Coherence Analysis between ECG and EEG Signals

A DISSERTATION

Submitted in partial fulfilment of the

Requirements for the award of the degree

Of

MASTER OF TECHNOLOGY

In

CONTROL AND INSTRUMENTATION ENGINEERING

By

GAVENDRA SINGH(Regd. No. 09206106)

Under the guidance of

Dr DILBAG SINGH(Associate Professor)

DEPARTMENT OF INSTRUMENTATION AND CONTROL ENGINEERING

Dr B R AMBEDKAR NATIONAL INSTITUTE OF TECHNOLOGY

JALANDHAR – 144011, PUNJAB (INDIA), JUNE 2011

CANDIDATE’S DECLARATION

I hereby declare that the work which is being presented in this dissertation entitled

“Coherence Analysis between ECG and EEG Signals ” submitted towards the partial

fulfilment of the requirements for the award of the degree of the Master of Technology in

Control and Instrumentation Engineering from Dr B R Ambedkar National Institute of

Technology Jalandhar, India, is an authentic record of my own work carried out from August

2010 to June 2011 under the supervision of Dr Dilbag Singh, Associate Professor,

Department of Instrumentation and Control Engineering, Dr B R Ambedkar National Institute

of Technology Jalandhar.

This matter in this dissertation report has not been submitted by me for of any other

degree or diploma.

Place: NIT Jalandhar Gavendra Singh

Date: June 2011

CERTIFICATE

This is to certify that the above statement made by the candidate is correct to the best of my

knowledge.

Dr Dilbag Singh

(Associate Professor)

Department of ICE

NIT Jalandhar-144011

i

Dr B R AMBEDKAR NATIONAL INSTITUTE OF TECHNOLOGY

JALANDHAR, (PUNJAB)

CERTIFICATE

This is to certify that dissertation entitled

“Coherence Analysis between ECG and EEG Signals”

Submitted By

GAVENDRA SINGH(Regd. No. 09206106)

May be accepted for the partial fulfilment for award of

Master of Technology in “Control and Instrumentation Engineering”

Internal External HODExaminer Examiner Department of ICE

Date:

ii

“Dedicated to my mother, Smt Rajvala Devi and my Father, Mr Kanchhi Singh for their

continued Inspiration, Encouragement, Love and Support”

iii

ACKNOWLEDGEMENT

At this momentous occasion of completing my research I would like to acknowledge the

contribution of all those benevolent people, I have been blessed to associate with. All the data

collection, theories, models would have failed to serve their purpose for me if blessing of the

Almighty would not have joined hands with my efforts.

My first and foremost offering of thanks goes to the architect who shaped my dream into the

reality, my guide and mentor Dr Dilbag Singh, Associate Professor, Department of

Instrumentation and Control Engineering, Dr B R Ambedkar National Institute of

Technology, Jalandhar. Perseverance, exuberance, positive approaches are just some of the

traits he has imprinted on my personality. He steered me through this journey with his

invaluable advice, positive criticism, stimulating discussions and consistent encouragement.

He took care to shine light of knowledge, when I was groping in the darkness of ignorance. If

I will stand proud of my achievements then undeniably he is the main creditor. It is my

privilege to be under his tutelage.

I express my sincere thanks to Dr A K Jain, Head, Department of

Instrumentation and Control Engineering, Dr B R Ambedkar National Institute of

Technology, Jalandhar. He provided me continuous help and guidance to complete my

dissertation.

With a grateful heart, I acknowledge the noble and gentle hand of support lent to me by Mr

Buta Singh, Research Scholar, for his valuable guidance at every step and cooperation for

data collection and analysis.

When talking about cooperation and help to complete this work how can I go without the

name of my arch-batch met throughout my journey, Mr Varun Gupta for her valuable

suggestions, consistent encouragement and to keep my approaches positive and my senier Mr

Madhwendra Nath Tripathi for his good help.

iv

Dated: June 2011 GAVENDRA SINGH

v

List of Tables and Figures

List of Tables and Figure

Table 1.1: Rhythmic brain activity

Table 1.2: Average respiratory rates, by age

Table 2.1: Experimental Hardware Setup

Table 2.2: Specification of data acquisition unit Biopac Inc. MP100

Table 2.3: ECG100C Specifications

Table 2.4: EEG100C Specifications

Table 2.5: RSP100C Specifications

Table 4.1: Different parameters of the acquired signals

Table 4.2: Signals Acquisition Settings

Table 4.3: Coherence analysis of results at different respiratory rates

Table 4.4: ECG and EEG signals from 1 to 25 subjects statistics

Table 4.5: ECG and EEG signals from 26 to 50 subjects statistics

Table 5.1: Coherence and phase coherence measure parameters for first subject

Table 5.2: Coherence and phase coherence measure parameters for second subject

Table 5.3: Coherence and phase coherence measure parameters for third subject

Table A.1: Lead Type Length Usage Note

Table A.2: TSD201 Specifications

Fig. 1.1: The Human Heart with Coronary Arteries

Fig. 1.2: Heart Valves

Fig. 1.3: Cardiac Conduction System

Fig. 1.4: The lobes and sulci of the cerebrum.

Fig. 1.5: Functional areas of the cerebrum

Fig. 1.6: Rhythmic brain activity

Fig. 1.7: Willem Einthoven, The string galvanometer that he invented in 1903.

Fig. 1.8: Experimental Setup of 12 Lead ECG Acquisition from Atria 6100

Fig. 2.1(a): Block Diagram of Multi-channel Data Acquisition System Biopac Inc. MP100

Fig. 2.1(b): Hardware Components of Multi-channel Data Acquisition System MP100

Fig. 2.2: Graph of Experimental Data Acquired using MP100 and Acqknowledge3.9.0

Fig. 2.3: Snap of Subject and Technician during Data Acquisition

Fig. 2.4: The electrode connections to the ECG100C for the measurement of Lead I

vi

List of Tables and Figures

Fig. 2.5(a): Bipolar EEG electrode leads placement

Fig. 2.5(b): International 10-20 electrode placement on the different brain regions

Fig. 2.6: The placement and connections for recording thoracic respiration effort

Fig. 2.7(a): Transform tool bar

Fig. 2.7(b): Graph window function tool bar

Fig. 2.7(c): Acquisition set up

Fig. 2.7(d): On opening the new graph, graphic journal

Fig. 2.7(e): Setting channel label

Fig. 2.7 (f): Set screen horizontal axis

Fig. 2.7(g): Set screen horizontal axis

Fig. 2.7(h): Channel selection, (i) Cursor tools

Fig. 2.7(j): Edit tool bar

Fig. 2.7(k): Clipboard

Fig. 2.7(l): Journal

Fig. 3.1: One signal lags by j unit with the other signal

Fig. 3.2(a): ECG signal having 5006 samples with sampling rate 500 samples/sec

Fig. 3.2(b): EEG Signal having 5006 samples with sampling rate 500 samples/sec

Fig. 3.2(c): Cross-correlation between the ECG signal and the EEG signal

Fig. 3.3(a): ECG signal having 5006 samples with sampling rate 500 samples/sec

Fig. 3.3(b): Auto-correlation of ECG signal

Fig. 3.3(c): EEG Signal having 5006 samples with sampling rate 500 samples/sec

Fig. 3.3(d): Auto-correlation of the EEG signal

Fig. 3.4(a): ECG signal having 5006 samples with sampling rate 500 samples/sec

Fig. 3.4(b): Auto power spectral density estimate of ECG signal

Fig. 3.4(c): EEG Signal having 5006 samples with sampling rate 500 samples/sec

Fig. 3.4(d): Auto power spectral density estimate of EEG signal

Fig. 3.5(a): ECG signal having 5006 samples with sampling rate 500 samples/sec

Fig. 3.5(b): EEG Signal having 5006 samples with sampling rate 500 samples/sec

Fig. 3.5(c): Cross-power spectral density between the ECG signal and the EEG signal

Fig. 3.6(a): ECG signal having 5006 samples with sampling rate 500 samples/sec

Fig. 3.6(b): EEG Signal having 5006 samples with sampling rate 500 samples/sec

Fig. 3.6(c): Coherence between the ECG signal and the EEG signal

vii

List of Tables and Figures

Fig. 4.1: ECG Signals at the respiratory rates. First signal in the figure 6.1 is at nearly

zero breathing rate for 9.99 seconds and similarly second and third signals at

the 10 to 12 BPM(breaths per minute) and 15 to 20 BPM

Fig. 4.2: EEG Signals respiratory rates. First signal in the Figure 6.1(a) is at nearly zero

breathing rate for 9.99 seconds and similarly second and third signals at the 10

to 12 BPM and 15 to 20 BPM

Fig. 4.3: Coherence between the ECG and EEG signals respiratory rates

Fig. 4.4: Phase Coherence between the ECG and EEG signals respiratory rates

Fig. 5.1: ECG signals and corresponding EEG signals of the first subject (S1)

Fig. 5.2(a): Coherence between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to 35 Hz)

for S1

Fig. 5.2(b): Coherence between ECG and EEG (C3-C4) for S1

Fig. 5.2(c): Coherence between ECG and EEG (P3-P4) for S1

Fig. 5.2(d): Coherence between ECG and EEG (O1-O2) for S1

Fig. 5.3(a): Coherence phase between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to

35 Hz) for S1

Fig. 5.3(b): Coherence phase between ECG and EEG (C3-C4) for S1

Fig. 5.3(c): Coherence phase between ECG and EEG (P3-P4) for S1

Fig. 5.3(d): Coherence phase between ECG and EEG (O1-O2) for S1

Fig. 5.4: ECG signals and corresponding EEG signals of the second subject (S2)

Fig. 5.5(a): Coherence between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to 35 Hz)

for S2

Fig. 5.5(b): Coherence between ECG and EEG (C3-C4) for S2

Fig. 5.5(c): Coherence between ECG and EEG (P3-P4) for S2

Fig. 5.5(d): Coherence between ECG and EEG (O1-O2) for S2

Fig. 5.6(a): Coherence phase between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to

35 Hz) for S2

Fig. 5.6(b): Coherence phase between ECG and EEG (C3-C4) for S2

Fig. 5.6(c): Coherence phase between ECG and EEG (P3-P4) for S2

Fig. 5.6(d): Coherence phase between ECG and EEG (O1-O2) for S2

Fig. 5.7: ECG signals and corresponding EEG signals of the second subject (S3)

Fig. 5.8(a): Coherence between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to 35 Hz)

for S3

Fig. 5.8(b): Coherence between ECG and EEG (C3-C4) for S3

viii

List of Tables and Figures

Fig. 5.8(c): Coherence between ECG and EEG (P3-P4) for S3

Fig. 5.8(d): Coherence between ECG and EEG (O1-O2) for S3

Fig. 5.9(a): Coherence phase between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to

35 Hz) for S3

Fig. 5.9(b): Coherence phase between ECG and EEG (C3-C4) for S3

Fig. 5.9(c): Coherence phase between ECG and EEG (P3-P4) for S3

Fig. 5.9(d): Coherence phase between ECG and EEG (O1-O2) for S3

Fig. 5.10(a): Box plot of Coherence between the ECG signals corresponding to the EEG

signals 1-EEG (Fp1-Fp2), 2-EEG (C3-C4), 3-EEG (P3-P4), 4-EEG (O1-O2)

of the First Subject (S1)

Fig. 5.10(b): Box plot of Coherence between the ECG signals corresponding to the EEG

signals of the Second Subject (S2)

Fig. 5.10(c): Box plot of Coherence between the ECG signals corresponding to the EEG

signals of the Third Subject (S3)

Fig. 5.11(a): Histogram of Coherence between the ECG signals corresponding to the EEG

signals 1-EEG (Fp1-Fp2), 2-EEG (C3-C4), 3-EEG (P3-P4), 4-EEG (O1-O2)

of the First Subject (S1)

Fig. 5.11(b): Histogram of Coherence between the ECG signals corresponding to the EEG

signals of the Second Subject (S2)

Fig. 5.11(c): Histogram of Coherence between the ECG signals corresponding to the EEG

signals of the Third Subject (S3)

Fig. 5.12(a): Box plot of Coherence phase between the ECG signals corresponding to the

EEG signals 1-EEG (Fp1-Fp2), 2-EEG (C3-C4), 3-EEG (P3-P4), 4-EEG

(O1-O2) of the First Subject (S1)

Fig. 5.12(b): Box plot of Coherence phase between the ECG signals corresponding to the

EEG signals of the Second Subject (S2)

Fig. 5.12(c): Box plot of Coherence phase between the ECG signals corresponding to the

EEG signals of the Third Subject (S3)

ix

Abbreviations and Acronyms

ABBREVIATIONS AND ACRONYMS

ANS Autonomic Nervous System

AV Atrio-Ventricular

BP Blood Pressure

BPM Beats per Minute, Breaths per Minute

BSL Biopac Student Lab

CAD Coronary Artery Disease

CFA Cardiac Field Artifact

CK Creatine Kinase

CNS Central Nervous System

CPSD Cross Power Spectrum Density

DC Direct Current

ECG Electrocardiogram

EEG Electroencephalogram

EKG Electrokardiagram

EMD Empirical Mode Decomposition

EMG Electromyogram

EOG Electrooculogram

EPS Electrophysiological Studies

EPSP Excitatory Postsynaptic Potential

ERS Event-Related Synchronization

FFT Fast Fourier Transform

GSR Galvanic Skin Resistance

HF High Frequency

HS Heart Sound

IMF Intrinsic Mode Function

IPSP Inhibitory Postsynaptic Potential

LF Low Frequency

MSC Magnitude Squared Coherence

PCG Phonocardiogram

PPG Photoplethysmograph

x

Abbreviations and Acronyms

PSD Power Spectrum Density

QRS Electrocardiogram QRS Complex

RSP Respiration

RSPR Respiration Rate

SA Sino-atrial

SKT Skin Temperature

TOE Transoesphageal Echocardiogram

UIM Universal Interfacing Module

USB Universal Serial Bus

VLF Very Low Frequency

xi

List of Publications

LIST OF PUBLICATIONS

Singh. G and Singh. D “Coherence Analysis between ECG Signal and EEG

Signal” BEATS-2010 conducted by NIT, Jalandhar, Punjab, India, Dec 2010, pp.33.

Singh. G, Gupta. V, Singh. D “Estimation of Coherence between ECG Signal and

EEG Signal”, IJECT, Volume 1 Issue 1, ISSN: 2230-7109(online), 2230-9543(print),

pp. 25-28, Dec 2010.

xii

Contents

CONTENTS

Candidate’s Declaration i

Certificate ii

Acknowledgements iv

List of figures and tables v

Abbreviation and acronyms ix

List of Publications x

Contents xi

Abstract xiv

CHAPTER 1: INTRODUCATION

1.1. A Brief Introduction of Human Heart and Electrocardiogram

1.1.1. Physiology of Human Heart

1.1.2. Significance of Heart as an organ

1.1.3. Chambers of Heart

1.1.4. Operation of Heart

1.1.5. Heart Ailments

1.1.6. Diagnostics techniques for Heart diseases

1.2. A Brief Introduction of Human Brain and Electroencephalogram.

1.2.1 Functional Areas Of Cerebrum

1.2.2 Electroencephalogram (EEG)

1.2.3 Rhythmic Brain Activity

1.3. Respiratory Rate

1.3.1 Optimum Breathing

1.4. Literature Review

1.5. Objective of This Thesis Work

CHAPTER 2: PHYSIOLOGICAL DATA ACQUISITION

2.1. MP100 System

2.2.1. Introduction

2.2.2. MP100 block Diagram.

2.2. ECG100C – Electrocardiogram Amplifier Module

2.3. EEG100C – Electroencephalogram Amplifier Module

xiii

Contents

2.4. RSP100C–Respiration pneumogram amplifier module

2.5. Various Functions of AcqKnowledge3.9.0 software

CHAPTER 3: COHERENCE AND PHASE COHERENCE FUNCTION

3.1 Introduction

3.1.1 Cross-Correlation

3.1.2 Auto-Correlation

3.1.3 Correlation Co-efficient Definition

3.2 Spectral Density Functions

3.3 Coherence Function from Spectral Analysis

CHAPTER 4: COHERENCE ANALYSIS BETWEEN ECG AND EEG SIGNAL

4.1. Introduction

4.2. ECG and EEG signals

4.3. Coherence and Phase Coherence between the ECG and EEG signals acquired

at different breathing rates

CHAPTER 5: RESULTS AND DISCUSSION

5.1. Coherence Analysis for first subject

5.1.1 ECG and EEG signals

5.1.2 Coherence

5.1.3 Phase coherence

5.2. Coherence Analysis for second subject

5.2.1 ECG and EEG signals

5.2.2 Coherence

5.2.3 Phase coherence

5.3. Coherence Analysis for third subject

5.3.1 ECG and EEG signals

5.3.2 Coherence

5.3.3 Phase coherence

5.4. Combine Coherence Analysis for all four subjects

5.5. Combine Phase Coherence Analysis for all four subjects

CHAPTER 6: CONCLUSION AND FUTURE SCOPE

REFERENCES

APPENDIX

xiv

Abstract

ECG or electrocardiogram and EEG or electroencephalogram are the very important

parameters when it comes to diagnosis and treatment of human heart and brain related

problems. For this reason signal processing of such signals are of utmost importance. A

continuous non-invasive, low cost and accurate monitoring of functioning of heart and brain

have been proven to be invaluable in various diagnostics applications and clinical

applications.

