DETECTION OF DISEASES USING ECG

T.C.
DOKUZ EYLUL UNIVERSITY
ENGINEERING FACULTY
ELECTRICAL & ELECTRONICS ENGINEERING
DEPARTMENT
DETECTION OF DISEASES USING ECG SIGNAL
Final Year Project Report
by
Serhat DAĞ
February, 2017
İZMİR

DETECTION OF DISEASES USING ECG SIGNAL
SERHAT DAĞ
(about.serhat@gmail.com)

I
ABSTRACT
In recent years, electrocardiogram (ECG) has been used as main method for the diagnosis of heart
disease. The purpose of this study is to design a decision algorithm which automatically detect heart
diseases. With this project, arrhythmia diseases will be detected more easily. This program can also be
used in education for physicians.
In this project , peak classification approach is used in ECG signal for determining various diseases .
As known the amplitudes and duration values of P-Q-R-S-T peaks determine the functioning of heart of
human. Therefore duration and amplitude of all peaks are found. R-R and P-R intervals are calculated.
Finally, we have obtained the necessary information for disease detection
This project is implemented by using MATLAB software. An interface was created to easily select
and process the signal. “.dat” format is used the for ECG signal data. We have detected bradycardia
and tachycardia.

II
ÖZET
Son yıllarda, kalp hastalığının tanısında ana yöntem olarak elektrokardiyogram (EKG)
kullanılmaktadır. Bu çalışmanın amacı, kalp hastalıklarını otomatik olarak tespit eden bir karar
algoritması tasarlamaktır. Bu proje sayesinde aritmi hastalıkları daha kolay tespit edilecektir. Bu
program aynı zamanda doktorlar için eğitimde de kullanılabilir.
Bu projede, çeşitli hastalıkların belirlenmesinde EKG sinyalinde pik sınıflandırma yaklaşımı
kullanılmaktadır. Bilindiği gibi P-Q-R-S-T piklerinin genlik ve süre değerleri insan kalbinin işleyişini
belirler. Bunun için tüm piklerin genliği ve süresi bulunur. R-R ve P-R aralıkları hesaplanır. Son olarak
hastalık tespiti için gerekli bilgileri elde etmiş oluruz
Bu proje MATLAB yazılımı kullanılarak gerçekleştirilmiştir. Sinyali kolayca seçme ve işleme için
bir arayüz oluşturuldu. EKG sinyal verileri için ".dat" biçimi kullanıldı. Bradikardi ve taşikardi
hastalıklarının tespiti yapıldı.

III
TABLE OF CONTENTS
ABSTRACT ............................................................................................................................................I
ÖZET..................................................................................................................................................... II
TABLE OF CONTENTS....................................................................................................................III
LIST OF FIGURES.............................................................................................................................IV
1.INTRODUCTION.............................................................................................................................. 1
1.1 GENERAL OVERVIEW..................................................................................................................... 1
1.2. THE ELECTROCARDIOGRAM ......................................................................................................... 1
1.3. DATA ............................................................................................................................................ 3
2.ARRHYTMIA .................................................................................................................................... 3
2.1. NORMAL SINUS RHYTHM ............................................................................................................. 4
2.2. SINUS TACHYCARDIA.................................................................................................................... 4
2.3. SINUS BRADYCARDIA ................................................................................................................... 4
2.4.HEART BLOCK ............................................................................................................................... 4
3. ALGORITHM OF PROGRAM...................................................................................................... 5
3.1. WAVELET TRANSFORM ................................................................................................................ 6
3.2 PEAKS DETECTION ....................................................................................................................... 11
4. DEMONSTRATION OF PROGRAM .......................................................................................... 11
5. COST ANALYSİS........................................................................................................................... 15
6. WORK PLAN.................................................................................................................................. 15
7. CONCLUSION................................................................................................................................ 16
8. REFERENCES................................................................................................................................ 17
9.APPENDIX-I .................................................................................................................................... 18
10.APPENDIX-II................................................................................................................................. 28
11.APPENDIX-III ............................................................................................................................... 29
12.APPENDIX-IV................................................................................................................................ 30
13.APPENDIX-V................................................................................................................................. 32

