not possible to better present signal segments with weaker amplitude or higher frequency. In this embodiment, the preferred heartbeat signal is a heartbeat signal provided by a high resolution electrocardiograph or an average signal electrocardiograph. In particular, since the heartbeat signal presented by the average signal electrocardiograph has the characteristics of low noise, it is more suitable for detecting the abnormal potential signal inside the QRS wave or the ventricular delay potential.

After the feature wave in the heartbeat signal is intercepted, the heartbeat signal detecting method provided by the embodiment of the present invention further performs adaptive signal decomposition on the feature wave to obtain a plurality of characteristic wave components (step S230). In general, the abnormal potential signal in the QRS wave has a smaller amplitude and a higher frequency than the QRS wave. Therefore, in this embodiment, the heartbeat signal detection method further decomposes the characteristic wave into an adaptive signal decomposition. A plurality of characteristic wave components of different frequency components are covered, whereby a characteristic wave component having the highest frequency component is selected from the plurality of characteristic wave components as a test component (step S240). In other words, the characteristic wave component of the high-frequency component covering the heartbeat signal is locked and obtained by using the characteristic that the abnormal potential signal inside the QRS wave has a relatively high frequency to detect.

In the embodiment of the present invention, the adaptive signal performed in step S230 is decomposed into an Empirical Mode Decompositioin (EMD). Empirical mode decomposition is often used for signal analysis of nonlinear (Non-sta-ion) signals, while the original signal undergoing empirical mode decomposition can be decomposed into a finite number of essential modal functions (Intrinsic Mode Function (IMF) and a mean trend component (Trend), the equation is as follows:

Where x(t) is the original signal, c _k (t) is the kth essential mode function ( k is 1~ N , N is a positive integer) and r _n (t) is the mean trend component. In general, since the empirical mode decomposition usually preferentially decomposes the higher frequency intrinsic mode function by eliminating the low frequency carrier of the original signal, the first intrinsic mode function c _l (t) usually covers the original signal. High frequency component. Based on the foregoing characteristics, if the characteristic wave (QRS wave) has an abnormal potential signal (abnormal potential signal in the QRS wave), since the abnormal potential signal inside the characteristic wave has a relatively high frequency (relative to the characteristic wave), By the empirical mode decomposition method, the test component (the characteristic wave component having the highest frequency component) selected in step S240 also generally contains an abnormal potential signal. In the embodiment of the present invention, when the original signal x(t) is a characteristic wave, the essential modal function c _k (t) is a characteristic wave component, and the first essential modal function c _l (t) is usually Selected as the test component.

However, even if the characteristic wave does not have an abnormal potential signal, when the characteristic wave is subjected to adaptive signal decomposition, a plurality of characteristic wave components are also generated. The test component selected in step S240 at this time may be only a characteristic wave component of a relatively large amplitude and a low frequency (relative to the abnormal potential signal) in the heartbeat signal. More specifically, the test component obtained in the case where the characteristic wave does not have the abnormal potential signal has a frequency component lower than that obtained when the characteristic wave has an abnormal potential signal. The frequency components covered by the test components. In other words, if only the adaptive signal decomposition is used to obtain the test component, the obtained test component will have a serious difference in frequency depending on the presence or absence of the abnormal potential signal.

In order to solve the foregoing situation, the present embodiment further proposes a method of adding an enhancement wave to a characteristic wave. FIG. 3 is a flow chart of a method for obtaining a plurality of characteristic wave components according to an exemplary embodiment of the invention. Referring to FIG. 3, the step of performing adaptive signal decomposition to obtain the characteristic wave component further includes adding an enhanced wave to the characteristic wave (step S232), performing empirical mode decomposition on the characteristic wave to obtain a plurality of characteristic wave components (step S234), and The enhancement wave is subtracted from the characteristic wave component having the highest frequency component (step S236). More specifically, by adding an enhancement wave whose relative characteristic wave is a high frequency, the characteristic wave has the highest frequency among the plurality of characteristic wave components obtained by the adaptive signal decomposition method regardless of whether or not the characteristic wave has an abnormal potential signal. The characteristic wave components of the component will all contain the aforementioned enhancement wave. In other words, the enhancement wave is utilized such that the frequency components covered by the characteristic wave component having the highest frequency component are similar. Finally, the enhancement wave is subtracted from the characteristic wave component having the highest frequency component, and the test component for determining whether the characteristic wave has an abnormal potential signal can be obtained. The enhancement wave is a sine wave having a frequency of 250 Hz and a root mean square value of 35 microvolts (μV) in this embodiment, but is not limited thereto. In other embodiments, the enhanced wave may be a sine wave signal having a relatively high frequency (relative to the characteristic wave QRS wave). FIG. 4 is a schematic diagram showing characteristic wave and characteristic wave components. Referring to FIG. 4, by adding an enhancement wave to the characteristic wave (a), the obtained test component (b) has a distinct high frequency component.