All the organs of the human body have some synchronism, association and

correlation to each other. In this work we investigate the coherence and phase coherence

between the ECG and EEG; means the association between the human brain and heart.

These signals have proper responses in some specific frequency bands. We acquired 50 ECG

and EEG signals simultaneously using Biopac Inc. Acqknowledge3.9.0 software and MP100

hardware for this work. All data collected from healthy subjects under the age group (21-36

years old) at the sampling rate is 500 samples/second. The number of samples used for the

analysis of association or correlation is 5006 for each signal. We also acquired respiratory

rate by setting the Calculate Channel in the Acqknowledge3.9.0 software simultaneously with

corresponding ECG, EEG and Respiratory signals. The data acquisition is done at different

Respiratory Rates. The different respiratory rates are 0-4 breaths/minute (Low Breathing

Rate), 10-12 breaths/minute (Normal Breathing) and 16-20 breaths/minute (High Breathing

Rate). The EEG signals acquired from the four different positions; the Frontal(F p1−Fp 2

),

Central(C3−C4), Parietal (P3−P4) and Occipital (O1−O2) Brain Regions.

A measure of ‘coupling’ and as a measure of a functional association (relationship)

between two signals (here, ECG Signal and EEG Signal); is interpreted as Coherence.

Mathematically the Coherence is the degree of relationship or association of frequency

spectra between the two signals (ECG and EEG) at a particular frequency. In this work the

Magnitude Squared Coherence (MSC) is investigated in the frequency band 0Hz to 35Hz.

Preliminary results obtained from the examination of three subjects show the existence of a

maximum MSC at the normal respiration rate respiration frequency and mean of MSCs of no

airflow, normal airflow and high airflow ECG and EEG signals is found to be continuously

decreasing in the frequency band(0-35Hz).

Abstract

Secondary results are found to be that the maximum mean of magnitude squared

coherence is among the three subjects’ coherence between the ECG and EEG signal from

parietal(P3−P4). Furthermore, the frequency content in both heart signals and brain signals

was calculated via power spectrum analysis with frequency band in the two organs was

investigated.

xv

Abstract

Chapter 1

INTRODUCTION

1.1. A Brief Introduction of Human Heart &

Electrocardiogram(ECG):

1.1.1. Physiology of Heart:The heart is a muscular organ in all vertebrates responsible for pumping blood

through the blood vessels by repeated, rhythmic contractions. The term cardiac (as in

cardiology) means "related to the heart" and comes from the Greek καρδιά, kardia, for

"heart".

The heart of a vertebrate is composed of cardiac muscle, an involuntary muscle

tissue which is found only within this organ. The average human heart, beating at 72 beats

per minute, will beat approximately 2.5 billion times during a lifetime (about 66 years). It

weighs on average 250 gm to 300 gm in females and 300 gm to 350 gm in males [1].

The heart is usually situated in the middle of the thorax with the largest part of

the heart slightly offset to the left (although sometimes it is on the right, see figure 1.1),

underneath the breastbone.

Figure1.1: The Human Heart with Coronary Arteries

xv

Abstract

The heart is usually felt to be on the left side because the left heart (left ventricle) is

stronger (it pumps to all body parts).

1.2.1. Significance of Heart as an organ:The function of the right side of the heart is to collect de-oxygenated blood, in the

right atrium, from the body (via superior and inferior vena cava) and pump it, via the right

ventricle, into the lungs (pulmonary circulation) so that carbon dioxide can be dropped off

and oxygen picked up (gas exchange). This happens through the passive process of diffusion.

The left side (see left heart) collects oxygenated blood from the lungs into the left atrium.

From the left atrium the blood moves to the left ventricle which pumps it out to the body (via

the aorta). On both sides, the lower ventricles are thicker and stronger than the upper atria.

The muscle wall surrounding the left ventricle is thicker than the wall surrounding the right

ventricle due to the higher force needed to pump the blood through the systemic circulation.

Starting in the right atrium, the blood flows through the tricuspid valve to the right ventricle.

Here it is pumped out the pulmonary semi-lunar valve and travels through the pulmonary

artery to the lungs. From there, blood flows back through the pulmonary vein to the left

atrium. It then travels through the mitral valve to the left ventricle, from where it is pumped

through the aortic semi-lunar valve to the aorta. The aorta forks and the blood is divided

between major arteries which supply the upper and lower body. The blood travels in the

arteries to the smaller arterioles, then finally to the tiny capillaries which feed each cell. The

(relatively) deoxygenated blood then travels to the venules, which coalesce into veins, then to

the inferior and superior venae cava and finally back to the right atrium where the process

began.

The impulses generated during the heart cycle produce electrical currents, which are

conducted through body fluids to the skin, where they can be detected by electrodes and

recorded as an electrocardiogram (ECG or EKG) [3].

1.3.1. Chambers of Heart:The outer layer of the pericardium surrounds the roots of your heart's major blood

vessels and is attached by ligaments to your spinal column, diaphragm, and other parts of

your body. The inner layer of the pericardium is attached to the heart muscle. A coating of

fluid separates the two layers of membrane, letting the heart move as it beats, yet still be

attached to your body.

xv

Abstract

Figure1.2: Heart Valves

The heart contains four chambers; two thin-walled atria separated from each other by

an inter-atrial septum and two thicker-walled ventricles possessing common wall in the inter-

ventricular septum.

Atria and ventricles are connected by a fibrous A-V ring. This ring is penetrated on the

right side by the tricuspid valve and on the left side by the mitral valve as shown in Fig 1.2.

The two valves consist of flaps or cusps, which are attached at the periphery of the valve ring.

The heart wall, which is composed of a cardiac muscle tissue, is referred as the

myocardium. The muscle cells of myocardium are classified into five functionally and

anatomically separate parts namely, sino-atrial (SA) node, atrio-ventricular (AV) node, His-

purkinje system, atrial muscle and ventricular muscle each having different characteristic

action potentials. They are mainly involved in the maintenance of two primary and well

synchronized physiological events, namely, the heart’s mechanical activity (pumping of the

blood) and the heart’s electrical activity (the transmission of electrochemical impulses for the

coordination of the heart’s effort). These two activities give rise to an orderly heart beat.

Four types of valves regulate blood flow through your heart shown in figure 1.2:

The Tricuspid valve regulates blood flow between the right atrium and right

ventricle.

The Pulmonary valve controls blood flow from the right ventricle into the

pulmonary arteries, which carry blood to your lungs to pick up oxygen.

The Mitral valve lets oxygen-rich blood from your lungs pass from the left

atrium into the left ventricle.

xv

Abstract

The Aortic valve opens the way for oxygen-rich blood to pass from the left

ventricle into the aorta, your body's largest artery, where it is delivered to the

rest of your body.

1.4.1. Operation of Heart:1.1.4.1. The Conduction System :

Electrical impulses from the heart muscle (the myocardium) cause the heart to

contract. This electrical signal begins in the sino-atrial (SA) node, located at the top of the

right atrium. The SA node is sometimes called the heart's "natural pacemaker." An electrical

impulse from this natural pacemaker travels through the muscle fibres of the atria and

ventricles, causing them to contract. Although the SA node sends electrical impulses at a

certain rate, the heart rate may still change depending on physical demands, stress, or

hormonal factors.

Some cardiac cells are self-excitable, contracting without any signal from the nervous

system, even if removed from the heart and placed in culture. Each of these cells has its own

intrinsic contraction rhythm. A region of the human heart called the sino-atrial node (SA

node), or pacemaker, sets the rate and timing at which all cardiac muscle cells contract. The

SA node generates electrical impulses, much like those produced by nerve cells. Impulses

from the SA node spread rapidly through the walls of the atria, causing both atria to contract

in unison. The impulses also pass to another region of specialized cardiac muscle tissue, a

relay point called the atrio-ventricular (AV) node, located in the wall between the right

atrium and the right ventricle. Here, the impulses are delayed for about 0.1s before spreading

to the walls of the ventricle. The delay ensures that the atria empty completely before the

ventricles contract.

xv

Abstract

Figure 1.3: Cardiac Conduction System

1.1.4.2. The Circulatory System:The heart and circulatory system make up the cardiovascular system. The heart works

as a pump that pushes blood to the organs, tissues, and cells of your body. Blood delivers

oxygen and nutrients to every cell and removes the carbon dioxide and waste products made

by those cells. Blood is carried from the heart to the rest of the body through a complex

network of arteries, arterioles, and capillaries. Blood is returned to your heart through venules

and veins. If all the vessels of this network in the body were laid end-to-end, they would

extend for about 60,000 miles (more than 96,500 kilo-meters), which is far enough to circle

the earth more than twice..!!

1.5.1. Heart Ailments:Heart ailments are increasing in the current years. Much hospital admission took place

because of the growing heart diseases. The common diseases among the heart ailments are

coronary arteries blockage that supplies blood to the person’s heart, it is popularly known as

CAD (Coronary Artery Disease). Heart can also gets affected by the defects in congenital

heart which are present since birth in one percent of infants approximately. There are many

advances that are taking position in the direction to cure any kind of heart disease. Some

years back the treatment for heart diseases were available but they were either surgery or with

medicines. But as the time passes with the development of new technologies and sciences

made any type of heart treatment very easy. The latest improvement is made in the area of

xv

Abstract

worldwide cardiology; several curative treatments are done by making the small cut of 1.0

mm in the groin or in the wrist vein.

These actions are taken as well as adopted to provide maximum relieve to the patients

along with minimal risk. Surgical bypass was the only treatment available twenty-six years

back for the artery blockage. But nowadays-countless situations can easily be managed and

handled by a simple method known as Angioplasty. By this treatment a patient can walk after

twelve hours and can go back to their works after forty-eight hours, because heart ailments

treatment is very simple. But in some cases say fifteen to twenty percent cases it carries the

danger of re-blockage, specifically in those cases where patient is suffering from sugar

problem or Diabetes mellitus.

Heart does a number of main and useful works and its work is never ending till death.

There are many common diseases of heart and the treatment is also available but as the time

passes the treatment is becoming very expensive as well as risk is also there. The home

remedies can cure heart ailments also. The best technique to cure the heart ailments is to gear

up in the direction of the cleansing diet, since, it purifies the blood and the cleaner your blood

will be the less chances you will face of any kind of toxicity piling up in the area of your

heart.

1.1.5.a. Heart Attack:

Heart attack is caused when the blood flow to the heart muscle is obstructed and if the

blood supply is not re- established immediately the heart muscle begins to perish due to lack

of oxygen.

Heart attack is one of the most important mortality factors among Americans. Heart

attack can be best treated within the first hour of symptoms. Seek emergency medical aid as

timeliness is crucial in reducing the mortality rates. Heart attacks are normally caused due to

the thickening and narrowing of the coronary arteries due to the deposition of fatty materials

such as cholesterol. It blocks the free flow of oxygenated blood to the heart and results in a

condition called the coronary artery disease.

The first step in the event of a heart attack would be remove the blockage to ensure

blood supply as heart muscles will start dying to be replaced by scar tissue, if oxygen supply

remains cut off for more than 20 minutes. This in turn would have far reaching consequences

in the future. Serious impediments of heart disease can result in life threatening conditions

xv

Abstract

like heart failure or arrhythmias. Heart failure happens when the heart cannot pump adequate

blood throughout the body and the irregularity in the heart beat patterns might result in the

death of the patient if timely medical aid is not extended. Watch out for some of the common

symptoms such as chest discomfort, squeezing pain and cyclic pain along the neck, shoulders

and jaw region which lasts for a few minutes the disappear to come back after some time.

Breathlessness, stomach discomfort, fainting etc are also commonly seen. Do not take these

warning symptoms lightly and even if these vanish completely, it is recommended to consult

a doctor at the earliest. Call an ambulance to save time and keeping an aspirin tablet under the

tongue might help to minimize the damages caused by the blood clots in the artery.

A heart attack is alarming alright however of you can identify the heart attack

symptoms you can save a precious life of your near and dear ones. Majority of people have

the false notion that heart attack is always massive and serious where the person clasps his

chest and slumps to his chair just like in a movie. However in reality, heart attacks progress

slowly as a minor discomfort, which many tend to ignore. Recognizing these symptoms can

make all the difference between life and death. In case you notice any of these signs do

consult a doctor as fast action is what matters.

Remember timely help is the key in the management of heart attacks as there are

many effective anti coagulants such as aspirin and nitro-glycerine, which can stop a heart

attack in the initial stages itself. The sooner they are administered the greater would be the

chances of a full recovery. You can reduce the risk of heart attack by certain life style

changes. If you are a smoker try to quit this habit; reduce high blood pressure and cholesterol

and in case you have had a heart attack before follow the medications promptly as per the

doctor’s advice.

1.1.5.b. High blood pressure/ Hypertension:

The pressure exerted by the blood on the arteries as it is pumped through the blood

vessels in called blood pressure. If the blood pressure reading shows 140/90 mmHg or higher,

you have high blood pressure or hypertension. The normal blood pressure of a healthy adult

should be lower than 120/80 mmHg. If blood pressure is high, the heart pumps the blood

harder, which in turn leaves the arteries hardened resulting in atherosclerosis.

The blood pressure is measured by using a sphygmomanometer and one end of the

stethoscope is placed over an artery to record the readings. The pressure at which the first

xv

Abstract

pulse sound can be heard is called systolic blood pressure and the pressure level where the

sound disappears is the diastolic pressure and is expressed in millimetre of mercury (mm Hg).

High blood pressure and excess body weight are closely related. So, weight management is

the most important phase in the treatment of your high blood pressure. Requisite dietary

changes coupled with a well planned work out might help you to keep hypertension under

check.

In some people high blood pressure could be triggered by another disease then it is

called secondary hypertension, which gets back to normal once the root cause is solved.

Some of the causes for secondary hypertension include tumours, kidney ailments, narrowing

of the aorta, pregnancy etc.

While in the majority of cases no particular reason could be attributed to high blood

pressure and it is often referred to as primary hypertension. Hypertension is normally

associated with old age as the arteries get hardened with age and in some cases it shows a

hereditary history as well. In people suffering from salt sensitivity, the blood pressure shoots

up with the usage of salt. Such patients should desist from eating fast foods and deep fried

salty items and if you take care to avoid high sodium food sources, the blood pressure could

be managed easily. Alcohol consumption is yet another cause of heart disease and also for

high blood pressure; make sure that you do not consume more than two drinks a day to steer

clear of this risk. People with high blood pressure should take immediate medical aid to

control it in time.

1.6.1. Diagnostics Techniques for Heart Diseases:There are several tests available to diagnose possible heart disease. How the physician

decides which tests to perform (and how many) depends on factors such as your risk factors,

history of heart problems, current symptoms and the physician's interpretation of these

factors.

The tests usually begin with the simplest and may progress to more complicated ones.

Specific tests depend on your particular problem(s) and the physician's assessment. Tests that

do not involve inserting needles, instruments or fluids into the body are termed non-invasive.

Those that do are called invasive tests.

1.1.6.1. Non-Invasive Tests:

xv

Abstract

1.1.6.1.a. Electrocardiogram (ECG):

This is the most common test for heart conditions. It is a simple, painless test that

takes about 10 minutes. Every time the heart beats, natural electrical currents can be picked

up by electrodes placed on various points around the body. These natural electrical currents

are recorded on paper. The tracing records the heart rate and rhythm and whether the muscle

is conducting the electricity normally. Damaged heart muscle, or muscle that is short of

oxygen, will result in a different appearance on the tracing. The resultant tracing can give the

doctor a lot of information about your heart, but, like most tests, the ECG is not infallible. If

one have angina, the heart tracing may be normal if it is recorded at rest when one is free of

pain. In this case one may need an exercise ECG.

1.1.6.1.b. Phonocardiogram(PCG):

The physiological variability of the mechanical function of the heart is reflected in the

produced acoustic vibrations—the heart sounds. Heart sounds have been widely used in

clinical practices and the phonocardiography is the graphical continuous non-invasive,

recording of heart sounds.

Although the Electrocardiogram (ECG) signal provides reliable indications for

electrical dysfunctions related to the heart’s pacing and conduction systems, as well as for

conditions of myocardial ischemia. However, mechanical dysfunctions that are not

accompanied by electrical changes may not be reflected in the electrocardiogram. In addition,

patients with chronic heart disease such as heart failure often have enduring ECG

abnormalities [4], which reduce the efficacy of ECG monitoring in detecting worsening of the

disease. Hence in some cases of heart diseases, there is need of Phonocardiogram test.

