DWT Based Beat Rate Detection in ECG Analysis

advertisement
Applications of The Discrete
Wavelet Transform in Beat
Rate Detection
Advisor:Jian-Jiun Ding
Speaker:Jian-Hwu Wang
Date:11/25/2010
DISP Lab, GICE, NTU
1
Outline
• Introduction to Wavelet Transform
• Applications of the Discrete Wavelet Transform in
Beat Rate Detection
– DWT Based Beat Rate Detection in ECG
Analysis.
– Improved ECG Signal Analysis Using Wavelet
and Feature.
• Conclusion
• Reference
DISP Lab, GICE, NTU
2 /22
Introduction to wavelet transform
• Fourier transform is the well-known tool for
signal processing.

X ( f )   x(t )e j 2ft dt

• One limitation is that a Fourier transform can’t deal
effectively with non-stationary signal.
• Short time Fourier transform

X (t , f )   w(t   )x( )e j 2f d

where w(t ) is mask function
DISP Lab, GICE, NTU
3 /22
Introduction to wavelet transform
• Gabor Transform
– The mask function is satisfied with Gaussian
distribution.
• Uncertainly principle
 t f 
1
4
t x(t ) dt


, 
 x(t ) dt
2
2
where 
2
t
2
f X ( f ) df


 X ( f ) df
2
2
2
f
2
• We expected to occur a high resolution in time domain, and then
adjust  t2 or  2f .
DISP Lab, GICE, NTU
4 /22
Introduction to wavelet transform
• The principle of wavelet transform is based on the
concept of STFT and Uncertainly principle.
– A mother wavelet  (t ) .
1
t
 ( ) and translating  (t  b) .
– Scaling
a
a
• Sub-wavelets
 a ,b (t ) 
• Fourier transform
1
t b
(
)
a
a
 (t )  F[ (t )]
a,b (t )  F[ a,b (t )]
DISP Lab, GICE, NTU
5 /22
Introduction to wavelet transform
• Continuous wavelet transform(CWT)
wa ,b
1
  a ,b , x(t ) 
a



x(t ) a ,b (
t b
)dt
a
• ICWT
1
x(t ) 
C


dadb
  wa,b a,b (t ) a 2
where C  

0
 ( w)
w
dw and
DISP Lab, GICE, NTU



 ( w) dw  
6 /22
Introduction to wavelet transform
• Discrete wavelet transform(DWT)
wm,n  x(t ), m,n  a0m / 2  f (t ) (a0m (t )  nb0 )dt
– Sub-wavelets
 m,n (t )  a0m/ 2 (a0m (t )  nb0 )
• IDWT
m, n  Z
x(t )   wm,n m,n (t )
m
n
DISP Lab, GICE, NTU
7 /22
DWT applications for
beat rate detection
• DWT Based Beat Rate Detection in ECG
Analysis
– The purpose of this paper is to detect heart beat
rate by the concept of discrete wavelet
transform, which is suitable for the non
stationary ECG signals as it has adeuate scale
values and shifting in time.
DISP Lab, GICE, NTU
8 /22
DWT Based Beat Rate
Detection in ECG Analysis
• ECG(Electrocardiogram) signal
DISP Lab, GICE, NTU
9 /22
DWT Based Beat Rate
Detection in ECG Analysis
• Preprocessing
– Denoise
• Baseline wandering
• Moving average method and subtraction procedure.
DISP Lab, GICE, NTU
10 /22
DWT Based Beat Rate
Detection in ECG Analysis
• Preprocessing
– Denoising : The wavelet transform is used prefiltering step for subsequent R spike detection
by thresholding of the coefficients.
• Decomposition.
• Thresholding detail coefficients.
• Reconstruction.
DISP Lab, GICE, NTU
11 /22
DWT Based Beat Rate
Detection in ECG Analysis
• Feature extraction using DWT
– Detect R-waves.
– Thresholding.
• Positive threshold.
• Negative threshold.
DISP Lab, GICE, NTU
12 /22
DWT applications for
beat rate detection
• Improved ECG Signal Analysis Using
Wavelet and Feature.
– This paper introduced wavelet to extract
features and then distinguish several heart beat
condition, such as normal beats, atrial
premature beats, and premature ventricular
contractions.
DISP Lab, GICE, NTU
13 /22
Improved ECG Signal Analysis Using
Wavelet and Feature.
• Some kinds of ECG signal:
Atrial premature beat
Normal beat
Premature ventricular
contractions
DISP Lab, GICE, NTU
14 /22
Improved ECG Signal Analysis Using
Wavelet and Feature.
• ECG signal analysis flow
DISP Lab, GICE, NTU
15 /22
Improved ECG Signal Analysis Using
Wavelet and Feature.
• Feature Extraction
– Matlab : wpdec function, the wavelet ‘bior5.5’.
DISP Lab, GICE, NTU
16 /22
Improved ECG Signal Analysis Using
Wavelet and Feature.
• Feature Extraction
– Energy
1 N
2
E( j)n 
(
x

m
)

i
N  1 i 1
– Normal Energy
E( j )norm_ n 
E( j)n
E( j )12  E( j ) 22    E( j ) 2n
– Entorpy
N
Ent( j )log_ n   log(xi2 )
i 1
DISP Lab, GICE, NTU
17 /22
Improved ECG Signal Analysis Using
Wavelet and Feature.
• Feature Extraction
– Clustering
DISP Lab, GICE, NTU
18 /22
Improved ECG Signal Analysis Using
Wavelet and Feature.
• Method 1
wavelet: bior5.5, decomposition level: 1 and 3 with Method 1(●: normal
beats, □: atrial premature beats, ○ : premature ventricular contractions)
DISP Lab, GICE, NTU
19 /22
Improved ECG Signal Analysis Using
Wavelet and Feature.
• Method 2
wavelet: bior5.5, decomposition level: 1 and 3 with Method 2(●: normal
beats, □: atrial premature beats, ○ : premature ventricular contractions)
DISP Lab, GICE, NTU
20 /22
Conclusion
• Wavelet analysis is widely used in many
application. Because it provides both time and
frequency information, can overcome the
limitation of Fourier transform.
• We can learn about the wavelet transform which
is able to detect beat rate of signals and to classify
the difference of signals.
• We also use the wavelet transform on the
other beat rate detection.
DISP Lab, GICE, NTU
21 /22
Reference
• Chui, C.K. (1992). An Introduction to Wavelets. Academic
Press, San Diego, CA.
• S.S. Joshi, C.V. Ghule, "DWT Based Beat Rate Detection in
ECG Analysis," Proc. of IEEE International Conference and
Workshop on Emerging Trends in Technology(ICWET 2010),
pp. 765-769, 2010.
• A. Matsuyama, M. Jonkman, F. de Boer, ”Improved ECG
Signal Analysis Using Wavelet and Feature Extraction”,
Methods Inf. Med.,vol. 46, pp.227-230, 2007
DISP Lab, GICE, NTU
22 /22
Q&A
DISP Lab, GICE, NTU
Download