Referring back to FIG. 2, the detection method of the heartbeat signal then calculates the ratio of the root mean square value of the test component to the root mean square value of the characteristic wave to obtain the evaluation parameter (step S250). The root mean square value of the test component, the root mean square value of the characteristic wave, and the ratio of the foregoing are as follows:

Where AIQP is the root mean square value of the test component, AQR is the ratio of the root mean square value of the test component to the root mean square value of the characteristic wave (QRS wave), QRSD is the signal length of the characteristic wave interval, and on and off are characteristic waves. The starting point and ending point, c(t) is the test component and x(t) is the characteristic wave. In more detail, AIQP represents the energy rms value of the test component, and AQR is an evaluation parameter obtained by normalizing the energy RMS AIQP of the test component using the energy value of the test component.

Finally, the detection method of the heartbeat signal determines whether the heartbeat signal is an abnormal heartbeat signal according to the evaluation parameter AQR (step S260). FIG. 5 is a flow chart of a method for determining whether a heartbeat signal is an abnormal heartbeat signal according to an exemplary embodiment of the invention. Referring to FIG. 5, in the embodiment of the present invention, the step of determining whether the heartbeat signal is an abnormal heartbeat signal further comprises determining whether the evaluation parameter is within a reference value range (step S262). If the evaluation parameter is outside the reference value range, it is determined that the characteristic wave has an abnormal potential signal, and the heartbeat signal is an abnormal heartbeat signal (step S264). On the other hand, if the evaluation parameter is within the reference range, it is judged that the characteristic wave does not have the abnormal potential signal (step S266).

In detail, the abnormal potential signal in the QRS wave has only a relatively small amplitude compared to the QRS wave. However, the abnormal potential signal still affects the amplitude of the characteristic wave, thereby causing the amplitude of the QRS wave. A slight difference is produced. The abnormal potential signal based on the QRS wave has a relatively high frequency characteristic. If the heartbeat signal is an abnormal heartbeat signal, the test component obtained in step S240 includes the root mean square of the abnormal potential signal in the QRS wave. The value AIQP is in principle greater than the root mean square value AIQP of the test component obtained in the case of a normal normal heartbeat signal. In other words, if the heartbeat signal is detected separately from the patient with ventricular arrhythmia and the average person, and the evaluation parameter AQR is obtained, the patient's evaluation parameter AQR should be greater than the average person's evaluation parameter AQR .

Table 1 shows the results of analysis of the heartbeat signal and the evaluation parameter AQR for patients with ventricular arrhythmia (26 persons) and the average person (42 persons). The measured heartbeat signal contains the heartbeat signals measured by the three segments of X, Y, and Z (ie, three vertical directions), so the evaluation parameters obtained are labeled as AQR_X , AQR_Y , and AQR_Z , respectively .

As can be seen from the above table, the evaluation parameter AQR of patients with ventricular arrhythmia, in principle, is greater than the average person's evaluation parameter AQR . By detecting and counting the average person's evaluation parameter AQR , one or more reference ranges can be summarized. When the heartbeat signal is detected, if the evaluation parameter AQR is outside the reference range, the heartbeat signal is an abnormal heartbeat signal.