Especially, The heart valvular diseases i.e. Mitral valve Stenosis, Mitral regurgitations,

Arterial-Septal defect etc are reflected from Phonocardiogram test.

1.1.6.1.c. Holter Monitoring:

The purpose of Holter monitoring is to look for heart rhythm problems over a 24 to

48-hours period. The Holter monitor is a small, portable, battery powered ECG machine worn

when at home. It will record your heart rate and rhythm over a period of time. You will be

asked to keep a diary of activities and any symptoms that you experience while the Holter

monitor is being worn. At the end of the time period, the monitor needs to be returned to the

hospital and a technician will view the recorded information.

1.1.6.1.d. Echocardiogram:

xv

Abstract

This test uses sound waves to study the movement of the heart's chambers and valves.

This is particularly useful as you can assess different areas of the heart while it is beating.

The echo sound waves create an image on the monitor as an ultrasound transducer probe is

passed over the chest and heart.

1.1.6.1.e. Exercise Stress Test (treadmill test or exercise ECG):

Some heart problems only show up when the heart is working hard. To assess this, it

is necessary to monitor the heart when you are exercising. A continuous ECG is done to

achieve this. The test takes a maximum of 10 minutes. You will be connected to an ECG

machine and blood pressure monitoring facilities. You then walk on a treadmill which will

slowly increase in speed and incline. At various stages you will have blood pressure and ECG

recorded.

1.1.6.2. Invasive Tests:

1.1.6.2.a. Echocardiogram Stress Test:

This is similar to a resting echo test. It is performed on people who need to have a

exercise ECG but are unable to walk any great distance due to mobility problems. Medication

is given via an IV cannula to simulate exercise. Heart function and rhythm are monitored.

1.1.6.2.b. Transoesphageal Echocardiogram (TOE):

A TOE gains more information about how your heart is functioning from a closer

position, without the chest wall blocking ultrasound recordings. Because your heart sits next

to the oesophagus you get clearer pictures. This test uses the same special sound wave

technology that a regular echocardiogram uses, but the pictures are taken by inserting a

special probe into the oesophagus (the food tube that connects the mouth with the stomach)

rather than placing it on the patient's chest wall. A local anaesthetic spray is administered to

the back of the mouth to numb that area and prevent gagging. You will be given sedation to

make you relaxed and drowsy but not completely asleep. Your nurse will advise you when

you can eat and drink again, this are about one to two hours.

1.1.6.2.c. Electrophysiological Studies (EPS):

Your cardiologist might refer you for electrophysiology studies if you have an

abnormal heart rhythm or palpitations. Fine tubes called electrode catheters are introduced

through a vein and/or artery, usually in the groin. They are then gently moved into position in

the heart, where they stimulate the heart and record electrical impulses. This type of

xv

Abstract

investigation assists the doctor to make a definitive diagnosis and plan treatment for

arrhythmia management.

1.1.6.2.d. Blood Tests:

Various blood tests may be performed depending on your type of heart disease. These

all help to build a picture of the nature of your disease.

Included might be assays for:

Electrolytes

Full blood count

Hormone levels

Blood clotting times

Cardiac enzymes.

In recent years the most commonly used blood test to measure the level of cardiac

muscle damages are proteins called troponins. The level of troponins in the blood helps to

give a quick and accurate idea of the amount of muscle damage after a heart attack.

1.1.6.2.e. Cardiac Troponins:

Cardiac troponins measurements help either confirm or exclude a heart attack in a

person who may be having, or recently had, a cardiac event. They also help decide what

treatments a person with unstable angina may need. Troponins T and Troponins I are proteins

that are part of the heart or cardiac muscle. When heart muscle injury occurs, these proteins

are released. Troponins T and I are more sensitive to heart muscle damage than the enzyme

creatine kinase (CK). This makes them a valuable test to detect mild heart attacks. They can

be detected in blood as early as three hours after a heart attack associated chest pain starts.

The levels peak at 10 to 24 hours and can still be detected up to five to 10 days later. This

means that if you have had chest pain for several days a heart attack can still be detected.

1.2. A Brief Introduction of Human Brain & Electroencephalogram(EEG):

The brain is the center of the nervous system in all vertebrate and most invertebrate

animals. The brain controls the other organ systems of the body, either by activating muscles

or by causing secretion of chemicals such as hormones and neurotransmitters.

The brain constitutes about one-fiftieth of the body weight and lies within the cranial

cavity. The parts of the brain are cerebrum, midbrain, pons, medulla oblongata and

xv

Abstract

cerebellum. For descriptive purposes each hemisphere of the cerebrum is divided into lobes

which take the names of the bones of the cranium under which they lie:

• Frontal

• Parietal

• Temporal

• Occipital

The boundaries of the lobes are marked by deep sulci (fissures). These are the central,

lateral and parieto-occipital sulci.

Figure1.4: The lobes and sulci of the cerebrum.

1.2.1 Functional Areas of Cerebrum

The main areas of the cerebrum associated with sensory perception and voluntary

motor activity are known but it is unlikely that any area is associated exclusively with only

one function. Except where specially mentioned, the different areas are active in both

hemispheres.

There are three main varieties of activity associated with the cerebral cortex:

• Mental activities involved in memory, intelligence, sense of responsibility, thinking,

reasoning, moral sense and learning are attributed to the higher centres

• Sensory perception, including the perception of pain, temperature, touch, sight,

hearing, taste and smell

xv

Abstract

• Initiation and control of skeletal (voluntary) muscle contraction.

Figure 1.5: Functional areas of the cerebrum (Courtesy: Ross and Wilson - Anatomy and Physiology in Health and Illness)

1.2.2 Electroencephalogram (EEG)

Richard Caton (1842–1926), a scientist from Liverpool, England, used a

galvanometer and placed two electrodes over the scalp of a human subject and thereby first

recorded brain activity in the form of electrical signals in 1875. Since then, the concepts of

electro-(referring to registration of brain electrical activities) encephalo- (referring to emitting

the signals from the head), and gram (or graphy), which means drawing or writing, were

combined so that the term EEG was henceforth used to denote electrical neural activity of the

brain.

The Central Nervous System generally consists of nerve cells and glia cells, which are

located between neurons. Each nerve cell consists of axons, dendrites, and cell bodies. Nerve

cells respond to stimuli and transmit information over long distances. An axon is a long

cylinder, which transmits an electrical impulse and can be several metres long in vertebrates.

Dendrites are connected to either the axons or dendrites of other cells and receive impulses

from other nerves or relay the signals to other nerves. In the human brain each nerve is

connected to approximately 10,000 other nerves, mostly through dendritic connections.

The activities in the CNS are mainly related to the synaptic currents transferred

between the junctions (called synapses) of axons and dendrites, or dendrites and dendrites of

cells. A potential of 60–70 mV with negative polarity may be recorded under the membrane

of the cell body. This potential changes with variations in synaptic activities. If an action

xv

Abstract

potential travels along the fibre, which ends in an excitatory synapse, an excitatory

postsynaptic potential (EPSP) occurs in the following neuron. If two action potentials travel

along the same fibre over a short distance, there will be a summation of EPSPs producing an

action potential on the postsynaptic neuron providing a certain threshold of membrane

potential is reached. If the fibre ends in an inhibitory synapse, then hyperpolarization will

occur, indicating an inhibitory postsynaptic potential (IPSP)

Following the generation of an IPSP, there is an overflow of cations from the nerve

cell or an inflow of anions into the nerve cell. This flow ultimately causes a change in

potential long the nerve cell membrane. Primary transmembranous currents generate

secondary inonal currents along the cell membranes in the intra- and extracellular space. The

portion of these currents that flow through the extracellular space is directly responsible for

the generation of field potentials. These field potentials, usually with less than 100 Hz

frequency, are called EEGs when there are no changes in the signal average and DC if there

are slow drifts in the average signals, which may mask the actual EEG signals. A

combination of EEG and DC potentials is often observed for some abnormalities in the brain

such as seizure.

An EEG signal is a measurement of currents that flow during synaptic excitations of

the dendrites of many pyramidal neurons in the cerebral cortex. When brain cells (neurons)

are activated, the synaptic currents are produced within the dendrites. This current generates a

magnetic field measurable by electromyogram (EMG) machines and a secondary electrical

field over the scalp measurable by EEG systems.

1.2.3 Rhythmic Brain Activity

EEG signal voltage amplitude ranges from about 1 to 100 micro volts peak to peak at

low frequencies (0.5 to 100 Hz) at the cranial surface. At the surface of cerebrum, signal may

be 10 times stronger.

In healthy adults, the amplitudes and frequencies of EEG signals change from one

state of a human to another, such as wakefulness and sleep. The characteristics of the waves

also change with age. There are five major brain waves distinguished by their different

frequency ranges and amplitudes. These frequency bands from low to high frequencies

respectively are called alpha (α), theta (θ), beta (β), delta (δ), and gamma (γ).

1.2.3.a Delta wave:

xv

Abstract

Delta waves lie within the range of 0.5–4 Hz. These waves are primarily associated

with deep sleep and may be present in the waking state. Theta waves lie within the range of

4–7.5 Hz. Theta waves appear as consciousness slips towards drowsiness. Theta waves have

been associated with access to unconscious material, creative inspiration and deep

meditation. A theta wave is often accompanied by other frequencies and seems to be related

to the level of arousal. The theta wave plays an important role in infancy and childhood.

Figure 1.6: Rhythmic brain activity (Courtesy: EEG Signal Processing; Saeid Sanei and J.A. Chambers)

1.2.3.b Theta waves:

For theta waves the frequency lies between 4 Hz to 8Hz. Theta are normally seen

young children. It may be seen in drowsiness or arousal in older children and adults. It can

also be seen in meditation. Excess theta for age represents abnormal activity. It can be seen as

a focal disturbance in focal sub cortical lesions; it can be seen in generalised distribution in

diffuse disorder or metabolic encephalopathy or deep midline disorders or some instances of

hydrocephalus.

xv

Abstract

1.2.3.c Alpha Wave:

Alpha waves appear in the posterior half of the head and are usually found over the

occipital region of the brain. They can be detected in all parts of posterior lobes of the brain.

For alpha waves the frequency lies within the range of 8–13 Hz, and commonly appears as a

round or sinusoidal shaped signal. However, in rare cases it may manifest itself as sharp

waves. In such cases, the negative component appears to be sharp and the positive component

appears to be rounded, similar to the wave morphology of the rolandic mu (μ) rhythm. Alpha

waves have been thought to indicate both a relaxed awareness without any attention or

concentration. The alpha wave is the most prominent rhythm in the whole realm of brain

activity. An alpha wave has higher amplitude over the occipital areas and has amplitude of

normally less than 50 μV.

1.2.3.d Sensorimotor rhythm / mu rhythm:

Mu rhythm is alpha-range activity that is seen over the sensorimotor cortex. It

characteristically attenuates with movement of contralateral arm (or mental imagery of

movement of the contralateral arm)

1.2.3.e Beta Wave:

A beta wave is the electrical activity of the brain varying within the range of 14–26

Hz (though in some literature no upper bound is given).A beta wave is the usual waking

rhythm of the brain associated with active thinking, active attention, focus on the outside

world, or solving concrete problems, and is found in normal adults. A high-level beta wave

may be acquired when a human is in a panic state. Rhythmical beta activity is encountered

chiefly over the frontal and central regions. Importantly, a central beta rhythm is related to

the rolandic mu rhythm and can be blocked by motor activity or tactile stimulation. The

amplitude of beta rhythm is normally under 30 μV. The frequencies above 30 Hz (mainly up

to 45 Hz) correspond to the gamma range (sometimes called the fast beta wave). Although

the amplitudes of these rhythms are very low and their occurrence is rare, detection of these

rhythms can be used for confirmation of certain brain diseases. The regions of high EEG

frequencies and highest levels of cerebral blood flow (as well as oxygen and glucose uptake)

are located in the frontocentral area. The gamma wave band has also been proved to be a

good indication of event-related synchronization (ERS) of the brain and can be used to

demonstrate the locus for right and left index finger movement, right toes, and the rather

broad and bilateral area for tongue movement.

xv

Abstract

Table 1.1: Rhythmic brain activity

Delta δ Theta θ Alpha α Beta β Gamma γ

Frequency (Hz) 0.5- 4 4 - 8 8 - 13 13 - 22 22 - 50

Amplitude (μV) < 100 < 100 < 20 < 20 < 2

Predominant Location ------ ------ Occipital

Area Frontal Area ------

Found inInfants,

Deep sleep etc...

Children, sleeping

adults etc...

Light sleep, Eyes closed

High state of wakefulness

Sensory stimulation

1.3. Respiratory Rate:Respiration rate, pulmonary ventilation rate or ventilation rate) is the number of

breaths a living being, such as a human, takes within a certain amount of time (frequently

given in breaths per minute).

The human respiration rate is usually measured when a person is at rest and simply

involves counting the number of breathes for one minute by counting how many times the

chest rises. Respiration rates may increase with fever, illness, or other medical conditions.

When checking respiration, it is important to also note whether a person has any difficulty

Respiratory rates measurement in children less than five years, for a 30 second or 60

second period, suggesting the 60 seconds resulted in the least variability. Another study

found that rapid respiratory rates in babies, counted using a stethoscope, were 20–50% higher

than those counted from beside the cot without the aid of the stethoscope.

1.3.1 Optimum Breathing:

A trained, systematic approach to deep breathing may lower respiration rates in

cardiac patients, helping them to maintain healthy blood oxygen levels and become more

physically fit. In one study, 15 cardiac patients were assigned to one of two experimental

groups. One of the groups learned "complete yoga breathing," a style of respiration that

encourages slow, deep breathing at a rate of about six breaths per minute. Those patients

continued practicing the breathing method at home for an hour a day. After a month, the

xv

Abstract

patients practicing the breathing technique breathed more slowly, had higher levels of blood

oxygen, and performed better on exercise tests.

Table 1.2: Average respiratory rates, by age:

Age Average Respiratory Rate (BPM)

1. Newborns Average 44 breaths per minute

2. Infants 20–40 breaths per minute

3. Preschool children 20–30 breaths per minute

4. Older children 16–25 breaths per minute

5. Adults 12–20 breaths per minute

6. Adults during strenuous

exercise

35–45 breaths per minute

7. Athletes 60–70 breaths per minute

Respiratory minute volume is the volume of air which can be inhaled (inhaled minute

volume) or exhaled (exhaled minute volume) from a person's lungs in one minute. The value

of respiratory rate as an indicator of potential respiratory dysfunction has been investigated

but findings suggest it is of limited value.

One study found that only 33% of people presenting to an emergency department with

oxygen saturation below 90% had an increased respiratory rate.

An evaluation of respiratory rate for the differentiation of the severity of illness in

babies less than 6 months found it not to be very useful. Approximately half of the babies had

a respiratory rate above 50 breaths per minute, thereby questioning the value of having a

"cut-off" at 50 breaths per minute as the indicator of serious respiratory illness. It has also

been reported that factors such as crying, sleeping, agitation and age have a significant

influence on the respiratory rate.

1.4. Literature review:1.4.1 History of Electrocardiogram (ECG):

Alexander Muirhead is reported to have attached wires to a feverish patient's wrist to

obtain a record of the patient's heartbeat while studying for his Doctor of Science (in

electricity) in 1872 at St Bartholomew's Hospital. An initial breakthrough came when Willem

xv

Abstract

Einthoven, working in Leiden, Netherlands, used the string galvanometer that he invented in

1903. Using today's self-adhesive electrodes Einthoven's subjects would immerse each of

their limbs into containers of salt solutions from which the ECG was recorded. Einthoven

assigned the letters P, Q, R, S and T to the various deflections, and described the

electrocardiographic features of a number of cardiovascular disorders. In 1924, he was

awarded the Nobel Prize in Medicine for his discovery. Though the basic principles of that

era are still in use today, there have been many advances in electrocardiography over the

years. The instrumentation, for example, has evolved from a cumbersome laboratory

apparatus to compact electronic systems that often include computerized interpretation of the

electrocardiogram.

Figure 1.7: Willem Einthoven, working in Leiden, Netherlands, used the string galvanometer that he invented in 1903.

By definition a 12-lead ECG will show a short segment of the recording of each of the

12-leads. This is often arranged in a grid of 4 columns by three rows, the first columns being

the limb leads as lead-I, lead-II and lead-III, the second column the augmented limb leads as

lead-aVR, lead-aVL and lead- aVF and the last two columns being the chest leads as leads V 1

, V 2,V 3, V 4, V 5 and V 6. It is usually possible to change this layout so it is vital to check the

labels to see which lead is represented. Each column will usually record the same moment in

xv

Abstract

time for the three leads and then the recording will switch to the next column which will

record the heart beats after that point. It is possible for the heart rhythm to change between

the columns of leads.