IV
LIST OF FIGURES
FIGURE 1-REPRESENTS A TYPICAL ECG WAVES ..................................................................................... 1
FIGURE 2-. SHAPES OF THE ELECTROCARDIOGRAM ................................................................................ 2
FIGURE 3-THE HEART’S ANATOMY ......................................................................................................... 3
FIGURE 4-DESCRIPTION OF THE ALGORITHMS ........................................................................................ 5
FIGURE 5-SCALING FUNCTION FOR WAVELET DAUBECHIES 8 ................................................................ 7
FIGURE 6-SCALING FUNCTION COEFFICIENT FOR DB8............................................................................. 8
FIGURE 7-WAVELET FUNCTION FOR WAVELET DAUBECHIES 8 (DB8) .................................................... 8
FIGURE 8-WAVELET FUNCTION COEFFICIENT FOR DB8.......................................................................... 9
FIGURE 9-RELATIONSHIP BETWEEN DIGITAL WAVELET COEFFICIENTS ................................................ 10
FIGURE 10-THREE LEVEL WAVELET DECOMPOSITION TREE ................................................................. 10
FIGURE 11-APPROXIMATION COEFFICIENT OF SIGNAL AT LEVEL 4...................................................... 11
FIGURE 12-DEMONSTRATION OF PROGRAM.......................................................................................... 12
FIGURE 13-ORIGINAL SIGNAL (PROPERTIES OF SIGNAL IS DESCRIBED IN 1.3 DATA).......................... 12
FIGURE 14- COEFFICIENTS OF SIGNAL AT 1 LEVEL DECOMPOSITION.................................................. 13
FIGURE 15-COEFFICIENTS OF SIGNAL AT 2 LEVEL DECOMPOSITION.................................................... 13
FIGURE 16-COEFFICIENTS OF SIGNAL AT 3 LEVEL DECOMPOSITION ................................................... 14
FIGURE 17-COEFFICIENTS OF SIGNAL AT 4 LEVEL DECOMPOSITION ................................................... 14
FIGURE 18-WORK PLAN ........................................................................................................................ 15
FIGURE 19-WAVELET FAMILIES............................................................................................................ 28
FIGURE 20-ANALYSIS RESULTS OF THE SIGNALS IN THE DATA FOLDER ............................................... 29
FIGURE 21-APPROXIMATION COEFFICIENTS OF SIGNAL LEVELS FOR DB8........................................... 30
FIGURE 22-DETAIL COEFFICIENTS OF SIGNAL LEVELS FOR DB8 .......................................................... 31
FIGURE 23- ATR. FILE FOR 101.DAT FILE............................................................................................... 32
FIGURE 24-HEADER FILE FOR 101.DAT FILE.......................................................................................... 32

1
1.INTRODUCTION
1.1 General Overview
The population increases day by day. Therefore people don't get enough benefit from health services.
It is impossible for patients to be kept under the supervision of a physician at all times. Speed and time
are very important for physicians. This project is intended to help doctors to analyze the ECG signal.
Doctors will save time thanks to program. This program can also be used for education for.
1.2. The Electrocardiogram
Open and see method were used for diseases of internal organs by physician. It meant that people
were always suffering. This was also a very difficult situation for both physicians and patients.
Technology has improved. Now doctors use non-invasive methods. For example urine and blood tests.
Various signals are taken from the organs in the human body. The information generated by these signals
is in this group. For examples EEG (electrical activity of the brain, electroencephalogram), EMG
(electrical activity of the muscles, electromyogram). and ECG. ECG signal analysis was performed in
this project. Detailed information about this signal is given below.
The electrical current is spread through the body when the heart undergoes depolarization (The
electrical activation of the tissues is positive) repolarization (The discharge of electrical charge of the
tissues ) [1]. This electrical activity generated by heart. It can be measured by an array of electrodes
placed on the body surface. These records are called electrocardiograms (ECG or EKG).
Over the past few years, there has been an increased trend toward processing of the electrocardiogram
(ECG) using microcomputers. A survey of literature in this reach area indicates that system based on
microcomputers can perform needed medical services in extremely efficient manner [2].
Figure 1: Represents a typical ECG waves. [2]

2
The heart beat period of time curve (typical ECG waveforms) is examined in figure 1. There is a
sharpness in the figure. Periodic sharpness can be seen in figure 2. These sharpness associated with the
contraction of the ventricular. These are called “depolarization” [3]. Muscle fibers lost resting potential
during depolarization [2]. Signal prepared to go slightly negative deviation as a rule, it continues with a
big positive taper [2]. Followed by a second negative deviation income. The cons traction of the
ventricles is called “QRS complex”. A oscillation is seen after QRS complex. This oscillation also is
called ST range or ST wave. Now the potential difference is immeasurable. A zero line is drawn at this
time is called the isoelectric line. This is followed by T-wave ( representing the depolarization). The P
wave indicates that they are in the excited state to the atrium. Finally U waves is not always seen [1]. It
is typically small . U waves are thought to represent repolarization of the papillary muscles. A ECG
signal can be examine in figure 2 (This is a theoretical signal. The actual ECG signal is not as smooth
as this signal) .
Figure 2 . Shapes of the electrocardiogram : signal heart might produce assuming a 4000 Hz
sampling rate. The heart rate for this signal is approximately 89 beats per minute, and the peak
voltage of the signal is 3.5 millivolts [4]
An electrocardiogram is a measurement of the electrical activity of the heart muscle which can be
obtained from the surface of the skin and from different angles like Figure 3. When the heart muscle
contracted and pumping the blood for all parts of body, action potentials will be released through the
mechanical process within the heart muscle which leads to electrical activity.

