KR101504487B1 - Real Time System for Measuring Fetal Heart Rate - Google Patents

Abstract

A heart rate measuring system for a fetus according to an embodiment of the present invention includes: an adaptive filter which removes only the electrocardiogram signal of a mother from the electrocardiogram signal of the mother and the electrocardiogram signal of the fetus; an online empirical mode decomposition (EMD) which decomposes a signal extracted from the adaptive filter into multiple intrinsic mode function (IMF) signals; a window designating unit which designates a window based on the peak value of a phonocardiogram signal of the fetus; and a synthesizing unit which synthesizes the window designated by the window designating unit with the electrocardiogram signal selected among the IMF signals decomposed by the online EMD. A signal included in the window is drawn as a final electrocardiogram signal of a fetus among signals synthesized by the synthesis unit. According to an embodiment, the present invention is able to handle a situation fast according to the state of a fetus by consistently monitoring the state of the fetus since the electrocardiogram of the fetus can be measured in real time.

Description

[0001] The present invention relates to a system for measuring the heart rate of a fetus in real time (Real Time System for Measuring Fetal Heart Rate)

The present invention relates to a system for measuring a heart rate of a fetus, and more particularly, to a system for measuring the heart rate of a fetus from a mother and measuring it in real time.

The fetal pulse refers to the overall sustained pulse rate in the absence of significant changes in fetal pulse for labor or other reasons. The fetal heart rate is confirmed from 6 weeks of gestation by vaginal ultrasound and the fetal heart rate of gestation is normal from 120 to 190 per minute. At term infants, the fetal pulse rate is more than twice that of adults, so 120 to 160 pulse rates per minute are normal, and the average is measured at about 144 times per minute.

The fetal heart rate is usually 120 to 160 in normal range, and even if the fetal condition is normal, the fetal heart rate is normal. Therefore, the simple fetal heart rate is not of great medical significance. However, if the fetal heart rate is lower than the weight 120 for more than 10 minutes 100-119 / min is considered a mild decline, 80-100 / min is a moderate decline, and below this is considered a severe decline. If the fetal heart rate is less than 90 or more than 40 minutes in the baseline heart rate is more than 1 minute, the condition of the fetus is likely to be bad and careful observation is required.

Conventional methods for measuring the heart rate of a fetus include a measurement method using a Doppler ultrasound signal, a blind source separation, and a method using an ECG and a heart sound together.

However, the most widely used Doppler ultrasound method has a disadvantage in that it is difficult to measure the heartbeat variation of the fetus due to errors in the movement of the mother or the fetus, the error between the heart sound and the calculated heart rate, Cost is also high and there is a problem that continuous measurement is impossible.

The method using blind source separation is a method to measure fetal electrocardiogram from maternal electrocardiogram using Independent Component Analysis (ICA) by measuring several electrocardiograms in the maternal abdomen. By using several electrocardiogram channels, relatively accurate fetal electrocardiogram Can be measured. However, the method using ICA requires that the electrocardiogram be extracted by visual confirmation at the time of measurement. The electrocardiogram can be measured only when the number of channels is equal to or more than a certain number. The ICA algorithm requires a large amount of computation for signal processing, It is difficult to measure.

The method of using ECG and heart sound also uses the ICA technique to extract the electrocardiogram of the fetus by detecting the state of the uterus using the electrocardiogram signal and the heart sound of the mother, It is not suitable as a system for measuring heart rate in real time.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems and it is an object of the present invention to provide a system for measuring a heart rate of a fetus which reduces the occurrence of a delay time by data processing and reduces the number of channel electrodes, The purpose of this system is to provide a system that can do this.

The system for measuring the heart rate of a fetus according to an embodiment of the present invention includes an adaptive filter for removing only an electrocardiogram signal of a mother in an electrocardiogram signal of a mother and a fetus; An online EMD (Empirical mode decomposition) for decomposing the signal extracted from the adaptive filter into a plurality of IMF (Intrinsic mode functions) signals; A window specifying unit for specifying a window based on a peak value of a signal of the heart of a fetus; And a synthesizer for synthesizing an ECG signal selected from the IMF signals decomposed by the Online EMD by a window specified by the window designator, wherein, among the signals synthesized by the synthesizer, And is derived from the final electrocardiogram signal of the fetus.