Figure 1.8: Experimental Setup of 12 Lead ECG Acquisitions from Atria 6100

1.4.2 History of Electrocardiogram (EEG):

A timeline of the history of EEG is given by Swartz. Richard Caton (1842–1926), a

physician practicing in Liverpool, presented his findings about electrical phenomena of the

exposed cerebral hemispheres of rabbits and monkeys in the British Medical Journal in 1875.

In 1890, Polish physiologist Adolf Beck published an investigation of spontaneous

electrical activity of the brain of rabbits and dogs that included rhythmic oscillations altered

by light.

In 1912, Russian physiologist, Vladimir Vladimirovich Pravdich-Neminsky published

the first animal EEG and the evoked potential of the mammalian (dog). In 1914, Napoleon

Cybulskiand Jelenska-Macieszyna photographed EEG-recordings of experimentally induced

seizures.

German physiologist and psychiatrist Hans Berger (1873–1941) recorded the first human

EEG in 1924.

The acquisition of EEG is described in the next chapter 2 (Physiological data

acquisition).

1.4.3 Work done so far in Coherence analysis of Physiological Signals:1.4.3.a A Study of Heart Rate and Brain System Complexity and Their Interaction in Sleep-

Deprived Subjects by AK Kokonozi, EM Michail, IC Chouvarda, and NM Maglaveras

xv

Abstract

In 2008 , we investigate the synchronization of the dynamic behaviour of heart rate

(ECG) and brain (EEG) signals using sample entropy as a measure of complexity. EEG and

ECG recordings were collected during experiment with sleep-deprived subjects exposed to

real field driving conditions. The degree to which brain and heart complexity loose

complexity in a synchronous manner, indicating a possible interaction between the two

systems is investigated. Furthermore, the frequency content in both heart rate and brain

signals was calculated via power spectrum analysis and the association of synchronisation

patterns with prevalent frequencies in the two systems was investigated.

This study shows interesting preliminary results for the synchronization of heart and

brain signals during real field driving conditions. A slight interaction between these systems

can be observed, which varies during the driving task, depending on the drivers fatigue and

special events.

In order to assess and determine the correlation of EEG and ECG signals, we need to

further investigate their dynamic behaviour under different driving conditions (more Subjects

of different age and health status in different drowsiness level and driving conditions) [26].

1.4.3.b Interaction between Sleep EEG and ECG Signals during and after Obstructive Sleep

Apnea Events with or without Arousals by AH Khandoker, CK Karmakar, and M

Palaniswami

In 2008, this is a preliminary attempt to directly investigate the interactions of sleep

EEG and ECG signals during normal, OSA breathing event and events following its

termination with or without arousal in non-REM (NREM) and REM sleep stages. ECG and

EEG signals were collected from 10 patients with OSA and 5 healthy subjects. Coherence

between two signals (Coherence between ECG and EEG) over different frequency

bands(range:0~40Hz) were calculated for normal breathing events, OSA events and events

following OSA terminations (with/without arousals) in NREM as well as REM sleep. In

normal breathing events, overall Coherence in REM sleep is higher than that in NREM sleep.

Significant (p<0.01) differences of Coherence between OSA events with and without arousals

were found in NREM sleep over 0.5-25 Hz bands but in REM sleep over 3.0-12 Hz. This

research could be useful in understanding cardiac dysfunction in sleep apnoea patient.

The limitation of this study is that proper separation of the effect of cardiac electric

field from heart cycle related brain potentials was not considered. Although the cardiac field

xv

Abstract

artifact (CFA) in the EEG is not relevant in most clinical and experimental situations because

of its small magnitude, but in connection with heart cycle-related EEG averaging it turns out

to be a problem. In this study we have not used any computational method to remove the

CFA from the averaged EEG data [3].

1.4.3.c Coherence Analysis between Respiration and PPG Signal by Bivariate AR Model by

Yue-Der Lin, Wei-Ting Liu, Ching-Che Tsai, and Wen-Hsiu Chen

In 2009, PPG is a potential tool in clinical applications. Among such, the relationship

between respiration and PPG signal has attracted attention in past decades. In this research, a

bivariate AR spectral estimation method was utilized for the coherence analysis between

these two signals. Ten healthy subjects participated in this research with signals measured at

different respiratory rates. The results demonstrate that high coherence exists between

respiration and PPG signal, whereas the coherence disappears in breath-holding experiments.

These results imply that PPG signal reveals the respiratory information. The utilized method

may provide an attractive alternative approach for the related researches.

The existence of coherent peak can be determined by checking whether the

corresponding pole inside the unit circle is prominent or not. It has been shown that the

coherence spectrum is sensitive and specific to the respiration in this research. It may be

possible to acquire the respiratory information from PPG signal by single-channel AR

method with the consideration of poles around the respiratory frequency [1].

1.4.3.d Single-Trial EEG-EMG Coherence Analysis Reveals Muscle Fatigue-Related

Progressive Alterations in Corticomuscular Coupling by Qi Yang, Vlodek Siemionow,

Wanxiang Yao, Vinod Sahgal, and Guang H. Yue

In 2010, voluntary muscle fatigue is a progressive process. A recent study

demonstrated muscle fatigue-induced weakening of functional corticomuscular coupling

measured by coherence between the brain [electroencephalogram (EEG)] and muscle

[electromyogram (EMG)] signals after a relatively long-duration muscle contraction.

Comparing the EEG-EMG coherence before versus after fatigue or between data of two long-

duration time blocks is not adequate to reveal the dynamic nature of the fatigue process. The

purpose of this study was to address this issue by quantifying single-trial EEG-EMG

coherence and EEG, EMG power based on wavelet transform. The energy of both the EEG

and EMG signals decreased significantly with muscle fatigue. This provides extra

xv

Abstract

information to demonstrate a time course of dynamic adaptations of the functional

corticomuscular coupling, as well as brain and muscle signals during muscle fatigue [9].

1.4.3.e On the Recording Reference Contribution to EEG Correlation, Phase Synchorony,

and Coherence by Sanqing Hu, Matt Stead, Qionghai Dai, and Gregory A. Worrell

The degree of synchronization in electroencephalography (EEG) signals is commonly

characterized by the time-series measures, namely, correlation, phase synchrony, and

magnitude squared coherence (MSC). However, it is now well established that the

interpretation of the results from these measures are confounded by the recording reference

signal and that this problem is not mitigated by the use of other EEG montages, such as

bipolar and average reference. In this paper, we analyze the impact of reference signal

amplitude and power on EEG signal correlation, phase synchrony, and MSC. We show that,

first, when two nonreferential signals have negative correlation, the phase synchrony and the

absolute value of the correlation of the two referential signals may have two regions of

behaviour characterized by a monotonic decrease to zero and then a monotonic increase to

one as the amplitude of the reference signal varies in ¿. It is notable that even a small change

of the amplitude may lead to significant impact on these two measures. Second, when two

nonreferential signals have positive correlation, the correlation and phase-synchrony values

of the two referential signals can monotonically increase to one (or monotonically decrease to

some positive value and then monotonically increase to one) as the amplitude of the reference

signal varies in ¿. Third, when two nonreferential signals have negative cross-power, the

MSC of the two referential signals can monotonically decrease to zero and then

monotonically increase to one as reference signal power varies in¿. Fourth, when two

nonreferential signals have positive cross-power, the MSC of the two referential signals can

monotonically increase to one as the reference signal power varies in¿. In general, the

reference signal with small amplitude or power relative to the signals of interest may decrease

or increase the values of correlation, phase synchrony, and MSC.

However, the reference signal with high relative amplitude or power will always

increase each of the three measures. In our previous paper, we developed a method to identify

and extract the reference signal contribution to intracranial EEG (iEEG) recordings. In this

paper, we apply this approach to referential iEEG recorded from human subjects and directly

investigate the contribution of recording reference on correlation, phase synchrony, and

xv

Abstract

MSC. The experimental results demonstrate the significant impact that the recording

reference may have on these bivariate measures [2].

1.5. Objective of this Thesis Work:The main objectives of this thesis work are:

Acquisition of Electrocardiogram, Electroencephalogram and Respiratory signal

simultaneously.

Calculation of respiratory rate simultaneously with the ECG, EEG and Respiration

Analysis of coherence and phase coherence between the ECG and EEG signals

acquired from the different brain regions to investigate that which region of brain is

more functionally associated to the corresponding heart signal.

The study of heart and heart sound chosen for Thesis work due to following reasons:

According to the world health organisation (WHO), Indians are the much greater risk

of contracting heart diseases and brain disorders than other nationalities.

As per the recent report of THE TIMES OF INDIA, dated April 28, 2010, Dr. Pratap

C. Reddy, Founder Chairman, Appllo Hospital, Says, “Reports point to India

becoming the heart disease capital of the world, if we haven’t become it already. This

is a dubious distinction. ”

According to recent report of the Indian health ministry, 10 percent of adults suffer

from hypertension and the country is home to 25-30 millions diabetics. The number of

deaths from heart attack is projected to increase to two millions in 2010.

xv

Abstract

Chapter 2

PHYSIOLOGICAL DATA ACQUISITION

Physiological signals like ECG, EEG, Heart Rate, Air flow and respiration rate etc.

are measured by using Biopac Inc. MP100 System. The MP100 unit takes incoming signals

and converts them into digital signals that can be processed with your computer. Data

collection generally involves taking incoming signals (usually analog) and sending them to

the computer, where they are (a) displayed on the screen and (b) stored in the computer’s

memory (or on the hard disk). In this dissertation work, ECG, EEG and respiratory signals

are measured by using mainly MP100 system simultaneously. The ECG, EEG and respiratory

signals are recorded for different subject and for the same subject under different condition.

The physiological signals are acquired from 65 persons in the age group 19-36 years for this

study. The physiological signals are acquired in the software environment of Biopac

Acqknowledge3.9.0 on the computer that is connected through the USB data cable.

2.1 MP100 System:

2.1.1 Introduction

The MP100 system is computer based data acquisition system that perform many of

the same function as a chart recorder or other data viewing device, but is superior to such

device in that it transcends the physical limits commonly encountered (such as paper width or

speed).The MP data acquisition unit (MP100) is the heart of MP system. The MP unit takes

incoming signal and converts them into digital signal that can be processed with your

computer. MP system can be used for a wide range of application, including

ECG

EEG

EMG

EOG

xv

Abstract

GSR

Evoked response

Plethsmography

Pulmonary Function etc.

Data collection generally involves taking incoming signal (usually analog) and

sending them to computer , where they are (a) displayed on the screen and (b)store on the

computer memory(or on the hard disk).These signals can then be stored for future

examination much as a word processor stores a document or a statistics program saves a data

file. Graphical and numerical representation of the data can also be produced for use for other

program.

The MP100 system offers USB –ready data acquisition and analysis. It record

multiple channels differing sample rates, at speed up to 70kHz .The system is designed to

satisfy the following Medical Safety Test Standards affiliated with IEC601-1:

1. Creep age and air Clearance

2. Dielectric Strength

3. Patient Leakage Current

2.1.2 MP100 Block Diagram

Figure 2.1(a): Block Diagram of Multi-channel Data Acquisition System Biopac Inc. MP100

xv

Abstract

Figure 2.1(b): Hardware Components of Multi-channel Data Acquisition System Biopac Inc. MP100

Table 2.1: Experimental Hardware Setup

Sr. No.

Acquisition Hardware Hardwares Name

1. Data acquisition unit MP100C2. Universal interface module UIM100C3. USB adapter USB1W(PC) or USB1M(Macintosh)4. Transformer AC100A5. Cables CBLSERA cable, CBLS100 cable set

2.1.2.a Signal source

Transducers are biopotential electrodes are the source of signal for the biopack

system. These sensors are systematically placed on the subject and connected to the signal

conditioning modules through connecting leads. Ag-Agcl disposable electrodes are used for

ECG recording.

2.1.2.b Signal conditioning modules

Signals from the sensor are given to signal conditioning module for amplification and

filtering. The MP100 system has 9 signal conditioning module for different physiological

variable. It has 3 ECG100c modules for ECG recording, two EEG100C module for ECG

recording, SKT100C is for temperature recording, and RSP100c module for respiration rate

measurement. Besides the above module it also has two general purpose amplifier modules

which can be used for displacement sensor or blood pressure measurement. No. of channel

can be extended up to 16.

2.1.2.c UIM100 C Universal Interfacing module

The UIM100C universal interfacing module is the interface between the MP100 and

external device. Typically the UIM100C is used to input preamplifier signal and digital

signal to the MP100 acquisition unit. Other signal connect to various signal condition

xv

Abstract

modules. The universal interface module is designed to serve as a general purpose interface

to most type of laboratory equipment. The UIM100C consist of sixteen 3.5mm miniphone

jack connector for analog input, to 3.5mm miniphone connector for analog output, and screw

terminals for the sixteen digital lines, external trigger and supply voltage.

The UIM100C allow access to sixteen analog input and two analog output on one

side, and sixteen digital i/o lines, and external triggers, and supply voltage on the other side.

The UIM100C is designed to compatible with a variety of different input device, including

the biopack series of signal conditioning amplifier.

Connection between UIM100C and the MP100 acquisition unit are made via two

cable one for analog signal and one digital signal. Use the 6 meters cables included with your

system to connect the UIM100C to the acquisition unit. UIM100C enables the main device to

communicate with other device. It control the polling and interrupt in the communication

[Appendix].

2.1.2.d MP100 data acquisition unit

MP100 system has and internal microprocessor to control the data acquisition and

communication with the computer. There are sixteen analog input channel, two analog output

channel , sixteen digital channel that can be used for either input or output , and an external

trigger input.

Table 2.2: Specification of data acquisition unit Biopac Inc. MP100

Sr. No

Parameters Corresponding Values

1. No. of Analog Channel 162. Input Voltage Range ±10V3. Accuracy ±0. 0034. A/D Resolution 16 Bits5. No. of Digital Channel 166. Output Voltage Range ±10V7. Output Derive Current ±5mA8. No. of Calculation Channel 16

xv

Abstract

0.00000 0.50000 1.00000 1.50000 2.00000 2.50000seconds

-0.001322

0.000000

0.001322

0.002645

Volts

EEG

-0.047506

0.000000

0.047506

0.095011

Volts

ECG

9.052734

9.429932

9.807129

10.184326

Volts

RSP

173.754473

175.474814

177.195156

178.915497

BPM

Hea

rt R

ate

19.920319

19.946879

19.973440

20.000000

BPM

Res

pira

tion Rat

e

Figure 2.2: Graph of Experimental Data Acquired using MP100 and Acqknowledge3.9.0

Figure 2.3: Snap of Subject and Technician during Data Acquisition using MP100 and Acqknowledge3.9.0

2.2 ECG100C – Electrocardiogram Amplifier Module

The electrocardiogram amplifier module (ECG100C) is a single channel, high gain,

differential input, biopotential amplifier designed specifically for monitoring the heart’s

electrical activity [Appendix].

2.2.1 Applications:

Einthoven’s triangle potential measurement (3-lead ECG)

Chaos investigations (heart rate variability)

xv

Abstract

Heart arrhythmia analysis

Exercise physiology studies

The ECG100C will connect directly to any of BIOPAC Systems, Inc.’s series of Ag-

AgCl lead electrodes. Use two shielded electrodes (EL208S) for the signal inputs and one

unshielded electrode (EL258S) for the ground.

2.2.2 ECG100C Calibration

The ECG100C is factory set and does not require calibration.

Table 2.3: ECG100C Specifications

Sr. No. Parameters Corresponding Range values1. Gain 500, 1000, 2000, 5000 2. Output Selection Normal, R-wave indicator3. Output Range ±10V (analog)4. Frequency Response Low Pass

Filter35Hz, 150Hz

5. High Pass Filter 0.05Hz, 1.0Hz6. Notch Filter 50dB rejection @ 50/60Hz7. Noise Voltage 0.1µV rms – (0.05-35Hz)8. Signal Source Electrodes (three electrode leads required)9. Z (input) Differential 2MΩ10. Input Voltage Range Gain Vin (mV)

500 ±20 1000 ±10 2000 ±5 5000 ±2

11. Weight 350 grams12. Dimensions 4cm (wide) x 11cm (deep) x 19cm (high)

xv

Abstract

Figure 2.4: The electrode connections to the ECG100C for the measurement of Lead I. Signals from this electrode montage can be used to calculate BPM and general-purpose ECG applications.

2.3 EEG100C – Electroencephalogram Amplifier Module The electroencephalogram amplifier module (EEG100C) is a single-channel, high-

gain, differential input, biopotential amplifier designed specifically for monitoring the

neuronal activity of the brain [Appendix].