3
Figure 3: The heart’s anatomy with waveforms from different specific part of the heart . [3]
1.3. Data
ECG signals are collected from Physionet MIT-BIH arrhythmia database. ECG signals are
described by a text header file (.hea), a binary file (.dat) and a binary annotation file (.atr). Header file
consists of detailed information such as number of samples, sampling frequency, format of ECG signal,
type of ECG leads and number of ECG leads, patient’s history and the detailed clinical information.[5]
In binary data signal file, raw ECG recordings were sampled at 360 Hz with an 12-bit resolution over
the 10 mV range. Binary data file store raw data signals (12 bits used for each sample). Binary data
files are created with reference to header files [5]. The signal is stored in 212 format. Other files are .atr
files (annotation file). “.atr ” files contain some comments for the record files. So which storages some
experts note about the analysis of signal quality results [6].
2.ARRHYTMIA
For a normal healthy person the ECG comes off as a nearly periodic signal with depolarization
followed by repolarization at equal intervals(as figure 2). However, sometimes this rhythm becomes
irregular.
Cardiac arrhythmia (also dysrhythmia) is a term for any of a large and heterogeneous group of
conditions in which there is abnormal electrical activity in the heart. The heart beat may be too fast or
too slow, and may be regular or irregular. Arrthymia comes in varieties. It may be described as a flutter
in chest or sometimes “racing heart” [7]. The diagnosis of Arrthymia requires Electrocardiogram. By
studying ECG, Doctors can diagnose the disease and prescribe the required medications. Most of time
arrythmia’s are harmless and happen in healthy people free of heart disease. However, some abnormal
heart rhythms can be serious or even deadly [8]. Having other types of heart disease an also increase the
risk of arrhytmia’s.

4
2.1. Normal Sinus Rhythm
The first measurement is known as the "P-R interval" and is measured from the beginning of the
upslope of the P wave to the beginning of the QRS wave. This measurement should be between 0.12
and 0.20 seconds(120-200 ms)[7]. Duration of PR is represented “res” in MATLAB code.
When measuring the PR interval, the lead with the longest PR interval should be chosen. (In some
leads, the initial part of the PR interval may be isoelectric. This may be misinterpreted as a short PR
interval) [9] . The R-R intervals are constant; the rhythm is regular for normal sinus rhythm. Duration
of RR interval should be between 480 and 600 ms [7] ( Duration of RR interval is represented res_2
in MATLAB codes)
2.2. Sinus tachycardia
Generally, occurs normally in exercises and stress. Other causes may be under lying medical
problems (Anemia, Fever ,Blood loss…[10]) Electrical signal is faster than usual. The heart rate is
fast. But heart beats is properly. Therefore RR interval is constant and regular. Due to the above
reasons, duration of RR peak should be shorter than normal sinus (it is between 0.45 and 0.48
second). PR interval is between 0.12–0.20 seconds.
2.3. Sinus Bradycardia
Electrical signal is slower than usual. The heart rate is slower. But heart beats is properly. Therefore
RR interval is constant and regular. PR interval is between 0.12–0.20 seconds. Causes of bradycardia
are Sleep, hypothermia, some drugs etc. [10] Duration of RR peak should be bigger than tachycardia
and sinus rhythm.
2.4.Heart Block
Heart block is an abnormal heart rhythm where the heart beats too slowly. It's slower than
bradycardia. PR interval is bigger than 0.20 second [9].

5
3. ALGORITHM OF PROGRAM
Figure 4: Description of the Algorithms
In this project we will implement a algorithm in MATLAB. The program will firstly extract the
characteristics of ECG. After that we will find the location and amplitude of ECG signal. That's why
we can find the problem that cause to patient.
Apply Wavelet Transform for Decomposed Signal
R peak detection (Find the maximum value of ECG signal and
locate Rloc )
P peak detection (Using window Rloc-90 to Rloc-10, find the
maximum)
Q peak detection (The minima in the window of Rloc-40 to
Rloc-10 )
50)
S Peak Detection (The minima in the window of Rloc+5 to
Rloc+40)
T Peak Detection (Using window of Rloc+25 to Rloc+90, find
the maximum)
Calculate PR and PR Intervals
Decision for Patient Healthy
or not Healthy
Take a RAW ECG Signal
From Data Set

6
3.1. Wavelet Transform
Recently, wavelets transform is applied to electro cardiology for several problems . Wavelet
transform is used for including data compression, analysis of ventricular late potentials, and the
detection of ECG characteristic points [11]. The wavelet transform can separate high frequency
component an low frequency component in time domain. In addition wavelet transform is able to
decompose signals at various resolutions. It allows accurate feature extraction from non-stationary
signals like ECG [11] .
The Continuous Wavelet Transform of a signal is defined as f (t), and it is represented as (the sum
of over all time of the signal multiplied by scaled [11] ):
W(a, b) = ∫ f(t)ᴪa,b (t)
+∞
−∞
dt ………… ( 3.1)
where
ᴪa,b(t) =
1
√a
ᴪ∗
(
t−b
a
) ……….( 3.2)
Where * denotes complex conjugation and , ᴪ∗
(
𝑡−𝑏
𝑎
) is a window function called the mother
wavelet, a’ is a scale factor ,b’ is also a translation factor [12].
Wavelets are defined by the wavelet function ψ(t) (i.e. the mother wavelet) and scaling function φ(t)
(also called father wavelet) in the time domain.
Let a=𝑎0
−𝑟
, b=ka0
−r
b0 and a0 = 2 , and b0 = 1 from dyadic transformation (dyadic
transformation are samples of wavelet transforms following a geometric sequence of ratio 2). Value of
j and k is integer and t is a variable in the continuous space (j is number of level and k is shifting value
).
We can define scaling and wavelet function as below
Φj,k(t) = 2
j
2ΦN(2j
t − k) ……(3.3) , ᴪj,k(t) = 2
j
2ᴪ(2k
t − k) ……….(3.4)
We define wavelet and scaling function Φj,k(t), ᴪa,b(t) ∈ L2
(R), which is limited in time domain.
The continuous wavelet transform is the sum of the signal multiplied by the shifted and scaled versions
of the main wavelet in the time domain. As a result of these operations, wavelet coefficients are obtained
depending on the scale and position.
In this project we use discrete wavelet transform with ‘wavedec’ function. Because , discrete wavelet
transform is most practical . It provides enough information for signal. It offers a significant reduction
in computation time, it mean it is faster .

