Method and system for extracting ventricular fibrillation signals in electrocardiogram using spline interpolation with uniform phase ensembles

Patent metadata

Field	Value
Publication number	US-9451898-B2
Application number	US-201614993584-A
Country	US
Kind code	B2
Filing date	Jan 12, 2016
Priority date	Jul 2, 2014
Publication date	Sep 27, 2016
Grant date	Sep 27, 2016

Abstract

This invention discloses a system and a method for extracting VF signal in ECG recorded during uninterrupted CPR. The present invention provides a method for extracting a Ventricular fibrillation (VF) signal in Electrocardiography (ECG), comprising: receiving an ECG signal; adding a plurality of shadowing functions to the ECG signal, to obtain a plurality of modification signals; decomposing the plurality of modification signals by using an Empirical Mode Decomposition (EMD) method, to generate a plurality of Intrinsic Mode Functions (IMFs); calculating the sum of IMFs in different frequency regions based on time sequence, dividing by a number of the shadowing signal, to obtain a plurality of modification intrinsic mode functions; combining the plurality of modification IMFs with the same property, to obtain a shape function; modeling the shape functions to obtain a compression signal; and subtracting the compression signal from the ECG signal based on time sequence, to obtain the VF signal.

First claim

What is claimed is:

1. A computer-implemented method for extracting a Ventricular fibrillation (VF) signal in Electrocardiography (ECG), the method comprising:
(A) receiving an ECG signal;
(B) adding a plurality of shadowing functions to the ECG signal, to obtain a plurality of modification signals, wherein each pair of shadowing functions is conjugate and has same phase difference;
(C) decomposing the plurality of modification signals by using an Empirical Mode Decomposition (EMD) method, to generate a plurality of Intrinsic Mode Functions (IMFs);
(D) calculating the sum of IMFs in different frequency regions based on time sequence, dividing by a number of the shadowing signal, to obtain a plurality of modification intrinsic mode functions;
(E) combining the plurality of modification IMFs with the same property, to obtain a shape function;
(F) modeling the shape function, to obtain a compression signal; and
(G) subtracting the compression signal from the ECG signal based on time sequence, to obtain the VF signal.

2. The method according to claim 1 , wherein the ECG signal is a function of amplitude and time.

3. The method according to claim 1 , the steps further comprises:
(B1) adding a first shadowing signal to the ECG signal, to obtain a first modification signal;
(B2) generating a plurality of first intrinsic mode functions by using the EMD method, wherein the first intrinsic mode function is an amplitude expression of the first modification signal in the different frequency regions;
(B3) adding a second shadowing signal to the ECG signal to obtain a second modification signal;
(B4) generating a plurality of second intrinsic mode functions by using the EMD method, wherein the second intrinsic mode function is an amplitude expression of the second modification signal in the different frequency regions; and
(B5) repeating to add a np-1 shadowing signal and a np shadowing signal to the ECG signal, to obtain the plurality of intrinsic mode functions, wherein the phase difference of the shadowing signals are constant.

4. The method according to claim 3 , wherein the first shadowing function and the second shadowing function are conjugate.

5. The method according to claim 1 , the steps further comprises:
(C1) selecting the intrinsic mode functions corresponding to a first frequency region, calculating the sum of the intrinsic mode functions based on the time sequence, then dividing by the a number of the shadowing signal, to obtain a first modification intrinsic mode function; and
(C2) repeating step C1 until to obtain the modification intrinsic mode functions from all frequency regions.

6. The method according to claim 3 , wherein a function of the shadowing signal is calculated according to the following expression:
w i ( t )=ε*cos(2 πt/T w −2 πi/np )

wherein c is the amplitude of the shadowing signal, T w is a period of the shadowing signal, np is the number of the shadowing signal and i is any number into 0 to np-1.

7. The method according to claim 1 , the steps further comprises:
(B1) adding a first-level of first shadowing signal to the ECG signal to obtain a first-level of first modification signal, and generating a plurality of first-level of first intrinsic mode functions by using the EMD method, wherein the first-level of first intrinsic mode function is an expression amplitude of the first-level of first modification signal in the first frequency region;
(B2) adding a first-level of second shadowing signal to the ECG signal to obtain a first-level of second modification signal, and generating a plurality of first-level of second intrinsic mode functions by using the EMD method, wherein the first-level of second intrinsic mode function is an expression amplitude of the first-level of second modification signal in the first frequency region;
(B3) repeating step B1 to B2 to add a first level of np-1 shadowing signal and a first level of np shadowing signal to the ECG signal, to obtain the plurality of first level intrinsic mode functions, wherein the phase difference of the shadowing signals are constant;
(B4) calculating the sum of the first level intrinsic mode functions based on the time sequence, then dividing by the number of shadowing signal, to obtain a first modification intrinsic mode function; and
(B5) repeating step B1 to B4 to add a m level of shadowing signal to the ECG signal, to obtain a plurality of m level of intrinsic mode functions, calculating the sum of m level of the intrinsic mode functions based on the time sequence, then dividing by the number of shadowing signal to obtain a first level of m modification intrinsic mode function, until to obtain the modification intrinsic mode functions from all frequency regions.

8. The method according to claim 7 , wherein a m level of function of the shadowing signal is calculated according to the following expression:
w i ( t )=ε*cos(2 πt /2 m −2 πi/np )

wherein the np-1 shadowing signal and the np shadowing signal is a m level of the shadowing function, wherein m is corresponding to the m level of the modification intrinsic mode function.

9. A system for extracting a VF signal in ECG comprising:
a signal collecting unit, to receive an ECG signal;
a signal processing unit electrically connected to the signal collecting unit, to add a plurality of shadowing functions to the ECG signal, to obtain a plurality of modification signals, wherein each pair of shadowing functions is conjugate and has same phase difference, then decomposing the plurality of modification signals by using an Empirical Mode Decomposition (EMD) method, to generate a plurality of Intrinsic Mode Functions (IMFs), calculating the sum of IMFs in different frequency regions based on time sequence, dividing by a number of the shadowing signal, to obtain a plurality of modification intrinsic mode functions, combining the plurality of modification IMFs with the same property, to obtain a shape function, modeling the shape functions to obtain a compression signal, and subtracting the compression signal from the ECG signal based on time sequence to obtain the VF signal; and
a signal outputting unit connected to the signal processing unit, to output the VF signal.

10. The system according to claim 9 , wherein the ECG signal is a function of amplitude and time.

Assignees

• Univ Nat Central

Inventors

• Lo Men-Tzung
• Wang Yung-Hung
• Lin Chen
• Young Hsu-Wen Vincent
• Chang Hsiang-Chih
• Lin Lian-Yu
• Hsieh Wan-Hsin
• Ma Matthew Huei-Ming
• Hu Kun

Classifications

• A61B5/046 Primary

  Human Necessities · mapped topic

• A61B5/04014

  Human Necessities · mapped topic

• A61B5/11

  Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb {(A61B5/1038 takes precedence; motion detection to correct for motion artifacts in physiological signals A61B5/721)} · CPC title

• A61B5/7207 Primary

  of noise induced by motion artifacts · CPC title

• A61B5/04012

  Human Necessities · mapped topic