2.3.1 Applications:

Conventional EEG (16 channel, unipolar or bipolar)

Sleep studies

Epilepsy investigations

Evoked responses

Tumor pathology studies

Cognition studies

The EEG100C will connect directly to any of BIOPAC Systems, Inc.’s series of Ag-

AgCl lead electrodes. Typically, EL503 electrodes are recommended for evoked response

measurements. Use two shielded electrodes (LEAD110S) for the signal inputs and one

unshielded electrode (LEAD110) for ground.

2.3.2 EEG100C Calibration

xv

Abstract

The EEG100C is factory set and does not require calibration. To confirm the accuracy

of the device, use the CBLCALC.

Table 2.4: EEG100C Specifications

Sr. No. Parameters Corresponding Range values1. Gain 500, 1000, 2000, 5000 2. Output Selection Normal, Alpha Wave indicator3. Output Range ±10V (analog)4. Frequency Response Low Pass

Filter35Hz, 100Hz

5. High Pass Filter 0.1Hz, 1.0 Hz6. Notch Filter 50dB rejection @ 50/60Hz7. Noise Voltage 0.1µV rms – (0.05-35Hz)8. Signal Source Electrodes (three electrode leads required)9. Z (input) Differential 2MΩ10. Input Voltage Range Gain Vin (mV)

5000 ±2 10000 ±1 20000 ±0.5 50000 ±0.2

11. Weight 350 grams12. Dimensions 4cm (wide) x 11cm (deep) x 19cm (high)

Figure 2.5(a): Bipolar EEG electrode leads placement, (b) International 10-20 electrode placement on the

different brain regions

2.4 RSP100C - Respiration pneumogram amplifier module

The RSP100C respiration pneumogram amplifier module is a single channel,

differential amplifier designed specifically for recording respiration effort. The RSP100C is

designed for use in the following [Appendix].

2.4.1 Applications:

xv

Abstract

Allergic responses analysis

Exercise physiology studies

Psychophysiological investigations

Respiration rate determination

Sleep studies

The RSP100C works with the TSD201 respiration transducer to measure abdominal

or thoracic expansion and contraction. The RSP100C includes a lower frequency response

selection switch that permits either absolute (DC) or relative (via a 0.05 high pass filter)

respiratory effort measurements. The following illustration shows the placement and

connections for recording thoracic respiration effort using the RSP100C and the TSD201

respiration transducer.

2.4.2 Frequency Response Characteristics

The 0.05Hz lower frequency response setting is a single pole roll-off filter. The 0.5Hz

lower frequency response setting is a two pole roll-off filter. Modules can be set for 50 or

60Hz notch options, depending on the destination country.

2.4.3 RSP100C Calibration

None required.

Table 2.5: RSP100C Specifications

Sr. No. Parameters Corresponding Range Values1. Gain 10, 20, 50, 1002. Output Selection Normal, Alpha Wave indicator3. Output Range ±10V (analog)4. Frequency Response: Low Pass Filter 1Hz, 10 Hz 5. Frequency Response: High Pass Filter 0.05 Hz, 0.5 Hz6. Notch Filter 50dB rejection @ 50/60Hz7. Noise Voltage 0.2µV rms – amplifier contribution 8. Signal Source TSD2019. Excitation Voltage ±0.5 V 10. Weight 350 grams11. Dimensions 4cm (wide) x 11cm (deep) x 19cm

(high)

xv

Abstract

Figure 2.6: The placement and connections for recording thoracic respiration effort using the RSP100C and the TSD201 respiration transducer.

2.5. Various Functions of AcqKnowledge3.9.0 software:

(a) (b)

xv

Abstract

(c)

(d) (d)

(e)

xv

Abstract

(f) (g)

(h) (i)

(j) (k) (l)

Figure 2.7: (a) Transform tool bar, (b) Graph window function tool bar, (c) Acquisition set up, (d) On opening the new graph, graphic journal, (e) Setting channel label, (f) Set screen horizontal axis, (g) Set screen horizontal

axis, (h) Channel selection, (i) Cursor tools, (j) Edit tool bar, (k) Clipboard, (l) Journal

For detail description goes to appendix (www.biopac.com).

Chapter 3

COHERENCE AND PHASE COHERENCE FUNCTION

3.1. IntroductionIt is necessary to be able to quantify the degree of interdependence of one process

upon another, or to establish the similarity between one set of data and another. Correlation

can be defined mathematically and can be quantify. Consider how to data sequences each

consisting of simultaneously sampled values taken from the two corresponding waveforms or

signals. If the two signals varied similarly point for point, then a measure of their correlation

might be taken by the sum of the products of the corresponding pairs of points [39].

3.1.1. Cross-correlation

xv

x

nj

Abstract

The Discrete cross-correlation r12 between two data sequences x1(n) and x2(n) each

containing N data might therefore be written as

r12=∑n=0

N−1

x1 (n ) x2 (n )(3.1)

The definition of cross-correlation, however produces a result whitch depends on the

number of sampling points taken. This is corrected for the normalizing the result to the

number of points by dividing by N . Alternatively this may be regarded as averaging the sum

of products. Thus an improved definition is

r12=1N ∑

n=0

N−1

x1 (n ) x2 (n )(3.2)

However, the signals are highly correlated, although they are out of phase.

Figure 3.1: One signal lags by j unit with the other signal

As illustrated in the above figure 3.1 this is equivalent to changing x2(n) to x2 (n+ j ) ,

where j represents the amount of lag which is the number of sampling points by which x2 has

been sifted to the left. An alternative, but equivalent, procedure is to sift x1 to the right. The

formula for the cross-correlation thus becomes

r12 ( j )= 1N ∑

n=0

N−1

x1 (n ) x2 (n+ j )(3.3)

r21 (− j )= 1N ∑

n=0

N−1

x2 (n ) x1 (n− j )(3.4)

xv

Abstract

Of course, it is also possible to consider correlation in the continuous time domain,

and some analog signal correlation is implemented in this way, in the continuous domain

n → t and j→ τ and

r12 (τ )= limT →∞

1T ∫

−T /2

T /2

x1 (t ) x2(t+ τ )dt (3.5)

However, if x1 (t ) and x2 ( t ) are periodic with period T o the above equation 3.5 simplified to

r12 (τ )= limT →∞

1T o

∫−T o /2

To /2

x1 ( t ) x2(t +τ )dt(3.6)

If the signals are finite energy signals, for example non-periodic pulse-type signals,

then average evaluated over time T as T →∞ is not taken because 1/T → 0 and r12(τ ) is

always vanishingly small. For this case above equation 3.6 is used in the principle.

r12 (τ )=∫−∞

+∞

x1 ( t ) x2(t +τ )dt(3.7)

In practice, a finite record length will be processed and so equation 3.7 will be

applied:

r12 (τ )= 1T ∫

0

T

x1 (t ) x2(t+τ )dt (3.8)

xv

Abstract

0 1000 2000 3000 4000 5000 6000-0.2

0

0.2A

mpl

itude

(V)

No. of samples of ECG Signal(a)

0 1000 2000 3000 4000 5000 6000-2

0

2

4x 10

-3

No. of samples of EEG Signal(b)

Am

plitu

de(V

)

-80 -60 -40 -20 0 20 40 60 800.085

0.09

0.095

Lags b/w ECG and EEG(C3-C4)(c)

Cro

ss-C

orre

latio

n

Figure 3.2(a): ECG signal having 5006 samples with sampling rate 500 samples/sec (b) EEG Signal having 5006 samples with sampling rate 500 samples/sec (c) Cross-correlation between the ECG signal and the EEG

signal3.1.2. Autocorrelation

Autocorrelation is the cross-correlation of a signal with itself. Informally, it is the

similarity between observations as a function of the time separation between them. It is a

mathematical tool for finding repeating patterns, such as the presence of a periodic signal

which has been buried under noise, or identifying the missing fundamental frequency in a

signal implied by its harmonic frequencies. It is often used in signal processing for analyzing

functions or series of values, such as time domain signals [38].

3.1.2.a Continuous Auto-Correlation:

The autocorrelation function gives an average measure of the time-domain properties

of a signal waveform. It is defined as (3.9):

r11 ( τ )= limT → ∞

1T ∫

−T /2

T /2

x1 ( t ) x1( t+ τ)dt (3.9)

xv

Abstract

This is the average product of the signal, x1 (t ), and a time-shifted version of itself,

x1 (t +τ ). The expression above applies to the case of a continuous signal of infinite duration.

In practice, the intervals must be finite and it is necessary to use a modified version as given

by (3.10).

r11 ( τ )= limT → ∞

1T ∫

−T /2

T /2

x1 (t ) x1( t+τ)dt (3.10)

The autocorrelation function may be applied to deterministic as well as random

signals. Each of the frequency components in the signal x1 (t ) produces a corresponding term

in the autocorrelation function having the same period in the time-shifted variable, τ , as the

original component has in the time variable, t. The amplitude is equal to half of the squared

value of the original.

3.1.2.b Discrete Auto-Correlation:

Discrete auto-correlation is the cross-correlation between the discrete/sampled signal

and signal itself and given as

r11 ( j )= 1N ∑

n=0

N−1

x1 (n ) x1 (n+ j )(3.11)

r22 (− j )= 1N ∑

n=0

N−1

x2 (n ) x2 (n− j )(3.12)

where j is called the lag between the two sampled signals. x1 (n ) and x2 (n ) are the

discrete/sampled signals. N is the length of the sampled signal, means no. of samples taken in

the sampled signal and(0≤ n ≤ N−1)[39].

xv

Abstract

-100 -50 0 50 1001

2

3

4

5

6

7

Lags b/w ECG and ECG itself(b)

Aut

o-co

rrela

tion

0 1000 2000 3000 4000 5000 6000-2

-1

0

1

2

3x 10

-3

No. of samples of EEG Signal(c)

Am

plitu

de(V

)

0 1000 2000 3000 4000 5000 6000-0.05

0

0.05

0.1

0.15A

mpl

itude

(V)

No. of samples of ECG Signal(a)

-100 -50 0 50 1002

2.5

3

3.5

4

4.5

5

5.5x 10

-3

Lags b/w EEG & EEG itself(d)

Aut

o-co

rrela

tion

Figure 3.3(a): ECG signal having 5006 samples with sampling rate 500 samples/sec (b) Auto-correlation of ECG signal (c) EEG Signal having 5006 samples with sampling rate 500 samples/sec (d) Auto-correlation of the

EEG signal3.1.3. Correlation Co-efficient Definition:

It is a measure of the strength of linear association between two variables. Correlation

will always between -1.0 and +1.0. If the correlation is positive, we have a positive

relationship. If it is negative, the relationship is negative.

How to Interpret a Correlation Coefficient

The sign and the absolute value of a correlation coefficient describe the direction and the

magnitude of the relationship between two variables.

The value of a correlation coefficient ranges between -1 and 1.

The greater the absolute value of a correlation coefficient, the stronger

the linear relationship.

The strongest linear relationship is indicated by a correlation coefficient of -1 or 1.

The weakest linear relationship is indicated by a correlation coefficient equal to 0.

A positive correlation means that if one variable gets bigger, the other variable

tends to get bigger.

xv

Abstract

A negative correlation means that if one variable gets bigger, the other variable

tends to get smaller.

3.2. Spectral Density Functions:

Spectral density functions can be derived in several ways. One method takes the

Direct Fourier Transform of previously calculated autocorrelation and cross-correlation

functions to yield the two-sided spectral density functions given in (3.13).

Sxx ( f )=∫−∞

+∞

R xx ( τ ) e− j 2 πfτ dτ (3.13)

Syy ( f )=∫−∞

+∞

Ryy (τ ) e− j 2 πfτ dτ (3.14)

Sxy ( f )=∫−∞

+∞

R xy (τ ) e− j 2πfτ dτ (3.15)

These integrals always exist for finite intervals. The quantities Sxx(f ) and Syy ( f ) are

the auto spectral density functions of signals x(t) and y(t) respectively,

Example of auto power spectral density as:

0 1000 2000 3000 4000 5000 6000-0.05

0

0.05

0.1

0.15

No. of Samples of ECG Signal(a)

Am

plitu

de(V

)

0 0.2 0.4 0.6 0.8 1-70

-60

-50

-40

-30

-20

-10

Normalized Frequency ( rad/sample)(b)

Pow

er/fr

eque

ncy

(dB

/rad/

sam

ple) Welch Power Spectral Density Estimate

0 1000 2000 3000 4000 5000 6000-2

-1

0

1

2

3x 10

-3

No. of Samples of EEG Signal(c)

Am

plitu

de(V

)

0 0.2 0.4 0.6 0.8 1-80

-75

-70

-65

-60

-55

-50

Normalized Frequency ( rad/sample)(d)

Pow

er/fr

eque

ncy

(dB

/rad/

sam

ple) Welch Power Spectral Density Estimate

Figure 3.4(a): ECG signal having 5006 samples with sampling rate 500 samples/sec (b) Auto power spectral density estimate of ECG signal (c) EEG Signal having 5006 samples with sampling rate 500 samples/sec (d)

Auto power spectral density estimate of EEG signal

and Sxy(f )is the cross-spectral density function between x (t) and y (t ).

xv

Abstract

Example of cross power spectral density as:

0 1000 2000 3000 4000 5000 6000-0.2

0

0.2

No. of samples of ECG Signal(a)

Am

plitu

de(V

)

0 1000 2000 3000 4000 5000 6000-2

0

2

4x 10

-3

No. of samples of EEG Signal(b)

Am

plitu

de(V

)

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-100

-50

0

Normalized Frequency ( rad/sample)(c)

Pow

er/fr

eque

ncy

(dB

/rad/

sam

ple)

Welch Cross Power Spectral Density Estimate

Figure 3.5(a): ECG signal having 5006 samples with sampling rate 500 samples/sec (b) EEG Signal having 5006 samples with sampling rate 500 samples/sec (c) Cross-power spectral density between the ECG signal and

the EEG signalIn terms of physically measurable one-sided spectral density functions where f varies over

(0 , ∞), the results are given in (3.16):

G xx ( f )=2S xx ( f )(3.16)

G yy ( f )=2 S yy (f )(3.17)

G xy ( f )=2 Sxy ( f )(3.18)

Note that the cross-spectrum is a complex function with real and imaginary functions, where

C xy ( f ) is the coincident spectral density function (cospectrum) and Q xy(f ) is the quadrature

spectral density function (quad-spectrum) as shown in (3.19).

G xy ( f )=2∫−∞

+ ∞

R xx(τ )e− j2 πfτ dτ=C xy ( f )− jQ xy (f )(3.19)

xv

Abstract

In complex polar notation, the cross-spectral density becomes (3.20).

G xy ( f )=|G xy ( f )|e− j θxy(f )(3.20)

where

|G xy ( f )|=2√C xy2 ( f )+Qxy

2 ( f )(3.21)

θxy (f )=tan−1(Q xy ( f )C xy ( f ) )(3.22)

Here, the auto spectra of the input G xx ( f ) and the cross-spectra G xy ( f ). Nevertheless, the

complete frequency response function with gain and phase can be obtained when both G xx ( f )

and G xy ( f )are known.

3.3. Coherence Function from Spectral Analysis:

Coherence is the degree of relationship or association of frequency spectra between

the ECG and EEG signals at a particular frequency. Notes the peaks in the coherence

spectrum in the figure 3.6(a) as, at 0.01Hz, 5Hz, 31Hz, corresponding approximately to the

different respiratory rates as zero(0-4 BPM), 10-14BPM (Normal Breathing Rate) and 16-

20BPM(High Breathing Rate). The spectral content for each lead is highly similar regardless

of the lead configuration, although the actual energy at each frequency may differ. The

magnitude squared coherence estimate between two signals x (ECG Signal) and y (EEG

Signal), is

γ xy2 =Cxy ( f )=

|Pxy ( f )|2

Pxx ( f )×P yy ( f ) (3 .23)

Here C xy( f ) or γ xy2

is the magnitude squared coherence between the ECG and EEG signals.

Coherence phase is given as

θ( f )=tan−1Im Pxy Re Pxy (3 . 24 )

xv

Abstract

Where Pxx( f ) is the power spectral estimate of x (ECG Signal), Pyy ( f ) is the power spectral

estimate of y (EEG Signal), and P xy is the cross power spectral estimate of x and y.

Coherence is a function of frequency with C xy( f ) ranging between 0 and 1 and indicates how

well signal x corresponds to signal y at each frequency. The degree of synchronization in

electroencephalography (EEG) signal and ECG signal is commonly characterized by

coherence phase and magnitude squared coherence (MSC).