7
𝑊ᴪ(j, k)=
1
√M
∑ f(n)ᴪj,k(n)𝑛 …..(3.5) and 𝑊𝛷(𝑗0, k)=
1
√ 𝑀
∑ f(n)Φj0,k(n)n …(3.6)
j≥ 𝑗0 and n∈ 𝑍
1
√ 𝑀
is normalization constant.
Φ 𝑁(t) = ∑ hΦ(n)√2 Φ(2t − n)𝑛 …….. (3.7) and
ᴪ(t)= ∑ hᴪ(n)√2 Φ(2t − n)𝑛 ………(3.8)
where
𝛷(𝑡) = {
1 when 0 ≤ t ≤ 1
0 otherwise
…..(3.9) and ᴪ(𝑡) = {
1 when 0 ≤ t ≤ 1/2
−1 when
1
2
≤ t ≤ 1
0 otherwise
..(3.10)
ℎᴪ(𝑛) and ℎ 𝛷(𝑛) is wavelet and scaling function coefficient. These coefficients indicate
characteristic characteristics of wavelet and scaling function. There are several wavelet families like
Harr , Daubechies , Coeiflets, Symlets, Morlet, Mexican Hat, Meyer ect [12] ( wavelet form can be
examined in appendix 2). Selecting a wavelet function which closely matches the signal to be processed
is of most importance in wavelet applications Daubechies is chosen (d8 scaling function is used) .
Daubechies wavelet family are similar in shape to QRS complex and their energy spectrum are
concentrated around low frequencies [13] .
Figure 5: Scaling function for wavelet Daubechies 8 (db8) [14]

8
Figure 6: Scaling function coefficient for db8 [14]
Figure 7: Wavelet function for wavelet Daubechies 8 (db8) [14]

9
Figure 8: Wavelet function coefficient for db8 [14]
If we use (3.4) and (3.8) for (3.7)
𝑊ᴪ(j, k)=
1
√M
∑ f(n)ᴪj,k(n)𝑛 where , ᴪj,k(n) = 2
j
2ᴪ(2j
n − k)
𝑊ᴪ(j, k)=
1
√M
∑ f(n)𝑛 2
j
2ᴪ(2j
n − k)….(*) after we can write by 3.8
ᴪ(2j
n − k)= ∑ ℎᴪ(𝑛′)√2 𝛷(2j+1
𝑛 − 2𝑘 − 𝑛′)𝑛′ …(**) After that we can write into * equation
𝑊ᴪ(j, k)=
1
√M
∑ f(n)𝑛 2
j
2 ∑ ℎᴪ(𝑛′)√2 𝛷(2j+1
𝑛 − 2𝑘 − 𝑛′)𝑛′
Let n’=m-2k
𝑊ᴪ(j, k)=
1
√M
∑ f(n)𝑛 2
j
2 ∑ ℎᴪ(𝑚 − 2𝑘)√2 𝛷(2j+1
𝑛 − 𝑚)𝑚
𝑊ᴪ(j, k)=
1
√M
∑ ℎᴪ(𝑚 − 2𝑘)𝑚 ∑ f(n)2
j+1
2 𝛷(2j+1
𝑛 − 𝑚)𝑛
𝑊𝛷(j+1, m) =
1
√M
∑ f(n)2
j+1
2 𝛷(2j+1
𝑛 − 𝑚)𝑛 (From equation * )
So we can write 𝑊ᴪ(j, k)= ∑ ℎᴪ(𝑚 − 2𝑘)𝑚 𝑊𝛷(j+1, m) This means,
𝑊ᴪ(j, k)= ℎᴪ(−𝑛)* 𝑊ᴪ(j+1, m) and we can apply same operation for approximation coefficient
Therefore 𝑊𝛷(j, k)= ℎ 𝛷(−𝑛)* 𝑊𝛷(j+1, m) for n=2k and k≥ 0

10
The last equations determine relationship between digital wavelet coefficients of adjacent scales.
Wavelet coefficient is scaling and scaling time-reversed for each step. Block diagram can be examined
in figure 9.
Figure 9: Relationship between digital wavelet coefficients [11]
After wavelet transform is applied two component emerge (with wavedec function in MATLAB).
First component is approximation components (it has lower frequency function) and high scale
information other component is detail components have higher frequency and less information.
Approximation components (coefficient) occurs after low band pass filter (associated with scaling
function). Detail coefficients can be obtained after high band pass filter (associated with wavelet
transform).
The original signal is filtered by half band low pass and high pass filter. This is done for each
coefficient. The width of the filter will be reduced by half for each level. These operations consist of
decomposition tree. Operations can be repeated many times. In this project the signal decompose at level
8. We can see three level wavelet decomposition tree in figure 10.
Figure 10: Three level wavelet decomposition tree [13]
𝒂𝒋+𝟏(𝒏) = ∑ 𝒉(𝟐𝒏 − 𝒌)𝒂𝒋(𝒏)𝟐𝒏
𝒌=𝟏 And 𝒅𝒋+𝟏(𝒏) = ∑ 𝒈(𝟐𝒏 − 𝒌)𝒂𝒋(𝒏)𝟐𝒏
𝒌=𝟏