The method of measuring a heart rate of a fetus according to an embodiment of the present invention includes measuring an electrocardiogram signal of a mother and a fetus in an abdomen of a mother as an input signal; Measuring the electrocardiographic signal of the mother in the heart of the mother as a reference signal; Measuring the heart sound signal of the fetus in the abdomen of the mother; Estimating and removing a reference signal included in the input signal through an adaptive filter; Decomposing the signal extracted by the adaptive filter into a plurality of IMF signals through Online EMD; Setting a window based on a peak value of the heart sound signal of the fetus; And deriving a final electrocardiogram section of the fetus by synthesizing the IMF signal selected from the signals decomposed through the Online EMD and the window.

The heart rate measuring system of the fetus according to the embodiment of the present invention can measure the electrocardiogram of the fetus in real time so that the state of the fetus can be continuously monitored and it is possible to cope with the condition of the fetus rapidly.

In addition, in the embodiment, by using three channel electrodes used for fetal electrocardiogram measurement, it is easier to measure compared with the method using more than 10 multi-channel electrodes in measuring the fetal heart rate, and the fetal heart rate is continuously There are advantages that are easy to monitor.

1 is a view showing an apparatus for measuring a heart rate of a fetus according to an embodiment;
2 is a block diagram showing a system for measuring the heart rate of a fetus according to the embodiment
Figure 3 shows signal rejection through the adaptive filter of Figure 2;
4 is a flowchart illustrating an Online EMD method according to an embodiment;
5 illustrates IMF signal extraction according to the Online EMD method of FIG. 4; FIG.
6 is a diagram showing an IMF signal according to the Online EMD method
7 is a diagram showing a system structure of Online EMD
8 is a view showing a problem caused in a process of synthesizing signals by window buffers
9 is a view showing a method for solving the problem in Fig. 8
10 is a view showing a heart sound of a fetus
Fig. 11 is a diagram showing extraction of a fetal electrocardiogram signal using a fetal heart tone
12 is a graph showing an electrocardiogram signal of a fetus estimated by the heart rate measuring system of a fetus according to the embodiment
13 is a graph comparing root mean square errors according to voltage magnitudes of input initial signals
FIG. 14 is a diagram showing a result of evaluating real-time performance of a fetal heart rate measuring system according to the embodiment

The embodiments of the present invention will be described in detail with reference to the accompanying drawings, but the present invention is not limited to these embodiments. In describing the present invention, a detailed description of well-known functions or constructions may be omitted for the sake of clarity of the present invention.

1 is a diagram showing a system for measuring the heart rate of a fetus according to an embodiment.

Referring to FIG. 1, the heart rate measuring system of the embodiment includes a controller 2 configured by a computer or the like, a measurement unit 3 including a DAQ (Data Acquisition) for measuring electrocardiogram and heart sound, and a plurality of channel electrodes Lt; / RTI >

The channel electrode includes a first electrode 4 attached to the chest of the mother to measure the electrocardiogram of the mother, a second electrode 5 attached to the abdomen of the mother to measure the electrocardiogram of the mother and the fetus, A third electrode 6 may be provided for measuring the degree. In this embodiment, a real-time system using an algorithm for extracting the fetal heartbeat number by comparing and analyzing the data collected from each electrode is proposed and will be described in detail below.

2 is a block diagram illustrating a system for measuring the heart rate of a fetus according to an embodiment of the present invention. The embodiment is a method for measuring the heart rate of a fetus using an electrocardiogram (S2) of a mother and a fetus and a heart sound S3 of a fetus measured in a maternal abdomen as an input signal, The fetal electrocardiogram is extracted using the signal S1.

In order to estimate the electrocardiogram of the fetus, an adaptive filter (15) is used to remove the electrocardiogram (S1) of the mother from the signal (S2) measured at the abdomen of the mother and the signal is converted into an empirical mode decomposition 16). After extracting only the electrocardiogram of the fetus from the disassembled signal, the heart rate of the fetus is measured by searching the fetal electrocardiogram section using the heart sound of the fetus (S3).