Example of coherence b/w ECG and EEG (C3−C4) signals:

0 1000 2000 3000 4000 5000 6000-0.2

0

0.2

Am

plitu

de(V

)

No. of samples of ECG Signal(a)

0 1000 2000 3000 4000 5000 6000-2

0

2

4x 10

-3

No. of samples of EEG Signal(b)

Am

plitu

de(V

)

0 5 10 15 20 25 30 350

0.5

1

Coh

eren

ce

Frequency Band(Hz)(c)

Figure 3.6(a): Figure 3.5(a): ECG signal having 5006 samples with sampling rate 500 samples/sec (b) EEG Signal having 5006 samples with sampling rate 500 samples/sec (c) Coherence between the ECG signal and the

EEG signalThe coherence is a frequency domain function with observed values ranging from 0 to

1. At each frequency where the coherence function is performed, it represents the fraction of

the power output related to input. If the coherence function is less than 1, then there are three

possible explanations:

1. There is noise in the system or

2. The system has some nonlinearity generating energy at other frequencies or

xv

Abstract

3. There are other inputs into the system that have not be accounted for [38].

Chapter 4

xv

Abstract

COHERENCE ANALYSIS BETWEEN ECG AND EEG

Primarily we analysed the coherence and the phase coherence between the ECG and

EEG signals acquired at the different respiratory rates. The ECG and EEG signals acquired

from the normally healthy subject at the different respiratory rates as

Zero airflow (0-4breaths/min).

Normal airflow (10-12 breaths/min).

High airflow (16-20breaths/min).

The ECG and EEG signals acquired simultaneously with the respiratory signal. Data

collection is done in the Biomedical Instrumentation Laboratory in the department of

Instrumentation and Control Engineering, National Institute of Technology Jalandhar. The

Multi-channel data acquisition Biopac Inc. MP100 system is used to acquire data from the

healthy subject of the age group (20-35 Years).

4.1. ECG and EEG Signals:

0 1000 2000 3000 4000 5000 6000-0.5

0

0.5

1ECG Signals

0 1000 2000 3000 4000 5000 6000-0.5

0

0.5

1

0 1000 2000 3000 4000 5000 6000-0.5

0

0.5

1

Figure 4.1 ECG Signals at the respiratory rates. First signal in the figure 6.1 is at nearly zero breathing rate for 9.99 seconds and similarly second and third signals at the 10 to 12 BPM(breaths per minute) and 15 to

20 BPM.

xv

Abstract

0 1000 2000 3000 4000 5000 6000-0.2

-0.1

0

0.1

0.2EEG Signals

0 1000 2000 3000 4000 5000 6000-0.2

-0.1

0

0.1

0.2

0 1000 2000 3000 4000 5000 6000-40

-20

0

20

Figure 4.2 EEG Signals respiratory rates. First signal in the Figure 6.1(a) is at nearly zero breathing rate for 9.99 seconds and similarly second and third signals at the 10 to 12 BPM and 15 to 20 BPM.

Table 4.1 Different parameters of the acquired signals

Group Signals Max Min Mean Stddev

No Airflow

ECG 0.6845 -0.4263 0.0030 ±0.1096EEG 0.1529 -0.1462 0.0020 ±0.0487HR 112.149

5102.5641 106.0609 ±2.4184

Normal Airflow

ECG 0.6647 -0.4306 2.9511e-004

±0.1105

EEG 0.1407 -0.1602 -0.0024 ±0.0498HR 88.8888 70.5882 75.5919 ±4.8133

High Airflow

ECG 0.7526 -0.2954 0.0017 ±0.1158EEG 0.01760 -0.02082 0.00028 ±0.0050HR 92.3077 71.4285 82.3465 ±7.2561

Table 4.2 Signals Acquisition Settings

Signals AnalogChannel

Unit Window(Vertical)

ECG A3 mV -1000 to +1000(mV)EEG A1 mV -1000 to +1000(mV)

Airflow A6 V -2 to +2(V)HR C3 BPM 40 to 180 (BPM)

Resp. Rate

C6 BPM 0 to 25 (BPM)

xv

Abstract

4.2. Coherence and Phase Coherence between the ECG and EEG signals acquired at different breathing rates:

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

0.2

0.4

0.6

0.8Coherence amongs the ECG and EEG Signals

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

0.2

0.4

0.6

0.8

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

0.05

0.1

0.15

0.2

Figure 4.3 Coherence between the ECG and EEG signals respiratory rates

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5-2

-1

0

1

2Phase Coherence amongs the ECG and EEG Signals

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5-2

-1

0

1

2

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5-2

-1

0

1

2

Figure 4.4 Phase Coherence between the ECG and EEG signals respiratory rates

xv

Abstract

The coherence between the ECG signal and EEG signal at low airflow means low

respirator rate is found to be just lesser than 0.5 that is 0.4650 near the frequency 0.1Hz that

is also called the respiratory frequency. Similarly the coherence between the ECG and EEG

signal at the normal airflow means the normal respiratory rate is found to be more than 0.5

that is 0.6004 near the frequency 0.1Hz. Poor coherence is found between the ECG and

corresponding EEG signal in the frequency range (0¿0.5 Hz) at the higher respiratory rates.

Table 4.3 Coherence analysis of results at different respiratory rates

Group Results Max Min Mean StddevNo Airflow

Coherence 0.4650 0.0045 0.1208 ±0.0948Phase

Coherence1.3925 -1.3672 -

0.1626±0.6080

Normal Airflow

Coherence 0.6004 0.0033 0.0867 ±0.1215Phase

Coherence1.5423 -1.4865 -

0.2222±0.8914

High Airflow

Coherence 0.1882 0.0063 0.0866 ±0.0428Phase

Coherence1.5528 -1.5106 -

0.0351±0.8550

The continuous flow of blood throughout the human body is maintained by body

organ; called as Heart. The mechanical function of heart is reflected in the electrical form; is

called Electrocardiogram and also called ECG. A continuous non-invasive, low cost and

accurate monitoring of functioning of heart has been proven to be invaluable in various

diagnostics applications and clinical applications.

The electrical activity of the human brain is reflected as Electroencephalogram and

also called EEG. The EEG signals arises from the fact that these waveforms provide the non-

invasive diagnostic tool in a wealth of disorders that include epilepsy and comma assessment

in intensive care unit.

We acquired 50 ECG and EEG signals simultaneously using Biopac Inc.

Acqknowledge3.9.0 software and MP100 hardware for this work. All data collected from

healthy subjects under the age group (21-36 years old) at the sampling rate is 500

samples/second. The number of samples used for the analysis of association or correlation is

5006 for each signal. We also acquired respiratory rate by setting the Calculate Channel in

the Acqknowledge3.9.0 software simultaneously with corresponding ECG, EEG and

Respiratory signals. The data acquisition is done at different Respiratory Rates. The different

xv

Abstract

respiratory rates are 0-4 breaths/minute (Low Breathing Rate), 10-12 breaths/minute (Normal

Breathing) and 16-20 breaths/minute (High Breathing Rate). The EEG signals acquired from

the four different positions; the Frontal(F p1−Fp 2

), Central(C3−C4), Parietal (P3−P4) and

Occipital (O1−O2) Brain Regions.

Table 4.4 ECG and EEG signals fron 1 to 25 subjects statistics

Subjects Signals Max(V)

Min(V ) Mean(V)

Stddev(± V)

Median(V)

Subject 1 ECG 0.11200

-0.02319 0.01980 0.02905 0.00427

EEG 0.00458

-0.00153 0.00086 0.00064 0.00092

Subject 2 ECG 0.11200

-0.02777 0.01879 0.02744 0.00488

EEG 0.00549

-0.00214 0.00089 0.00077 0.00092

Subject 3 ECG 0.11108

-0.01495 0.01841 0.02668 0.00397

EEG 0.00275

-0.00183 0.00087 0.00063 0.00092

Subject 4 ECG 0.11383

-0.02411 0.01716 0.02809 0.00275

EEG 0.00244

-0.00153 0.00089 0.00059 0.00092

Subject 5 ECG 0.10376

-0.03326 0.01837 0.02854 0.00366

EEG 0.00366

-0.00122 0.00088 0.00064 0.00092

Subject 6 ECG 0.10651

-0.02045 0.01691 0.02619 0.00336

EEG 0.00366

-0.00122 0.00088 0.00065 0.00092

Subject 7 ECG 0.11200

-0.02594 0.01559 0.02653 0.00305

EEG 0.00336

-0.00275 0.00087 0.00060 0.00092

Subject 8 ECG 0.11169

-0.01678 0.01647 0.02574 0.00305

EEG 0.01099

-0.00916 0.00085 0.00196 0.00092

Subject 9 ECG 0.10254

-0.01495 0.01931 0.02817 0.00336

EEG 0.01312

-0.01343 0.00088 0.00146 0.00092

Subject 10 ECG 0.10590

-0.01648 0.01536 0.02531 0.00305

EEG 0.0061 -0.00488 0.00088 0.00081 0.00092

xv

Abstract

0Subject 11 ECG 0.1086

4-0.01831 0.01530 0.02511 0.00336

EEG 0.00488

-0.00214 0.00088 0.00067 0.00092

Subject 12 ECG 0.10986

-0.01740 0.01576 0.02639 0.00336

EEG 0.00397

-0.00183 0.00088 0.00060 0.00092

Subject 13 ECG 0.11230

-0.02777 0.01606 0.02724 0.00305

EEG 0.00336

-0.00153 0.00087 0.00057 0.00092

Subject 14 ECG 0.11200

-0.02319 0.01809 0.02837 0.00336

EEG 0.00397

-0.00153 0.00087 0.00057 0.00092

Subject 15 ECG 0.10193

-0.06805 0.02315 0.02898 0.00793

EEG 0.00336

-0.00092 0.00088 0.00056 0.00092

Subject 16 ECG 0.10651

-0.09705 0.01647 0.03005 0.00305

EEG 0.00305

-0.00031 0.00088 0.00055 0.00092

Subject 17 ECG 0.15625

-2.44995 -0.11052 0.38796 0.00183

EEG 0.00824

-0.00824 0.00088 0.00057 0.00092

Subject 18 ECG 0.11688

-0.06653 0.02311 0.03466 0.00397

EEG 0.00244

-0.00061 0.00088 0.00055 0.00092

Subject 19 ECG 0.10437

-0.06348 0.01668 0.02686 0.00397

EEG 0.00214

-0.00031 0.00088 0.00055 0.00092

Subject 20 ECG 0.09003

-0.00610 0.01249 0.02221 0.00183

EEG 0.00275

-0.00092 0.00090 0.00061 0.00092

Subject 21 ECG 0.09033

-0.00549 0.01157 0.02092 0.00183

EEG 0.00305

-0.00092 0.00090 0.00061 0.00092

Subject 22 ECG 0.09064

-0.00580 0.01179 0.02142 0.00153

EEG 0.00488

-0.00183 0.00090 0.00063 0.00092

xv

Abstract

Subject 23 ECG 0.09064

-0.00549 0.01219 0.02287 0.00153

EEG 0.00397

-0.00122 0.00089 0.00065 0.00092

Subject 24 ECG 0.09186

-0.00702 0.01139 0.02227 0.00153

EEG 0.00305

-0.00153 0.00089 0.00063 0.00092

Subject 25 ECG 0.09186

-0.00641 0.01217 0.02276 0.00183

EEG 0.00427

-0.00183 0.00090 0.00063 0.00092

Table 4.5 ECG and EEG signals from 26 to 50 subjects statistics

Subjects Signals

Max(V ) Min(V ) Mean(V)

Stddev(± V)

Median(V)

Subject 26 ECG 0.09918 -0.00610

0.01497 0.02582 0.00275

EEG 0.00275 -0.00061

0.00081 0.00050 0.00092

Subject 27 ECG 0.10071 -0.00580

0.01525 0.02598 0.00305

EEG 0.00275 -0.00061

0.00081 0.00049 0.00092

Subject 28 ECG 0.10040 -0.00580

0.01505 0.02580 0.00305

EEG 0.00244 -0.00061

0.00080 0.00049 0.00092

Subject 29 ECG 0.10468 -0.00793

0.01515 0.02763 0.00244

EEG 0.00641 -0.00244

0.00080 0.00057 0.00092

Subject 30 ECG 0.10468 -0.00793

0.01564 0.02803 0.00244

EEG 0.00519 -0.00122

0.00080 0.00054 0.00092

Subject 31 ECG 0.16174 -0.09430

0.01848 0.03851 0.00275

EEG 0.00946 -0.00732

0.00090 0.00091 0.00092

Subject 32 ECG 0.16235 -0.10895

0.02169 0.04358 0.00214

EEG 0.01740 -0.01343

0.00088 0.00108 0.00092

Subject 33 ECG 0.16327 -0.08636

0.02330 0.04173 0.00305

EEG 0.00671 -0.00275

0.00090 0.00086 0.00092

xv

Abstract

Subject 34 ECG 0.16052 -0.11261

0.02494 0.04402 0.00275

EEG 0.00702 -0.00275

0.00090 0.00084 0.00092

Subject 35 ECG 0.16571 -0.09796

0.02296 0.03991 0.00366

EEG 0.00732 -0.00641

0.00088 0.00111 0.00092

Subject 36 ECG 0.09460 -0.00885

0.01215 0.02384 0.00122

EEG 0.00488 -0.00183

0.00090 0.00070 0.00092

Subject 37 ECG 0.09399 -0.00793

0.01304 0.02470 0.00122

EEG 0.00458 -0.00153

0.00091 0.00069 0.00092

Subject 38 ECG 0.09277 -0.00763

0.01285 0.02398 0.00153

EEG 0.00519 -0.00183

0.00091 0.00074 0.00092

Subject 39 ECG 0.09277 -0.00824

0.01195 0.02326 0.00153

EEG 0.00549 -0.00244

0.00089 0.00113 0.00092

Subject 40 ECG 0.09216 -0.00702

0.01232 0.02333 0.00153

EEG 0.00732 -0.00275

0.00091 0.00071 0.00092

Subject 41 ECG 0.13885 -0.06561

0.01862 0.03619 0.00153

EEG 0.00549 -0.00275

0.00081 0.00126 0.00092

Subject 42 ECG 0.13855 -0.06042

0.01813 0.03632 0.00153

EEG 0.00397 -0.00214

0.00079 0.00126 0.00092

Subject 43 ECG 0.13885 -0.06073

0.01802 0.03617 0.00153

EEG 0.00641 -0.00366

0.00081 0.00128 0.00092

Subject 44 ECG 0.13794 -0.05371

0.01800 0.03600 0.00153

EEG 0.00519 -0.00305

0.00079 0.00135 0.00061

Subject 45 ECG 0.13916 -0.05127

0.01814 0.03536 0.00183

EEG 0.00519 -0.00183

0.00079 0.00122 0.00092

Subject 46 ECG 0.09308 - 0.01333 0.02417 0.00122

xv

Abstract

0.00702EEG 0.00641 -

0.002140.00091 0.00067 0.00092

Subject 47 ECG 0.09033 -0.00580

0.01108 0.02058 0.00153

EEG 0.00275 -0.00092

0.00090 0.00060 0.00092

Subject 48 ECG 0.09247 -0.00916

0.01256 0.02398 0.00122

EEG 0.00458 -0.00153

0.00090 0.00066 0.00092

Subject 49 ECG 0.09369 -0.00824

0.01276 0.02392 0.00122

EEG 0.00366 -0.00092

0.00090 0.00062 0.00092

Subject 50 ECG 0.09186 -0.00610

0.01120 0.02204 0.00153

EEG 0.00305 -0.00153

0.00089 0.00062 0.00092

xv

Abstract

Chapter 5

Results and Discussion

All the organs of the human body have some synchronism, association and correlation

to each other. In this work we investigate the coherence and phase coherence between the

ECG and EEG; means the association between the human brain and heart. These signals have

proper responses in some specific frequency bands. We acquired 50 ECG and EEG signals

simultaneously using Biopac Inc. Acqknowledge3.9.0 software and MP100 hardware for this

work. All data collected from healthy subjects under the age group (21-36 years old) at the

sampling rate is 500 samples/second. The number of samples used for the analysis of

coherence and phase coherence is 5006 for each signal. The EEG signals acquired from the

four different positions; the Frontal(F p1−Fp 2

), Central(C3−C4), Parietal (P3−P4) and

Occipital (O1−O2) Brain Regions.

5.1 Coherence analysis for first subject:5.1.1 ECG and EEG signals

0 1000 2000 3000 4000 5000 6000-0.2

0

0.2

(a)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-2

0

2

4x 10-3

(b)

0 1000 2000 3000 4000 5000 6000-0.2

0

0.2

(c)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-2

0

2

4x 10-3

(d)

0 1000 2000 3000 4000 5000 6000-0.2

0

0.2

(e)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-5

0

5x 10-3

(f)

0 1000 2000 3000 4000 5000 6000-0.2

0

0.2

No. of samples taken of ECG Signal(g)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-2

0

2

4x 10-3

No. of samples taken of EEG Signal(h)

Figure 5.1 (a) & (b) ECG signal and corresponding EEG (Fp1-Fp2) signal (Each signal is sampled at the sampling rate 500 samples/second and No. of samples taken for each signal is 5006) of the S1. (c) & (d) ECG

xv

Abstract

signal and corresponding EEG (C3-C4) signal. (e) & (f) ECG signal and corresponding EEG (P3-P4) signal. (g) & (h) ECG signal and corresponding EEG (O1-O2) signal.