11
3.2 Peaks detection
Peaks of the R waves have a largest amplitude (we can see in figure 1). After the decompose the signal
high frequency component is removed. R peaks is noticeable even if R peak are deformed.
Approximation coefficient signal of signal at 4 level can be examined on figure 11.
Figure 11: Approximation coefficient of signal At Level 4 (101.dat signal is used)
The locations of the maximum points of the signal were determined. Then these locations of R peaks
were found in the original signal. Finally, amplitude of R peaks were found for original signal.
Detection of R peaks is very important. Because other peaks are found according to R peaks. After R
peak detection we need to select a window of Rloc‐90 to Rloc‐10 and find the maximum, than that
maxima is P peak [15]. For location of Q peak we must select a window of Rloc‐40 to Rloc‐10 and find
the maximum [16]. To find S peak select a window of Rloc+5 to Rloc+40 and find the maximum [16].
Lastly we select a window of Rloc‐100 to Rloc‐50 and find the maximum for T peak [16]. Generally
window is used to search signal because of down sampling. Down sampling deviate location of wave
.
4. DEMONSTRATION OF PROGRAM
When we run the program, we encounter a interface which has multiple section. Firstly we select a
raw ECG signal from data file. Second step we need enter two value. First value represent number of
sample and other value is window size. Window size also represent part of sample to use. After that we
click on Apply Wavelet button. Wavelet transform apply to signal. That’s why we can see
approximation coefficient. Next step we click on Processing Button and we calculate P,Q,R ,S,T

12
(durations and amplitude) peaks. Finally , to read result on the interface we need to click on Result
Button.
Figure 12:Demonstration of Program ( 101.dat signal is used)
Approximation coefficient of 101.dat signal is drawn below. We can see R peak noticeable (4. level is
enough to consider R peaks )
Figure 13: Original signal (properties of signal is described in 1.3 DATA)

13
Figure 14: Coefficients of Signal At 1 Level Decomposition Very similar to the original
signal.
Figure 15: Coefficients of signal At 2 Level Decomposition , Better than the original
signal. But not smooth enough

14
Figure 16: Coefficients of Signal At 3 Level Decomposition , It still is not smooth
Figure 17: Coefficients of Signal At 4 Level Decomposition
High-frequency compounds are removed when the graphics are examined. As a result we can see R
peaks. Other important point is y axis of signal increased and x axis is reduced by scaling function (if
we examine axis we can notice that).

15
5. COST ANALYSİS
ECG Signal Compression is a software project that is created and developed via MATLAB. For this
project, MATLAB is provided by free license from the university to the students. For this project,
personal computer HP k53sv series with 2.4 Ghz processor, 6GB RAM, 500 GB Hard disc space, 5400
rpm hard disc driver NVIDIA GeForce GT 540M + Intel GMA HD graph card is used.
6. WORK PLAN
Figure 18 :Work Plan
Each step of the plant is interconnected . We have to pay attention to this every step of the way. The
table was created entirely according to the effort I spent.

16
7. CONCLUSION
Over the past few years, importance for human life has been known heart disease. ECG is the
main method is used in the diagnosis of heart disease. Purpose of project that the analysis of heart
disease with ECG signal. This project is implemented by using MATLAB software and on the basis of
the wave form obtain the peaks are classified and are used for the detection of the disease.
Wavelet decomposition applied for decomposed ECG signal. All R peaks are noticeable.
Daubechies wavelet family are similar in shape to QRS complex. Therefore Daubechies wavelet is
chosen for prototype wavelet used.
After wavelet transform is applied the ECG, we detect R peaks. The location of the other peaks
was determined with location of R peaks. Lastly , we calculated RR and PR intervals. These interval
is very important for detecting diseases. These intervals are important for determining the type of illness.
For example , electrical signal is faster than usual for tachycardia. Therefore duration of RR peak
should be shorter than normal sinus. PR interval is also important for us. It was a fixed interval in the
detection of tachycardia and bradycardia. This value was high in some patients' data. Thus Heart block
disease is added
It consist of effective, simple, non-invasive, low cost procedures for the diagnosis of various
disorders physicians for their for their decision and education. This project will reduce the workload
of the doctors. It would be easier for a doctor to verify later.