In another embodiment of the present invention, the method of determining whether the heartbeat signal is an abnormal heartbeat signal may not utilize the enhanced wave. In this embodiment, the heartbeat signal detecting method performs the adaptive signal decomposition (step S230), does not add the enhanced wave to the characteristic wave, but directly performs empirical mode decomposition on the characteristic wave (QRS wave) to obtain multiple The characteristic wave component is selected, and the test component is selected therefrom (step S240). The evaluation parameter AQR is obtained by calculating the root mean square value AIQP of the test component and the root mean square value of the characteristic wave (step S250), and determining whether the heartbeat signal is an abnormal heartbeat signal according to the evaluation parameter AQR (step S260).

In this embodiment, when the characteristic wave does not have the abnormal potential signal (the abnormal potential signal in the QRS wave), since the enhanced wave to the characteristic wave is not added, the test component selected in step S240 is the heartbeat signal. A characteristic wave component of relatively large amplitude and low frequency (relative to an abnormal potential signal). In other words, if the patient with ventricular arrhythmia and the average person detect their heartbeat signal separately and obtain the root mean square value AIQP of the test component and the evaluation parameter AQR , then the patient's evaluation parameter AQR should be lower than the average person's evaluation parameter. AQR . Therefore, the evaluation and statistics of the average person's evaluation parameter AQR can be detected, and finally one or more reference ranges can be summarized. When the heartbeat signal is detected, if the evaluation parameter AQR is outside the reference range, the heartbeat signal is an abnormal heartbeat signal.

In the embodiment of the present invention, the heartbeat signal detecting method mainly determines the QRS. Whether the wave has an abnormal potential signal to determine whether the heartbeat signal is an abnormal heartbeat signal. In other embodiments, in addition to determining whether the QRS wave has an abnormal potential signal, it is also possible to simultaneously detect and determine whether there is a ventricular delay potential in the end of the QRS wave to improve the accuracy of detecting the abnormal heartbeat signal.

In another embodiment of the present invention, a heartbeat signal detecting apparatus using the foregoing heartbeat signal detecting method is further provided. FIG. 6 is a schematic diagram of a device for detecting a heartbeat signal according to an exemplary embodiment of the invention. The detecting device 700 of the heartbeat signal includes a receiving unit 710, an intercepting unit 720, an adaptive signal decomposition unit 730, a selecting unit 740, a calculating unit 750, and a determining unit 760. The receiving unit 710 receives the heartbeat signal (for example, receives the heartbeat signal from the electrocardiograph), and the intercepting unit 720 is coupled to the receiving unit 710 to intercept the characteristic wave in the heartbeat signal. In this embodiment, the feature wave intercepted by the intercepting unit 720 is a QRS wave in the heartbeat signal. The adaptive signal decomposition unit 730 is coupled to the intercepting unit 720 to perform adaptive signal decomposition on the characteristic wave to obtain a plurality of characteristic wave components. In the present embodiment, the adaptive signal is decomposed into empirical mode decomposition, but the invention is not limited thereto. The selection unit 740 is coupled to the adaptive signal decomposition unit 730, and selects a characteristic wave component having the highest frequency component from the plurality of characteristic wave components as a test component. The calculating unit 750 is coupled to the intercepting unit 720 and the selecting unit 740, and calculates a ratio of the root mean square value of the test component to the root mean square value of the characteristic wave to obtain the evaluation parameter. The determining unit 760 is coupled to the calculating unit 750, and determines whether the heartbeat signal is an abnormal heartbeat signal according to the evaluation parameter. In more detail, the judging unit 760 judges whether or not the evaluation parameter is within the reference value range. If If the estimated parameter is outside the reference value range, the determining unit 760 determines that the characteristic wave has an abnormal potential signal, and the heartbeat signal is an abnormal heartbeat signal.

For detailed operation and setting of the heartbeat signal detecting device 700, please refer to the detailed description of the heartbeat signal detecting method, and details are not described herein again.