5.1.2 Coherence between ECG and EEG signals

Coherence between the ECG and corresponding EEG signals acquired from the four

prominent brain regions named as the Frontal(F p1−Fp 2

), Central(C3−C4), Parietal (P3−P4)

and Occipital (O1−O2) is investigated as:

It is shown in figure 5.2(a) the mean of coherence is found to be 0.14019 in the

frequency band (0-35Hz) and the maximum coherence is 0.99601 near the frequency 0.1Hz.

There are another two coherence peaks are found near the frequency range 16Hz to 21Hz.

It is shown in figure 5.2(b) the mean of coherence is found to be 0.13861 in the

frequency band (0-35Hz) and the maximum coherence is 0.99281 near the frequency 0.1Hz.

There are another two coherence peaks are found, one is near the frequency range 4.9Hz and

another near the 31Hz.

It is shown in figure 5.2(c) the mean of coherence is found to be 0.14399 in the

frequency band (0-35Hz) and the maximum coherence is 0.99142 near the frequency 0.1Hz.

There are another three coherence peaks are found, one is near the frequency 7Hz, second is

near the frequency 26Hz and third is near the frequency 35Hz.

It is shown in figure 5.2(d) the mean of coherence is found to be 0.15198 in the

frequency band (0-35Hz) and the maximum coherence is 0.99663 near the frequency 0.1Hz.

There are another two coherence peaks are found, one is near the frequency 9.5Hz and

another is near the frequency 35Hz.

Standard deviation means the variation in the coherence from the mean in both

directions (Upward or positive and Downward or negative) is generally increasing

continuously from the coherence between ECG and EEG signals from the Frontal(F p1−Fp 2

),

Central(C3−C4), Parietal (P3−P4) and Occipital (O1−O2) respectively.

xv

Abstract

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

(a)

Cohe

renc

e

Fp1-Fp2 y mean y median y std

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

(b)

C3-C4 y median y mean y std

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

Frequency Band(Hz)(c)

Cohe

renc

e

P3-P4 y mean y median y std

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

Frequency Band(Hz)(d)

O1-O2 y mean y median y std

Figure 5.2 (a) Coherence between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to 35 Hz) for S1. (b) Coherence between ECG and EEG (C3-C4). (c) Coherence between ECG and EEG (P3-P4). (d) Coherence

between ECG and EEG (O1-O2).5.1.3 Phase Coherence between ECG and EEG signals

Phase coherence is the measure of the phase induced by the one signal to another

signal at a particular frequency. Here, it is measured in radians. The mean of phase coherence

is found to be maximum when it is measured between the ECG signal and corresponding

EEG signal acquired from the frontal (F p1−Fp 2

) region.

0 5 10 15 20 25 30 35-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

(a)

Cohe

renc

e Ph

ase

Fp1-Fp2 y mean y median y std

0 5 10 15 20 25 30 35-1.5

-1

-0.5

0

0.5

1

1.5

2

(b)

C3-C4 y mean y median y std

0 5 10 15 20 25 30 35-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

Frequency Band(Hz)(c)

Cohe

renc

e Ph

ase

P3-P4 y mean y median y std

0 5 10 15 20 25 30 35-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

Frequency Band(Hz)(d)

O1-O2 y mean y median y std

xv

Abstract

Figure 5.3 (a) Coherence phase between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to 35 Hz) for S1. (b) Coherence phase between ECG and EEG (C3-C4). (c) Coherence phase between ECG and EEG (P3-P4). (d)

Coherence phase between ECG and EEG (O1-O2).Table 5.1 Coherence and phase coherence measure parameters for first subject

Signals Parameters Max Min Mean Stddev Median

ECG and EEG

(F p 1−Fp 1)

Coherence 0.99601 0.00032 0.14019 ±0.13377 0.11068

Phase Coherence 1.54025 -1.56089 0.03108 ±0.86714 0.08964

ECG and EEG

(C3−C4)

Coherence 0.99281 0.00068 0.13861 ±0.13516 0.10629

Phase Coherence 1.56003 -1.47487 -0.02502 ±0.92418 0.00E+000

ECG and EEG

(P3−P4)

Coherence 0.99142 0.00010 0.14399 ±0.14635 0.10466

Phase Coherence 1.55448 -1.55937 0.01754 ±0.91504 0.00E+000

ECG and EEG

(O1−O2)

Coherence 0.99663 0.00040 0.15198 ±0.15084 0.11045

Phase Coherence 1.55119 -1.56665 -0.08314 ±0.94767 -0.06879

5.2 Coherence analysis for second subject5.2.1 ECG and EEG signals

0 1000 2000 3000 4000 5000 6000-0.1

0

0.1

(a)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-5

0

5x 10-3

(b)

0 1000 2000 3000 4000 5000 6000-0.1

0

0.1

(c)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-2

0

2

4x 10-3

(d)

0 1000 2000 3000 4000 5000 6000-0.1

0

0.1

(e)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-5

0

5

10x 10-3

(f)

0 1000 2000 3000 4000 5000 6000-0.1

0

0.1

No. of Samples of ECG Signals(g)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-5

0

5x 10-3

No. of Samples of EEG Signals(h)

Figure 5.4 (a) & (b) ECG signal and corresponding EEG (Fp1-Fp2) signal (Each signal is sampled at the sampling rate 500 samples/second and No. of samples taken for each signal is 5006) of the S2. (c) & (d) ECG

xv

Abstract

signal and corresponding EEG (C3-C4) signal. (e) & (f) ECG signal and corresponding EEG (P3-P4) signal. (g) & (h) ECG signal and corresponding EEG (O1-O2) signal.

5.2.2 Coherence between ECG and EEG signals

Coherence between the ECG and corresponding EEG signals acquired from the four

prominent brain regions named as the Frontal(F p1−Fp 2

), Central(C3−C4), Parietal (P3−P4)

and Occipital (O1−O2) is investigated as:

It is shown in figure 5.5(a) the mean of coherence is found to be 0.13950 in the

frequency band (0-35Hz) and the maximum coherence is 0.99819 near the frequency 0.1Hz.

There is one coherence peak is found near the frequency 2.5Hz.

It is shown in figure 5.5(b) the mean of coherence is found to be 0.13569 in the

frequency band (0-35Hz) and the maximum coherence is 0.99569 near the frequency 0.1Hz.

There is one coherence peak is found near the frequency 1.5Hz.

It is shown in figure 5.5(c) the mean of coherence is found to be 0.16404 in the

frequency band (0-35Hz) and the maximum coherence is 0.99829 near the frequency 0.1Hz.

There are another three coherence peaks are found, one is near the frequency 21Hz, second

and third are near the frequency range 5.5Hz to 6Hz.

It is shown in figure 5.5(d) the mean of coherence is found to be 0.16092 in the

frequency band (0-35Hz) and the maximum coherence is 0.99381 near the frequency 0.1Hz.

There is coherence peaks is found, near the frequency 5Hz.

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

(a)

Cohe

renc

e

Fp1-Fp2 y mean y median y std

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

(b)

C3-C4 y mean y median y std

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

Frequency Band(Hz)(c)

Cohe

renc

e

P3-P4 y mean y median y std

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

Frequency Band(Hz)(d)

O1-O2 y mean y median y std

xv

Abstract

Figure 5.5(a) Coherence between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to 35 Hz) for S2. (b) Coherence between ECG and EEG (C3-C4). (c) Coherence between ECG and EEG (P3-P4). (d) Coherence

between ECG and EEG (O1-O2).

5.2.3 Phase coherence between ECG and EEG signals

Phase coherence is the measure of the phase induced by the one signal to another signal at a particular frequency. Here, it is measured in radians. The mean of phase coherence is found to be maximum when it is measured between the ECG signal and corresponding EEG signal acquired from the frontal (F p1

−Fp 2) region.

0 5 10 15 20 25 30 35-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

(a)

Cohe

renc

e Ph

ase

Fp1-Fp2 y mean y median y std

0 5 10 15 20 25 30 35-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

(c)

C3-C4 y mean y median y std

0 5 10 15 20 25 30 35-2

-1.5

-1

-0.5

0

0.5

1

1.5

Frequency Band(Hz)(c)

Cohe

renc

e Ph

ase

P3-P4 y mean y median y std

0 5 10 15 20 25 30 35-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

Frequency Band(Hz)(d)

O1-O2 y mean y median y std

Figure 5.6 (a) Coherence phase between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to 35 Hz) for S2. (b) Coherence phase between ECG and EEG (C3-C4). (c) Coherence phase between ECG and EEG (P3-P4). (d)

Coherence phase between ECG and EEG (O1-O2).Table 5.2 Coherence and phase coherence measure parameters for second subject

Signals Parameters Max Min Mean Stddev Median

ECG and EEG

(F p 1−Fp 1)

Coherence 0.99819 0.00070 0.13950 ±0.13801 0.08964

Phase Coherence 1.53778 -1.53407 -0.17700 ±0.89330 -0.17318

ECG and EEG

(C3−C4)

Coherence 0.99569 0.00036 0.13569 ±0.13322 0.09920

Phase Coherence 1.56504 -1.51907 0.11230 ±0.83720 0.15565

ECG and EEG

(P3−P4)

Coherence 0.99829 0.00186 0.16404 ±0.14793 0.14145

Phase Coherence 1.46363 -1.56290 -0.06682 ±0.90545 -0.04604

ECG and Coherence 0.99381 0.00072 0.16092 ±0.15508 0.11476

xv

Abstract

EEG

(O1−O2)

Phase Coherence 1.55846 -1.56160 -0.03770 ±0.91949 0.00E+000

5.3 Coherence analysis for third subject5.3.1 ECG and EEG signals

0 1000 2000 3000 4000 5000 6000-0.2

0

0.2

(a)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-2

0

2

4x 10-3

(b)

0 1000 2000 3000 4000 5000 6000-0.2

0

0.2

(c)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-2

0

2

4x 10-3

(d)

0 1000 2000 3000 4000 5000 6000-0.2

0

0.2

(e)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-2

0

2

4x 10-3

(f)

0 1000 2000 3000 4000 5000 6000-0.2

0

0.2

No. of Samples of ECG Signals(g)

Ampl

itude

(V)

0 1000 2000 3000 4000 5000 6000-2

0

2

4x 10-3

No. of Samples of EEG Signals(h)

Figure 5.7 (a) & (b) ECG signal and corresponding EEG (Fp1-Fp2) signal (Each signal is sampled at the sampling rate 500 samples/second and No. of samples taken for each signal is 5006) of the S3. (c) & (d) ECG

signal and corresponding EEG (C3-C4) signal. (e) & (f) ECG signal and corresponding EEG (P3-P4) signal. (g) & (h) ECG signal and corresponding EEG (O1-O2) signal.

5.3.2 Coherence between ECG and EEG signals

Coherence between the ECG and corresponding EEG signals acquired from the four

prominent brain regions named as the Frontal(F p1−Fp 2

), Central(C3−C4), Parietal (P3−P4)

and Occipital (O1−O2) is investigated as:

It is shown in figure 5.8(a) the mean of coherence is found to be 0.13443 in the

frequency band (0-35Hz) and the maximum coherence is 0.99429 near the frequency 0.1Hz.

It is shown in figure 5.8(b) the mean of coherence is found to be 0.13662 in the

frequency band (0-35Hz) and the maximum coherence is 0.99420 near the frequency 0.1Hz.

There are another two coherence peaks are found, one is near the frequency range 2Hz and

another near the 12Hz.

xv

Abstract

It is shown in figure 5.8(c) the mean of coherence is found to be 0.14209 in the

frequency band (0-35Hz) and the maximum coherence is 0.99733 near the frequency 0.1Hz.

There is one coherence peak is found near the frequency 32.5Hz.

It is shown in figure 5.8(d) the mean of coherence is found to be 0.15163 in the

frequency band (0-35Hz) and the maximum coherence is 0.99647 near the frequency 0.1Hz.

There are another three coherence peaks are found, one is near the frequency 6.5Hz and

second is near the frequency 22Hz and third is near the frequency 24Hz.

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

(a)

Cohe

renc

e

Fp1-Fp2 y mean y median y std

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

(b)

C3-C4 y mean y median y std

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

Frequency Band(Hz)(c)

Cohe

renc

e

P3-P4 y mean y median y std

0 5 10 15 20 25 30 350

0.2

0.4

0.6

0.8

1

Frequency Band(Hz)(d)

O1-O2 y mean y median y std

Figure 5.8 (a) Coherence between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to 35 Hz) for S3. (b) Coherence between ECG and EEG (C3-C4). (c) Coherence between ECG and EEG (P3-P4). (d) Coherence

between ECG and EEG (O1-O2).

5.3.3 Phase coherence between ECG and EEG signals

Phase coherence is the measure of the phase induced by the one signal to another signal at a particular frequency. Here, it is measured in radians. The mean of phase coherence is found to be maximum when it is measured between the ECG signal and corresponding EEG signal acquired from the frontal (F p1

−Fp 2) region.

xv

Abstract

0 5 10 15 20 25 30 35-2

-1.5

-1

-0.5

0

0.5

1

1.5

(a)

Cohe

renc

e Ph

ase

Fp1-Fp2 y mean y median y std

0 5 10 15 20 25 30 35-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

(b)

C3-C4 y mean y median y std

0 5 10 15 20 25 30 35-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

Frequency Band(Hz)(c)

Cohe

renc

e Ph

ase

P3-P4 y mean y median y std

0 5 10 15 20 25 30 35-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

Frequency Band(Hz)(d)

O1-O2 y std y median y mean

Figure 5.9 (a) Coherence phase between ECG and EEG (Fp1-Fp2) in Frequency Band (0 to 35 Hz) for S3. (b) Coherence phase between ECG and EEG (C3-C4). (c) Coherence phase between ECG and EEG (P3-P4). (d)

Coherence phase between ECG and EEG (O1-O2).

Table 5.3 Coherence and phase coherence measure parameters for third subject

Signals Parameters Max Min Mean Stddev Median

ECG and EEG

(F p1−Fp1)

Coherence 0.99429 0.00085 0.13443 ±0.12882 0.09564

Phase Coherence 1.49501 -1.54161 0.10652 ±0.83149 0.13572

ECG and EEG

(C3−C4)

Coherence 0.99420 0.00537 0.13662 ±0.13500 0.10517

Phase Coherence 1.56759 -1.55442 -0.04030 ±0.91074 -0.11559

ECG and EEG

(P3−P4)

Coherence 0.99733 0.00019 0.14209 ±0.14504 0.09462

Phase Coherence 1.54749 -1.55333 0.01010 ±0.88324 -0.02697

ECG and EEG

(O1−O2)

Coherence 0.99647 0.00211 0.15163 ±0.14304 0.11715

Phase Coherence 1.55132 -1.54963 -0.10359 ±0.92867 -0.19947

5.4 Combine coherence analysis for all three subjects

In the figure 5.10(a), (b) and (c) the maximum coherence upper quartile is in the

coherence between ECG and corresponding EEG signal (O1−O2) from all three subjects. The

number of coherence peaks (coherence value greater than ≥ 0.5) is found to be more in the

coherence between ECG and EEG signal (P3−P4) in all three subjects. It reflects that the

xv

Abstract

heart signal has relatively more functional association or relationship to the corresponding

brain signal (P3−P4) at a particular frequency band (0 to 35Hz).

0

0.2

0.4

0.6

0.8

1

1 2 3 4(a)

Cohe

renc

e

0

0.2

0.4

0.6

0.8

1

1 2 3 4(b)

Cohe

renc

e

0

0.2

0.4

0.6

0.8

1

1 2 3 4Boxplots for ECG and EEG Signals Coherence from

four Brain Regions(c)

Cohe

renc

e

Figure 5.10 (a) Coherence between the ECG signals corresponding to the EEG signals 1-EEG (Fp1-Fp2), 2-EEG (C3-C4), 3-EEG (P3-P4), 4-EEG (O1-O2) of the First Subject (S1). (b) Coherence between the ECG

signals corresponding to the EEG signals of the Second Subject (S2). (c) Coherence between the ECG signals corresponding to the EEG signals of the Third Subject (S3).

The figure 5.11 (a), (b) and (c) show the number of coherence values fall in the

different coherence limits for first, second and third subjects respectively. Here, in figure 5.11

(a), (b) and (c) the one coherence value greater than 0.9 is fall in the coherence limit (0.9 to 1)

xv

Abstract

of all subjects different four types of coherence calculations between the ECG signal and the

corresponding EEG signals. No coherence value is found in the coherence limits (0.7 to 0.8)

and (0.8 to 0.9).