17
8. REFERENCES
[1] H. Al-jobouri, “Designation of interface and software program as a base for ECG monitoring and
signal analysis,” M.S. thesis, Dept. of Medical Eng., AlNahrain Univ., Baghdad, Iraq, 2004
[2] Advances in Cardiac Signal Processing Rajendra Acharya U, Jasjit S. Suri, Jos A.E. Spaan an d S
.M. Krishnan, pp.3-13
[3] P. Gustafsso and J. Hagblad, “Miniaturised multi-channel wireless ECG connected to graphical
computer interface using bluetooth,” M.S. thesis, Dept. of Comp. Sci. and Electronics, Mälardalen
University, Sweden, 2007.
[4] http://www.mathworks.com/matlabcentral/fileexchange/35328, 01/02/2017
[5] THE MIT-BIH Arrhythmia Database On CD_ROM AND Software For Use With it, George B.
Moody and Roger G. Mark
[6] https://www.physionet.org/physiobank/database/mitdb/ ,01/02/2017
[7] A Research Work on Identification of Cardiac Disorders using LabVIEW Mohan Kumar
[8] F.A.Davis, “ECG_NOTES,” 2005
[9] The value of the electrocardiogram in the differential diagnosis of a tachycardia with a widened QRS
complex ,Hein J.J. Wellens. Author
[10] Diagnosis and Treatment of Sick Sinus Syndrome VICTOR ADÁN, M.D., Angel Medical Center,
Franklin, North Carolina LOREN A. CROWN, M.D., University of Tennessee Health Sciences Center,
Covington, Tennessee
[11] A Tutorial of the Wavelet Transform Chun-Lin, Liu February 23, 2010
[12] Digital Image Processing (3rd Edition) 3rd Edition by Rafael C. Gonzalez, pp.477-493
[13] ECG Feature Extraction Using Daubechies Wavelets S. Z. Mahmoodabadi, A. Ahmadian, M. D.
Abolhasani, Tehran University of Medical Sciences (TUMS), Tehran, Iran
[14] A Wavelet Transform Method to Detect P and S-Phases in Three Component Seismic Data Salam
Al-Hashmi, Adrian Rawlins, Frank Vernon
[15] Automatic Delineation of ECG Characteristics Points using Window Search & Multi-resolution
Wavelet Transform approach Amit Kumar Manocha, Mandeep Singh
[16] EKG Sinyallerinde Gürültü Gidermede Ayrık Dalgacık Dönüşümünde Farklı Ana Dalgacıkların
Ve Ayrıştırma Seviyelerinin Karşılaştırılması Cengiz Tepe, Hatice Sezgin
[17] Empirical Wavelet Transform-based Detection of Anomalies in ULF Geomagnetic Signals
Associated to Seismic Events with a Fuzzy Logic-based System for Automatic Diagnosis ,Omar Chavez
Alegria, Martin Valtierra-Rodriguez,

18
9.APPENDIX-I
%--------------Final Project------------------------------------------------
% DETECTION OF DISEASES USING ECG SIGNAL
%
% SERHAT DAĞ 2011502024
% February, 2016
%------------------------------------------------------------------------------
function varargout = ecg_wavelet(varargin)
% ECG_WAVELET MATLAB code for ecg_wavelet.fig
% ECG_WAVELET, by itself, creates a new ECG_WAVELET or raises the existing
% singleton*.
%
% H = ECG_WAVELET returns the handle to a new ECG_WAVELET or the handle to
% the existing singleton*.
%
% ECG_WAVELET('CALLBACK',hObject,eventData,handles,...) calls the local
% function named CALLBACK in ECG_WAVELET.M with the given input arguments.
%
% ECG_WAVELET('Property','Value',...) creates a new ECG_WAVELET or raises the
% existing singleton*. Starting from the left, property value pairs are
% applied to the GUI before ecg_wavelet_OpeningFcn gets called. An
% unrecognized property name or invalid value makes property application
% stop. All inputs are passed to ecg_wavelet_OpeningFcn via varargin.
%
% *See GUI Options on GUIDE's Tools menu. Choose "GUI allows only one
% instance to run (singleton)".
%
% See also: GUIDE, GUIDATA, GUIHANDLES
% Edit the above text to modify the response to help ecg_wavelet
% Last Modified by GUIDE v2.5 27-Oct-2016 01:03:26
% Begin initialization code - DO NOT EDIT
gui_Singleton = 1;
gui_State = struct('gui_Name', mfilename, ...

19
'gui_Singleton', gui_Singleton, ...
'gui_OpeningFcn', @ecg_wavelet_OpeningFcn, ...
'gui_OutputFcn', @ecg_wavelet_OutputFcn, ...
'gui_LayoutFcn', [] , ...
'gui_Callback', []);
if nargin && ischar(varargin{1})
gui_State.gui_Callback = str2func(varargin{1});
end
if nargout
[varargout{1:nargout}] = gui_mainfcn(gui_State, varargin{:});
else
gui_mainfcn(gui_State, varargin{:});
end
% End initialization code - DO NOT EDIT
% --- Executes just before ecg_wavelet is made visible.
function ecg_wavelet_OpeningFcn(hObject, eventdata, handles, varargin)
% This function has no output args, see OutputFcn.
% hObject handle to figure
% eventdata reserved - to be defined in a future version of MATLAB
% handles structure with handles and user data (see GUIDATA)
% varargin command line arguments to ecg_wavelet (see VARARGIN)
% Choose default command line output for ecg_wavelet
handles.output = hObject;
% Update handles structure
guidata(hObject, handles);
% UIWAIT makes ecg_wavelet wait for user response (see UIRESUME)
% uiwait(handles.figure1);
flag =1;
% --- Outputs from this function are returned to the command line.
function varargout = ecg_wavelet_OutputFcn(hObject, eventdata, handles)