As described above, the heartbeat signal detecting method and the heartbeat signal detecting device provided by the present invention mainly determine the measured condition by detecting an abnormal potential signal (especially an abnormal potential signal in the QRS wave) in the heartbeat signal. Does the person have signs of ventricular arrhythmia? The heartbeat signal detection method obtains a test component covering the high-frequency component of the heartbeat signal by adaptive signal decomposition, and checks whether the test component has an abnormal potential signal. In this way, it is possible to accurately and effectively determine whether the subject has a symptom of ventricular arrhythmia, thereby effectively overcoming/solving the problems described in the prior art.

While the present invention has been described above in the above exemplary embodiments, it is not intended to limit the scope of the present invention, and it is possible to make a few changes without departing from the spirit and scope of the invention. The scope of protection of the present invention is defined by the scope of the patent application. In addition, any of the objects or advantages or features of the present invention are not required to be achieved by any embodiment or application of the invention. In addition, the abstract sections and headings are only used to assist in the search of patent documents and are not intended to limit the scope of the invention.

Steps for S210~260‧‧‧ heartbeat signal detection method

Claims (10)

A heartbeat signal detecting method includes: receiving a heartbeat signal from a high resolution electrocardiograph or an average signal electrocardiograph; intercepting a characteristic wave in the heartbeat signal; performing an adaptive signal decomposition on the characteristic wave to obtain a plurality of features a wave component; selecting the characteristic wave component having the highest frequency component from the characteristic wave components as a test component; calculating a ratio of the root mean square value of the test component to the root mean square value of the characteristic wave to obtain An evaluation parameter; and determining whether the heartbeat signal is an abnormal heartbeat signal according to the evaluation parameter. The heartbeat signal detecting method according to claim 1, wherein the characteristic wave is a QRS wave in the heartbeat signal. The heartbeat signal detecting method according to claim 1, wherein the adaptive signal is decomposed into an empirical mode decomposition. The method for detecting a heartbeat signal according to claim 3, wherein the step of performing the adaptive signal decomposition to obtain the characteristic wave components further comprises: adding an enhancement wave to the characteristic wave; Empirical mode decomposition to obtain the characteristic wave components; and subtracting the enhancement wave from the characteristic wave component having the highest frequency component. For example, the method for detecting heartbeat signals as described in claim 1 of the patent application scope Determining whether the heartbeat signal is a step of the abnormal heartbeat signal, further comprising: determining whether the evaluation parameter is within a reference value range; and if the evaluation parameter is outside the reference value range, determining that the characteristic wave has an abnormal potential The signal, and the heartbeat signal is the abnormal heartbeat signal. A heartbeat signal detecting device includes: a receiving unit, receiving a heartbeat signal from a high resolution electrocardiograph or an average signal electrocardiograph; and an intercepting unit coupled to the receiving unit to intercept a characteristic wave in the heartbeat signal; An adaptive signal decomposition unit is coupled to the capture unit, and performs an adaptive signal decomposition on the characteristic wave to obtain a plurality of characteristic wave components; a selection unit coupled to the adaptive signal decomposition unit, from the The characteristic wave component having the highest frequency component is selected as a test component; a computing unit coupled to the intercepting unit and the selecting unit, calculating the root mean square value of the test component and the feature wave A ratio of the root mean square value is used to obtain an evaluation parameter; and a determining unit is coupled to the computing unit to determine whether the heartbeat signal is an abnormal heartbeat signal according to the evaluation parameter. The heartbeat signal detecting device according to claim 6, wherein the characteristic wave is a QRS wave in the heartbeat signal. The heartbeat signal detecting device according to claim 6, wherein the adaptive signal decomposed by the adaptive signal decomposition unit is decomposed into an empirical mode solution. The heartbeat signal detecting device according to claim 8, wherein the adaptive signal decomposition unit adds an enhancement wave to the characteristic wave, and performs the empirical mode decomposition on the characteristic wave to obtain the characteristic wave components. The enhanced wave is then subtracted from the characteristic wave component having the highest frequency component. The heartbeat signal detecting device according to claim 6, wherein the determining unit determines whether the evaluation parameter is within a reference value range, and if the evaluation parameter is outside the reference value range, the determining unit determines the characteristic wave There is an abnormal potential signal, and the heartbeat signal is the abnormal heartbeat signal.