Specifically, in the configuration of the embodiment, the adaptive filter 15 may be provided to remove the electrocardiogram signal S1 of the mother from the electrocardiogram signal S2 of the mother and fetus as the input signal. The adaptive filter 15 is a filter that implements an optimized algorithm based on an input signal so that its function is changed in a self-adaptive manner. In the present invention, by using an adaptive filter, a recursive least square method is used to preferentially remove only the electrocardiogram of a mother.

The cyclic least squares method is widely used in adaptive algorithms. It minimizes the sum of squared errors, has fast convergence ability, and provides an accurate solution for each hourly minimization problem. The reference signal included in the input signal is estimated through the reference signal in the adaptive filter when the signal S2 in which the electrocardiogram of the mother and the fetus are mixed is referred to as an input signal and the electrocardiogram signal S1 of the mother is referred to as a reference signal. Only the electrocardiogram signal of the mother is extracted. When this is removed from the input signal, the electrocardiogram signal of the mother is removed from the input signal.

FIG. 3 is a diagram illustrating signal rejection through the adaptive filter of FIG. 2. FIG. Referring to FIG. 3, in a signal obtained by combining a maternal electrocardiogram (mECG) and a fetal electrocardiogram (fECG), a circulatory least squares method is applied using an adaptive filter 15, Indicates that only the signal mECG is removed. Since the amplitude of the electrocardiogram signal of the mother is larger than that of the fetal electrocardiogram signal, it is possible to extract only the electrocardiogram of the fetus whose scale is reduced to within 0.1 mV by removing the electrocardiogram signal of the mother.

In addition, the signal that the fetal electrocardiogram signal is removed by removing the electrocardiogram signal of the mother can be resolved using the online EMD (empirical mode decomposition) method. Here, the EMD method is a new method of time-frequency analysis that adaptively and efficiently decomposes a signal. The EMD method is decomposed into a set of functions called IMFs (Intrinsic mode functions) defined by the signal itself, and the decomposed IMFs can be used as basis functions and filters because they preserve the inherent properties of the original signal. The signal decomposition by the EMD method is based on the local time scale characteristics of the signal, so it is suitable for nonlinear and abnormal signal processing and can be used for biological signal processing such as heart beat (ECG).

However, in the conventional EMD method, many repetitive calculations are generated in the process of extracting the IMF, and the amount of calculation is increased. Therefore, it is not easy to process the data in real time, and thus it is used for processing the already recorded data. Accordingly, the present embodiment intends to estimate the heartbeat number of a fetus through an online EMD method capable of decomposing and analyzing data in real time by improving the conventional EMD method.

Referring again to FIG. 2, a signal estimated to be a fetal electrocardiogram signal through the Online EMD 16 is decomposed into N, and in the selector 17, among the N decomposed signals using the Median Frequency, Signal.

FIG. 4 is a flowchart illustrating Online EMD according to an embodiment of the present invention, and FIG. 5 is a diagram illustrating a change of a signal according to FIG.

Referring to FIGS. 4 and 5, in the Online EMD, a signal in which a mother's heartbeat signal is removed is transmitted in a mixed signal of a maternal-fetal heartbeat number, and Online EMD serves to decompose the signal into N signals .

First, in step D1, the input signal r (t) is defined as x (t). Then, in step D2, a step of extracting all the poles of the signal x (t) is performed.

In step D3, the upper envelope Eu (t) is derived by connecting the upper side of the pole, and the lower envelope El (t) is derived by connecting the lower side of the pole. In step D4, the average of the upper envelope Eu (t) and the lower envelope El (t) is calculated to obtain an average signal m (t).

In step D5, the final signal h (t) is derived by subtracting the average signal m (t) from the input signal x (t). Next, in step D6, it is determined whether the final signal h (t) is an IMF. If not, the input signal is set to h (t) to perform the step D2 on the input signal again.

