Scribd
Upload
421 views

Image Processing of ECG Wave Using Matlab

The document loads ECG data and performs several signal processing techniques in MATLAB, including filtering, wavelet decomposition, and smoothing. It filters the raw ECG signal using a notch filter to remove noise, then applies wavelet decomposition at different levels to isolate different frequency components. It also performs smoothing to remove glitches before peak detection. Plots are generated throughout to visualize the processed signals.

Uploaded by

rakesh
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
421 views

Image Processing of ECG Wave Using Matlab

The document loads ECG data and performs several signal processing techniques in MATLAB, including filtering, wavelet decomposition, and smoothing. It filters the raw ECG signal using a notch filter to remove noise, then applies wavelet decomposition at different levels to isolate different frequency components. It also performs smoothing to remove glitches before peak detection. Plots are generated throughout to visualize the processed signals.

Uploaded by

rakesh
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 21

Image processing using

Matlab on ECG wave (.m file)

CASTORCLASSES
clc;
clear all;
close all;
val=load('100m.mat');
val=val.val;
f_s=360;%sampling frequency in hz
N=length(val);
t=[0:N-1]/f_s; % time period(total sample/Fs )
figure, plot(t,val);
title('Raw ECG Data plotting ')
xlabel('time')
ylabel('amplitude')
legend(' ORIGINAL ECG SIGNAL')

CASTORCLASSES
%%
half_total_sample=N/2; % N gives total
number of samples
val_new=val(1:half_total_sample);
t_new=[0:N/2-1]/f_s;
figure, plot(t_new,val_new,'k');
title('Half of the Raw ECG Data plotting ')
xlabel('time')
ylabel('amplitude')
legend('ORIGINAL ECG SIGNAL')

CASTORCLASSES
%%
w=50/(250/2);
bw=w;
[num,den]=iirnotch(w,bw); % notch filter
implementation
ecg_notch=filter(num,den,val);
figure,
N1=length(ecg_notch);
t1=[0:N1-1]/f_s;
plot(t1,ecg_notch,'r'); title('Filtered
ECG signal ')
xlabel('time')
ylabel('amplitude') CASTORCLASSES
%%
figure, subplot 211 % study useage of subplot under
help section
plot(t,val); title('Raw ECG Data plotting ')
xlabel('time')
ylabel('amplitude')
legend(' ORIGINAL ECG SIGNAL')
subplot 212
plot(t1,ecg_notch,'r'); title('Filtered ECG signal
')
xlabel('time')
ylabel('amplitude')
legend(' Flitered ECG SIGNAL')

CASTORCLASSES
%%
figure, plot(t(1:201),val(1:201),'r');
title('Data plotting for 0 to 0.8 time
frame')
xlabel('time')
ylabel('amplitude')
hold on
plot(t1(1:201),ecg_notch(1:201),'g');
legend('ORIGINAL ECG SIGNAL',' Flitered
ECG SIGNAL')
hold off

CASTORCLASSES
%%
[e,f]=wavedec(ecg_notch,10,'db6');% Wavelet implementation
g=wrcoef('a',e,f,'db6',8);
figure, plot(g),title('checking the recoefficient output')
ecg_wave=ecg_notch-g; % subtracting 10th level aproximation signal
%from original signal
ecg_smooth=smooth(ecg_wave); % using average filter to remove glitches
%to increase the performance of peak detection
N1=length(ecg_smooth);
t1=(0:N1-1)/f_s;
figure,
subplot 211 ,plot(t,val),title('Original ecg signal'),ylabel('amplitude')
xlabel('time')
subplot 212,plot(t1,ecg_smooth,'r'),title('Baseline Removed Ecg signal');
ylabel('amplitude'),xlabel('time')
figure,plot(t,val,'r')
hold on ;
plot(t1,ecg_smooth,'g'),ylabel('amplitude'),xlabel('time')
title('Filtered ECG signal after wavelet and smooth filter')
legend('ORIGINAL ECG SIGNAL',' Flitered ECG SIGNAL')
hold off
CASTORCLASSES
%%
[num,den]=wavedec(val,10,'db6');% Wavelet implementation for
% sampling frequency of 360
d1=wrcoef('d',num,den,'db6',1); % frequency (62-128)
d2=wrcoef('d',num,den,'db6',2); % frequency (31-62)
a1=wrcoef('a',num,den,'db6',1); % frequency (0-62)
a2=wrcoef('a',num,den,'db6',2); % frequency (0-31)

figure
subplot 511, plot(ecg_notch),ylabel('ecg'),xlabel('length')
legend('Sampling frequency 360')
title('decomposition for d1,d2,a1,a2')
subplot 512,plot(d1,'r'),ylabel('d1'),xlabel('length')
legend('freq 62-128')
subplot 513,plot(d2,'k'),ylabel('d2'),xlabel('length'),legend('freq 31-
62')
subplot 514,plot(a1,'g'),ylabel('a1'),xlabel('length'),legend('freq 0-
62')
subplot 515,plot(a2,'m'),ylabel('a2'),xlabel('length'),legend('freq 0-
31')
CASTORCLASSES
%%
d3=wrcoef('d',num,den,'db6',3); % frequency (15-31)
d4=wrcoef('d',num,den,'db6',4); % frequency (0-7)
a3=wrcoef('a',num,den,'db6',3); % frequency (7-15)
a4=wrcoef('a',num,den,'db6',4); % frequency (0-8)

figure
subplot 511, plot(ecg_notch),ylabel('ecg'),xlabel('length')
legend('Sampling frequency 360')
title('decomposition for d3,d4,a3,a4')
subplot 512,plot(d3,'r'),ylabel('d3'),xlabel('length'),legend('freq 15-
31')
subplot 513,plot(d4,'k'),ylabel('d4'),xlabel('length'),legend('freq 7-
15')
subplot 514,plot(a3,'g'),ylabel('a3'),xlabel('length'),legend('freq 0-
15')
subplot 515,plot(a4,'m'),ylabel('a4'),xlabel('length'),legend('freq 0-
8')

%%
CASTORCLASSES
CASTORCLASSES
CASTORCLASSES
CASTORCLASSES
CASTORCLASSES
CASTORCLASSES
CASTORCLASSES
CASTORCLASSES
CASTORCLASSES
CASTORCLASSES
CASTORCLASSES
CASTORCLASSES
CASTORCLASSES

You might also like