20
% varargout cell array for returning output args (see VARARGOUT);
% hObject handle to figure
% Get default command line output from handles structure
varargout{1} = handles.output;
% --- Executes on button press in Select.
function Select_Callback(hObject, eventdata, handles)
% hObject handle to Select (see GCBO)
% Hint: get(hObject,'Value') returns toggle state of Select
global nfile A flag
[fil pth]=uigetfile('.dat','Enter File Name Wich You Want to Work')
file=[pth fil];
fid=fopen(file,'r');
nfile=fread(fid);
nfile=imresize(nfile,[183 1]);
axes(handles.axes1)
plot(nfile,'-'), title('Original Signal')
figure(1),plot(nfile,'-'), title('Original Signal')
z=zeros(100,1);
A=[z;nfile;z]; % Add 100 zeros before and after the signal
axes(handles.axes2)
plot(A,'-'),title('Original Signal With Pad')
figure(2),plot(A,'-'),title('Original Signal With Pad')
flag=2;
% --- Executes on button press in Apply.

21
function Apply_Callback(hObject, eventdata, handles)
% hObject handle to Apply (see GCBO)
% Define global variables.
global nfile Qamp_mv Ramp_mv Samp_mv Tamp_mv P_1 Pamp_mv A flag c
%-----------Perform Discrete Wavelet Transform (daubechies 8)---------------
if flag==2
flag=3;
[c,l]=wavedec(nfile,8,'db8');
ca1=appcoef(c,l,'db8',1);
figure(2),plot(ca1,'-k'),title('Cofficients of Signal At l Level Decomposition')
figure(3),plot(ca2,'-k'),title('Cofficients of Signal At 2 Level Decomposition')
figure(4),plot(ca3,'-k'),title('Cofficients of Signal At 3 Level Decomposition')
figure(5),plot(ca4,'-r'),title('Cofficients of Signal At 4 Level Decomposition')
flag=3 ;
else
errordlg('First select ECG Signal')
end
% --- Executes on button press in Exit.
function Exit_Callback(hObject, eventdata, handles)
% hObject handle to Exit (see GCBO)
clc
close all
clear all
% --- Executes on button press in Result.

22
function Result_Callback(hObject, eventdata, handles)
% hObject handle to Result (see GCBO)
if flag==4
%% Setting handles q p r s t assigns the specified value of the handles to the specified property
Q,P,R,S,T
set(handles.q,'string',['Q=' num2str(Qamp_mv)])
set(handles.p,'string',['P=' num2str(Pamp_mv)])
set(handles.r,'string',['R=' num2str(Ramp_mv)])
set(handles.s,'string',['S=' num2str(Samp_mv)])
set(handles.t,'string',['T=' num2str(Tamp_mv)])
set(handles.textll,'string',['PERSON HEALT:-' P_1])
else
errordlg('Calculate values/apply wavelet') % Create error dialog box
end
end
function edit1_Callback(hObject, eventdata, handles)
% hObject handle to edit1 (see GCBO)
% Hints: get(hObject,'String') returns contents of edit1 as text
% str2double(get(hObject,'String')) returns contents of edit1 as a double
% --- Executes during object creation, after setting all properties.
function edit1_CreateFcn(hObject, eventdata, handles)
% handles empty - handles not created until after all CreateFcns called
% Hint: edit controls usually have a white background on Windows.

23
% See ISPC and COMPUTER.
if ispc && isequal(get(hObject,'BackgroundColor'), get(0,'defaultUicontrolBackgroundColor'))
set(hObject,'BackgroundColor','white');
end
function edit2_Callback(hObject, eventdata, handles)
% Hints: get(hObject,'String') returns contents of edit2 as text
% str2double(get(hObject,'String')) returns contents of edit2 as a double
% --- Executes during object creation, after setting all properties.
function edit2_CreateFcn(hObject, eventdata, handles)
% handles empty - handles not created until after all CreateFcns called
% Hint: edit controls usually have a white background on Windows.
% See ISPC and COMPUTER.
if ispc && isequal(get(hObject,'BackgroundColor'), get(0,'defaultUicontrolBackgroundColor'))
set(hObject,'BackgroundColor','white');
end
% --- Executes on button press in Proccesing.
function Proccesing_Callback(hObject, eventdata, handles)
% hObject handle to Proccesing (see GCBO)
flag=4;
per=get(handles.edit1,'string');

24
per=str2double(per);
if per>0 & per <=60
per=per;
else
errordlg('ENTER PERTAGE IN RANGE');
end
wndw=get(handles.edit2,'string');
wndw=str2double(wndw);
if wndw>2 & wndw <=20
wndw=wndw;
else
errordlg('ENTER window IN RANGE');
end
%-------------R Peak Detection--------
m1=max(c)*(per/100);
p=find(c>=m1);
p1=p;
p2=[];
last=p1(1);
sig_fnd=length(p1);
for i=2:sig_fnd
if (p1(i)>(last+10))
% In this step we find R peaks which are atleast 10 samples apart
last=p1(i);
p2=[p2 last];
end
end
p3=p2*8 %Multiply the current location with 8 to get the actual scale.
R_position=[];
for i=1:length(p3)
range=[p3(i)-wndw:p3(i)+wndw]
m=max(A(range))