If h (t) is determined to be IMF, h (t) is defined as c (t) which is the first IMF signal IMF1 in step D7. The value obtained by subtracting the c (t) from the input signal r (t) is set as an input signal for deriving the second IMF signal IMF2.

FIG. 6 shows a drawing in which continuous IMF signals are extracted using Online EMD in an input signal. As shown in FIG. 6, the signal transmitted through the Online EMD according to the embodiment includes a first IMF signal IMF1, a third IMF signal IMF3, a fourth IMF signal IMF2, IMF4) and the fifth IMF signal IMF5. The area under the curve of the decomposed signal is divided by the same area, and the corresponding median frequency is obtained, and the corresponding IMF signal is selected.

7 is a diagram showing the system structure of Online EMD. Referring to FIG. 7, the fetal heart rate measuring system according to the embodiment sets a buffer to extract fetal electrocardiogram signals in real time, forms a continuous IMF extracting means, and simultaneously extracts a plurality of IMFs.

That is, in the conventional EMD method, the second IMF signal IMF2 is extracted after all the first IMF signals IMF1 are extracted. However, in the online EMD method according to the embodiment, the entire IMF signal can be decomposed at once. Since the value of IMF1 is stored in the data of the buffer every time data is input, IMF2 can be obtained immediately by using this information in the next buffer data. By repeating this process, it is possible to simultaneously obtain several IMF signals from one signal in real time.

However, as shown in FIG. 8, a block-wise phenomenon in which a discontinuous section of data appears when data is concatenated for each buffer in the process of synthesizing signals by buffer and forming an IMF signal may occur.

Therefore, in order to solve this problem, it may include a process of overlapping the signals by 50% using a sliding window as shown in FIG.

Referring again to FIG. 2, the signal selected by the selecting unit 17 is subjected to windowing in the synthesizing unit 18, at which time the fetal heartbeat signal PCG is referred to.

FIG. 10 is a view showing the heart sound of the fetus. The signal S3 of the fetus is passed through the band-pass filter 13 to remove the signal of a specific frequency, and in the detection unit 18, Is detected. The window designation unit 19 designates a window including a time around 60 ms based on the peak of the heart sound signal.

11 is a diagram showing extraction of a fetal electrocardiogram signal using a fetal heart sound. Referring to Fig. 11, the signal selected as the electrocardiogram of the fetus in the selector 17 is synthesized with the window specified by the heart sound of the fetus.

12 is a diagram showing an electrocardiogram signal of a fetus estimated by the heart rate measuring system of a fetus according to the embodiment. Referring to FIG. 12, except for the electrocardiogram signal of the fetus windowed by the heart sound of the fetus, the fetal electrocardiogram signal can be finally estimated.

13 is a graph comparing root mean square errors according to voltage magnitudes of input initial signals. Referring to FIG. 13, there is shown a result of performing a performance evaluation using the simulation signal of the fetal heart rate measuring system according to the embodiment. To obtain the simulation signal, the electrocardiogram signals of the users A and B were merged, and the user B was exercised to satisfy the conditions of the fetal heart rate of 120 ~ 160 bpm.

User B measured the electrocardiogram and heart soundness together, and user A measured the electrocardiogram only in the abdomen and formed a simulation signal to apply the algorithm included in the heart rate measurement system according to the embodiment. The reference signal was evaluated based on the electrocardiogram signal measured by only the electrocardiogram of the user B.

As shown in the figure, as the rescaling of the electrocardiogram signal of the user B from 0.06 mV to 0.2 mV progresses, the electrocardiogram and the heart sound signal of the user B modeled on the fetus in the case of the signal having the value of 0.2 mV, The result shows that the electrocardiogram error of about 1.09 beat is obtained, and the result of the estimated electrocardiogram signal is within the error range with reliability.

FIG. 14 is a diagram showing a result of evaluating real-time performance of a fetal heart rate measuring system according to an embodiment. Referring to FIG. 14, at the sampling rate of 1K Sammples / s, the user A and the fetus model B simultaneously receive signals for 30 seconds and combine them. When the algorithm included in the fetal heart rate measuring system proposed in the embodiment is applied, the fetal electrocardiogram is estimated to have a processing speed of about 5.3 seconds. This means that the electrocardiogram measuring method according to the embodiment can extract the fetal electrocardiogram signal and the graph corresponding thereto in real time.