25
l=find(A(range)==m)
pos=range(l)
R_position=[R_position pos]
end
Ramp=round(A(R_position));
X=R_position;
Rloc=R_position;
y1=A;
for i=1:1
for j=1:length(X)
%----------------------P Peak Detection----
a=R_position(i,j)-90:R_position(i,j)-10;
m=max(y1(a));
b=find(y1(a)==m);
b=b(1);
b=a(b);
Ploc(i,j)=round(b);
Pamp(i,j)=round(m);
%------Q Peak Detection-----
a=R_position(i,j)-40:R_position(i,j)-10;
m=min(y1(a));
b=find(y1(a)==m);
b=b(1);
b=a(b);
Qloc(i,j)=round(b);
Qamp(i,j)=round(m);
%----S Peak Detection----
a=R_position(i,j)+5:R_position(i,j)+40;
m=min(y1(a));
b=find(y1(a)==m);
b=b(1);
b=a(b);
Sloc(i,j)=round(b);
Samp(i,j)=round(m);
%-------T Peak Detection----

26
a=R_position(i,j)+25:R_position(i,j)+90;
m=max(y1(a));
b=find(y1(a)==m);
b=b(1);
b=a(b);
Tloc(i,j)=round(b);
Tamp(i,j)=round(m);
end
end
clc
disp('VALUE OF AMPLITUDES(mv) OF P Q R S T ARE')
Pamp_mv=Pamp
%disp(Pamp)
Qamp_mv=Qamp
Ramp_mv=Ramp
Samp_mv=Samp
Tamp_mv=Tamp
disp('VALUE OF DURATION(ms) OF P Q R S T ARE')
Ploc_ms=Ploc
Qloc_ms=Qloc
Rloc_ms=Rloc
Sloc_ms=Sloc
Tloc_ms=Tloc
res=Ploc_ms+Qloc_ms+Rloc_ms; % P TO R İNTERVAL
res_2=Rloc_ms+Sloc_ms+Tloc_ms+Qloc_ms+Ploc_ms; % R to R interval
res
res_2
if res>=120 & res<=200
if res_2>=480 & res_2<=600
P_1='PERSON IS SUFFERING FROM BRADYCARDIA';
disp('PERSON IS SUFFERING FROM BRADYCARDIA')
elseif res_2<450 && res_2>430
P_1='PERSON IS SUFFERING FROM TACHYCARDIA';
disp('PERSON IS SUFFERING FROM TACHYCARDIA')

27
elseif res_2>450 && res_2<480
P_1='PERSON IS HEALTY';
disp('PERSON IS HEALTY')
else
P_1='Not Clear in ECG Signal';
disp('Not Clear in ECG Signal')
end
end
if res>200
P_1='PERSON IS SUFFERING FIRST DEGREE HEART BLOCK';
disp('PERSON IS SUFFERING FIRST DEGREE HEART BLOCK')
end
flag=4;

28
10.APPENDIX-II
Figure 19(a): Daubechies , Haar, Morlet, Symlets , Coiflets, Meyer wavelet function [17]
Figure 19 (b):Mexican Hat Wavelet Function [12]
Figure 19: Wavelet Families
Even if a signal is not well represented by one member of the Db family, it may still be efficiently
represented by another. Selecting a wavelet function which closely matches the signal to be processed
is of importance in wavelet applications [13]

29
11.APPENDIX-III
FILE Disease
100.dat PERSON IS HEALTY
101.dat PERSON IS SUFFERING FROM TACHYCARDIA
105.dat PERSON IS SUFFERING FROM BRADYCARDIA
109.dat Not Clear in ECG Signal
Figure 20: Analysis results of the signals in the data folder
Window size is 6 and 20 percent of the signal was used . Heart block 'could not be detected. But
Some signals have PR range close to 200 ms. PR need to be less than for detected bradycardia and
tachycardia.

30
12.APPENDIX-IV
Figure 21: Approximation Coefficients of Signal Levels for db8
The high-frequency components are removed by wavelet transform. Therefore last signal (8.Level) is
smooth and maximum value can be found. In addition effect of scaling function can be noticed.

31
Figure 22: Detail Coefficients of Signal Levels for db8
The low-frequency components are removed by wavelet transform. In addition effect of wavelet
function can be noticed.

32
13.APPENDIX-V
Figure 23: atr. File for 101.dat file (N: normal beat , Sub, Chan, Num is describe
abnormal beats ) [6]
Figure 24:header File for 101.dat file (101:signal number 2:two signal is used, 650000:
number of sample, 212:format of signal 200:gain of signals 1024 : Zero value of ADC, 955-29832
The checksums of the 650000 samples, third line is similar to second line, Last line determine sex
and age of the subject and data about the recording and drugs (Diapres is name of drug ) )[6].

DETECTION OF DISEASES USING ECG

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Viewers also liked

Viewers also liked (20)

Similar to DETECTION OF DISEASES USING ECG

Similar to DETECTION OF DISEASES USING ECG (20)

Recently uploaded

Recently uploaded (20)

DETECTION OF DISEASES USING ECG