Therefore, the heart rate measuring system of the fetus according to the embodiment of the present invention can measure the electrocardiogram of the fetus in real time, so that the state of the fetus can be continuously monitored, so that it is possible to promptly cope with the state of the fetus.

In the embodiment, by using three channel electrodes used for fetal electrocardiogram, it is easier to measure compared with the method using more than 10 multi-channel electrodes in the measurement of the fetal heart rate, and the fetal heart rate is continuously monitored .

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, It will be understood that various modifications and applications other than those described above are possible. For example, each component specifically shown in the embodiments of the present invention can be modified and implemented. It is to be understood that all changes and modifications that come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

2:
3:
4: first electrode
5: Second electrode
6: Third electrode
11, 12, 13: Bandpass filter
15: Adaptive filter
16: Online EMD
17:
18:
19:

Claims (12)

An adaptive filter which removes only the ECG signal of the mother in the fetal ECG signal;
An online EMD (Empirical mode decomposition) for decomposing the signal extracted from the adaptive filter into a plurality of IMF (Intrinsic mode functions) signals;
A window specifying unit for specifying a window based on a peak value of a signal of the heart of a fetus; And
And a synthesizer for synthesizing a window designated by the window designator to an electrocardiogram signal selected from the IMF signals decomposed by the Online EMD,
And a signal included in the window is derived as a final electrocardiogram signal of the fetus among the signals synthesized by the synthesizer. The method according to claim 1,
Wherein the adaptive filter is a fetal heart rate measurement system for removing only a maternal signal from a fetal electrocardiogram signal using a recursive least square method. The method according to claim 1,
And a selection unit for selecting an electrocardiogram of the fetus through the Median Frequency among the signals decomposed by the Online EMD. The method according to claim 1,
A fetal heart rate measuring system comprising a detector for detecting a peak value by measuring a fetal heart rate. The method according to claim 1,
Wherein the Online EMD includes a buffer for storing an IMF signal value to be decomposed, and derives a continuous IMF signal value using the value stored in the buffer. The method according to claim 1,
A fetal heart rate measuring system further comprising a first electrode for measuring the electrocardiogram signal of the mother, a second electrode for measuring the electrocardiogram signal of the mother and the fetus, and a third electrode for measuring the heart sound of the fetus. The method according to claim 6,
And a band-pass filter for filtering the electrocardiogram signal measured at the first to third electrodes to have a predetermined band value. Measuring an electrocardiographic signal of a mother and a fetus in an abdomen of a mother as an input signal;
Measuring the electrocardiographic signal of the mother in the heart of the mother as a reference signal;
Measuring the heart sound signal of the fetus in the abdomen of the mother;
Estimating and removing a reference signal included in the input signal through an adaptive filter;
Decomposing the signal extracted by the adaptive filter into a plurality of IMF signals through Online EMD;
Setting a window based on a peak value of the heart sound signal of the fetus; And
Extracting a final electrocardiogram section of the fetus by synthesizing the IMF signal and the window selected from the signals decomposed through the Online EMD;
Of fetal heart rate. 9. The method of claim 8,
Wherein the adaptive filter is a method of measuring heart rate of a fetus which removes only a maternal signal from a fetal electrocardiogram signal using a recursive least square method. 9. The method of claim 8,
The Online EMD is a method for measuring the heart rate of the fetus that selects the fetal electrocardiogram from the disassembled IMF signal through Median Frequency. 9. The method of claim 8,
Wherein the Online EMD includes a buffer for storing an IMF signal value to be decomposed, and performs a continuous IMF signal decomposition process using a value stored in the buffer. 9. The method of claim 8,
The step of deriving the final electrocardiogram section of the fetus by synthesizing the IMF signal and the window selected from the signals decomposed by the Online EMD,
Wherein the IMF signal included in the window is derived from a fetal electrocardiogram signal.

Images