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

20

40

60

80

(a)

No.

of C

oher

ence

val

ues

Coherence b/w ECG & EEG(Fp1-Fp2) of S1Coherence b/w ECG & EEG(C3-C4) of S1Coherence b/w ECG & EEG(P3-P4) of S1Coherence b/w ECG & EEG(O1-O2) of S1

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

20

40

60

80

(b)

No. o

f Coh

eren

ce v

alue

s

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

20

40

60

80

Coherence Range(c)

No. o

f Coh

eren

ce v

alue

s

Figure 5.11 (a) Coherence between the ECG signals corresponding to the EEG signals 1-EEG (Fp1-Fp2), 2-EEG (C3-C4), 3-EEG (P3-P4), 4-EEG (O1-O2) of the First Subject (S1). (b) Coherence between the ECG

signals corresponding to the EEG signals of the Second Subject (S2). (c) Coherence between the ECG signals corresponding to the EEG signals of the Third Subject (S3).

5.5 Combine phase coherence analysis for all three subjects

The figure 5.12(a), (b) and (c) provide the information how the coherence phase mean

vary in the three subjects among the different coherence investigated. The upper phase

coherence quartile (75 percentile of all phase coherence) is found to be maximum in the all

three subjects. The mean of phase coherence varies randomly in the three subjects’ coherence

investigated.

xv

Abstract

-1

0

1

1 2 3 4(a)

Cohe

renc

e Ph

ase

-1

0

1

1 2 3 4(b)

Cohe

renc

e Ph

ase

-1

0

1

1 2 3 4(c)

Cohe

renc

e Ph

ase

Figure 5.12 (a) Coherence phase between the ECG signals corresponding to the EEG signals 1-EEG (Fp1-Fp2), 2-EEG (C3-C4), 3-EEG (P3-P4), 4-EEG (O1-O2) of the First Subject (S1). (b) Coherence phase between the ECG signals corresponding to the EEG signals of the Second Subject (S2). (c) Coherence phase between the

ECG signals corresponding to the EEG signals of the Third Subject (S3).

xv

Abstract

Chapter 6

CONCLUSION AND FUTURE SCOPE

Conclusion:

This research is divided into two phases. In first phase we have evaluated the

coherence and phase coherence between the ECG signals and the EEG signals acquired from

the temporal region of the brain at the different respiratory rates. Here we acquired these

signals at the 500 samples/second (Sampling Rate). Number of sample points (samples) for

each signal is 5006. Data acquisition is done using Biopac Inc. MP100 and the software tool

AcqKnowledge3.9.0 by setting the corresponding calculate channel for getting respiratory

rates simultaneously with the ECG and EEG signals. We have found that the maximum

coherence mean is at the no airflow or zero respiratory rates (0-4 Breaths/min) between the

ECG and EEG signal. We have found that the minimum coherence mean is at the high

airflow or high respiratory rates (16-20 Breaths/min) between the signals. The peak of

coherence is found more than 0.5 at the respiratory frequency 0.1Hz in the coherence

between ECG and EEG signals at the zero respiratory rates and the normal respiratory rates.

In second phase, we acquired the data from 50 subjects and analyse the coherence. In

thesis we include the three subject’s data for finding the coherence and phase coherence

between ECG signals and the corresponding EEG signals acquired from the different brain

regions. The different brain regions are the Frontal(F p1−Fp 2

), Central(C3−C4), Parietal

(P3−P4) and occipital(O1−O2) from which the EEG signal acquired. The coherence and

phase coherence for each subject and each set of signals is evaluated using magnitude

squared coherence function.

For first subject, the maximum, mean of coherence is in the coherence between the

ECG and the EEG signal acquired from the (O1−O2) region. The maximum numbers of

coherence peaks are in the coherence between ECG and EEG signal acquired from (P3−P4)

region. The maximum, mean of phase coherence is in the phase coherence between the ECG

and EEG signal acquired from the frontal brain region(F p1−Fp 2

).

For second subject, the maximum, mean of coherence is in the coherence between the

ECG and the EEG signal acquired from the (O1−O2) region. The maximum numbers of

coherence peaks are in the coherence between ECG and EEG signal acquired from (P3−P4)

xv

Abstract

region. The maximum, mean of phase coherence is in the phase coherence between the ECG

and EEG signal acquired from the frontal brain region(F p1−Fp 2

).

Similarly for the third subject the above coherence measure is analysed.

In conclusion, the results of the investigation of interactions between spectral power

bands of ECG and EEG signals may contribute to a better understanding of physiological

mechanisms underlying the interactions between brain and heart during normal breathing but

need to be further investigated in a larger and more diverse sample of normal healthy

subjects, children, and old subjects at the unipolar EEG signals acquired from the particular

montages as F p1, Fp 2

, C3 , C z ,C 4 , P3 , P z , P4 ,O1 , O2etc . during different subject conditions.

Future Scope:

All the organs of the human body have some synchronism, association and correlation

to each other. In this work we investigate the coherence and phase coherence between the

ECG and EEG; means the association between the human brain and heart. Communication

between the heart and brain is actually a dynamic, ongoing, two-way dialogue, with each

organ continuously influencing the other's function. Research has shown that the heart

communicates to the brain in four major ways: neurologically (through the transmission of

nerve impulses), biochemically (via hormones and neurotransmitters), biophysically (through

pressure waves) and energetically (through electromagnetic field interactions).

Communication along all these conduits significantly affects the brain's activity. The

magnitude squared coherence between the two physiological signals provides the valuable

association between the corresponding physiological organs. We also analysed the phase

induced by the one physiological signal to another physiological signal. It is quantified, both

coherence and the phase coherence.

Now, we have the coherence spectrum, and we have analysed the numbers of

coherence peaks in the specified frequency band. The coherence peak reflects that the one

physiological signal is synchronised with another physiological signal at the particular

frequency. The phase coherence spectrum reflects that the one physiological signal induced,

how much phase (lead or lag corresponding to the positive phase and the negative phase) to

the another physiological signal.

REFERENCES

xv

Abstract

[1].Yue-Der Lin, Wei-Ting Liu, Ching-Che Tsai, and Wen-Hsiu Chen, “Coherence

analysis between respiration and PPG signal by bivariate AR model” World Academy

of Science, Engineering and Technology,2009,pp.847-852.

[2].Sanqing Hu, Matt Stead, Qionghai Dai, and Gregory A. Worrell, “On the recording

reference contribution to EEG correlation, phase synchrony, and coherence”, “IEEE

Transactions on Systems, Man, and Cybernetics—Part B: Cybernetics”, pp.1-11, 2010.

[3].AH Khandoker, CK Karmakar, M Palaniswami The University of Melbourne, Victoria,

Australia, “Interaction between sleep EEG and ECG signals during and after

obstructive sleep apnea events with or without arousals”. “Computers in Cardiology”,

2008, pp.685-688.

[4].Naveed R. Butt and Andreas Jakobsson, “Coherence spectrum estimation from

nonuniformly sampled sequences”, IEEE Signal Processing Letters, vol. 17, no. 4,

pp.339-342, April 2010.

[5].Andreas Jakobsson, Stephen R. Alty and Jacob Benesty, “Estimating and time-updating

the 2-D coherence spectrum”, IEEE Transactions on Signal Processing, vol. 55, no. 5,

pp.2350-2354, May 2007.

[6].David M. Simpson, Daniel A. Botero Rosas and Antonio Fernando C. Infantosi,

“Estimation of coherence between blood flow and spontaneous EEG activity in

neonates”, IEEE Transactions on Biomedical Engineering, vol. 52, no. 5, pp.852-858,

May 2005.

[7].Hernando Ombao and Sébastien Van Bellegem, “Evolutionary coherence of

nonstationary signals”, IEEE Transactions on Signal Processing, vol. 56, no. 6, pp.

2259-2266, June 2008.

[8].Luca Faes, Lamberto Widesott, Maurizio Del Greco, Renzo Antolini, and

Giandomenico Nollo, “Causal cross-spectral analysis of heart rate and blood pressure

variability for describing the impairment of the cardiovascular control in neurally

mediated syncope”, IEEE Transactions on Biomedical Engineering, vol. 53, no. 1,

pp.65-73, January 2006.

[9].Qi Yang, Vlodek Siemionow, Wanxiang Yao, Vinod Sahgal, and Guang H. Yue,

“Single-Trial EEG-EMG Coherence analysis reveals muscle fatigue-related progressive

xv

Abstract

alterations in corticomuscular coupling”, IEEE Transactions on Neural Systems and

Rehabilitation Engineering, vol. 18, no. 2, pp.97-106, April 2010.

[10]. David Ramírez, Javier Vía, Ignacio Santamaría and Louis L. Scharf, “Detection of

spatially correlated gaussian time series”, IEEE Transactions on Signal Processing, vol.

58, no. 10, pp.5006 -5015, October 2010.

[11]. Laura Astolfi, Febo Cincotti, Donatella Mattia, M. G. Marciani, Luis A. Baccalà,

Fabrizio de Vico Fallani, Serenella Salinari, Mauro Ursino, Melissa Zavaglia, and

Fabio Babiloni, “Assessing cortical functional connectivity by partial directed

coherence: simulations and application to real data”, IEEE Transactions on Biomedical

Engineering, vol. 53, no. 9, pp.1802-1812, September 2006.

[12]. Xiao Hu, Valeriy Nenov, Paul Vespa, and Marvin Bergsneider, “Characterization of

interdependency between intracranial pressure and heart variability signals: A causal

spectral measure and a generalized synchronization measure”, IEEE Trans Biomed

Eng.; vol.54, no.8, pp.1407-1417, August 2007.

[13]. Kurt Barbé, Rik Pintelon, and Johan Schoukens, “Welch method revisited:

Nonparametric power spectrum estimation via circular overlap”, IEEE Transactions on

Signal Processing, vol. 58, no. 2, pp.553-565, February 2010.

[14]. G. L. MOHNKERN, “Maximum likelihood estimation of magnitude-squared

coherence”, IEEE Transactions on Acoustics, Speech and Signal Processing. vo1. 36.

No. I, pp.130-132, January 1988.

[15]. Bcla Suki and Kenneth R. Lutchen, “Pseudorandom signals to estimate apparent

transfer and coherence functions of nonlinear systems: Applications to respiratory

mechanics”, IEEE Transactions on Biomedical Engineering. vol. 39, no. I I, pp.1142 -

1151, November 1992.

[16]. Vinod Chandran, Steve Elgar, and Barry Vanhoff, “Statistics of tricoherence”, lEEE

Transactions on Signal Processing, vol. 42, no. 12, pp.3330-3340, December 1994.

[17]. Nitish V. Thakor, Xuan Kong, and Daniel F. Hanley, “Nonlinear changes in brain’s

response in the event of injury as detected by adaptive coherence estimation of evoked

potentials”, IEEE Transactions on Biomedical Engineering, vol. 42, no. 1, pp.42-51,

January 1995.

xv

Abstract

[18]. Axel Clausen and Douglas Cochran, “An invariance property of the generalized

coherence estimate”, IEEE Transactions on Signal Processing, vol. 45, no. 4, pp.1065-

1167, April 1997.

[19]. Ramesh Srinivasan, Paul L. Nunez, and Richard B. Silberstein, “Spatial filtering and

neocortical dynamics: Estimates of EEG coherence”, IEEE Transactions on Biomedical

Engineering, vol. 45, no. 7, pp.814-826, July 1998.

[20]. Yan Xu, Simon Haykin, and Ronald J. Racine, “Multiple window time-frequency

distribution and coherence of EEG using slepian sequences and hermite functions”,

IEEE Transactions on Biomedical Engineering, vol. 46, no. 7, pp.861-866, July 1999.

[21]. Olivier Besson, Petre Stoica, and Alex B. Gershman, “Simple and accurate direction

of arrival estimator in the case of imperfect spatial coherence”, IEEE Transactions on

Signal Processing, vol. 49, no. 4, pp.730-737, April 2001.

[22]. Antonio Mauricio F. L. Miranda de Sá, Antonio Fernando C. Infantosi, and David M.

Simpson, “A statistical technique for measuring synchronism between cortical regions

in the EEG during rhythmic stimulation”, IEEE Transactions on Biomedical

Engineering, vol. 48, no. 10, pp.1211-1215, October 2001.

[23]. J. S. Paul, A. R. Luft, D. F. Hanley, and N. V. Thakor, “Coherence-weighted wiener

filtering of somatosensory evoked potentials”, IEEE Transactions on Biomedical

Engineering, vol. 48, no. 12, pp.1483-1488, December 2001.

[24]. Kamal Yacoub, “Relationship between multiple and partial coherence functions”,

IEEE Transactions on Information Theory, vol.IT-16, pp.668-672, November1979.

[25]. Vernon A. Benignus, “Estimation of the coherence spectrum and its confidence

interval using the Fast Fourier Transform”, IEEE Transactions on Audio and

Electroacoustics, vol.AU-17, no.2, pp.145-150, June 1969.

[26]. AK Kokonozi, EM Michail, IC Chouvarda, NM Maglaveras, “A study of heart rate

and brain system complexity and their interaction in sleep-deprived subjects” ISSN

0276−6574 969 Computers in Cardiology 2008, pp.969−971.

[27]. Gleb V. Tcheslavski and A. A. (Louis) Beex, “Phase synchrony and coherence

analyses of EEG as tools to discriminate between children with and without attention

xv

Abstract

deficit disorder”, Biomedical Signal Processing and Control 1 pp.151–161, September,

2006.

[28]. Aijun He, Xiaodong Yang, Xi Yang, Xinbao Ning, “Phase Synchronization in Sleep

Electroencephalogram”, “IEEE/ICME International Conference on Complex Medical

Engineering”, 2007, pp.1421-1424.

[29]. Aneta Stefanovska, “Complex But Not Complicated Cardiovascular and Brain

Interactions”, IEEE Engineering in Medicine and Biology Magazine, Nov/Dec 2007,

pp.25-29.

[30]. Tilmann Kluge and Manfred Hartmann, “Phase Coherent Detection of Steady-State

Evoked Potentials: Experimental Results and Application to Brain-Computer

Interfaces”, Proceedings of the 3rd International IEEE EMBS Conference on Neural

Engineering, May 2-5, 2007, pp. 425-429.

[31]. Disha Gupta, and Christopher J. James, “Narrowband vs. Broadband Phase

Synchronization Analysis Applied to Independent Components of Ictal and Interictal

EEG”, Proceedings of the 29th Annual International Conference of the IEEE EMBS

Cite International, August 23-26, 2007, pp. 3864-3867.

[32]. Xiaofeng Liu, Huan Qi, SupinWang, MingxiWan, “Wavelet-based estimation of EEG

coherence during Chinese Stroop task”, Computers in Biology and Medicine 36,

pp.1303-1315, Aug 2006.

[33]. Barry R. Greene, Geraldine B. Boylan, Richard B. Reilly, Philip de Chazal and Sean

Connolly, “Combination of EEG and ECG for improved automatic neonatal seizure

detection”, Clinical Neurophysiology 118, pp.1348-1359. March 2007.

[34]. MC Mantaras, MO Mendez, O Villiantieri, N Montano, V Patruno, AM Bianchi, S

Cerutti, “Non-parametric and Parametric Time-Frequency Analysis of Heart Rate

Variability during Arousals from Sleep”, Computers in Cardiology no.33, 2006,

pp.745-748.

[35]. Ahsan H. Khandoker, Chandan K. Karmakar, Marimuthu Palaniswami, “Analysis of

coherence between sleep EEG and ECG signals during and after obstructive sleep

apnea events”, 30th Annual International IEEE EMBS Conference, Vancouver, British

Columbia, Canada, August 20-24, 2008, pp. 3876-3879.

[36]. Mahmoud El-Gohary, James McNames, Tim Ellis and Brahm Goldstein, “Time

Delay and Causality in Biological Systems Using Whitened Cross-correlation

xv

Abstract

Analysis” Proceedings of the 28th IEEE EMBS Annual International Conference New

York City, USA, Aug 30-Sept 3, 2006, pp. 6169-6172.

[37]. Catarina S. Nunes, Teresa Mendonc¸a, Susana Br´as, David A. Ferreira, and Pedro

Amorim, “Modeling Anesthetic Drugs’ Pharmacodynamic Interaction on the Bispectral

Index of the EEG: the Influence of Heart Rate”, Proceedings of the 29th Annual

International Conference of the IEEE EMBS Cite International, Lyon, France August

23-26, 2007, pp. 6479-6482.

[38]. Charles S. Lessard, “Signal Processing Of Random Physiological Signals”, Coherence Function from Spectral Analysis, , First Edition, 2006 by Morgan & Claypool, pp.203-212.

[39]. Emmanuel C. Iteachar and Barrie W. Jervis, “Digital signal Processing: A Practical Approach”, Correlation and Convolution, second edition, Pearson education, pp.242-249.

APPENDIX

xv

Abstract

(A):

A.1. LEAD110 Series — Electrode leads

The LEAD110 Series, for use with disposable and other snap connector electrodes, are pinch

leads for easy connection between the EL500-series snap electrodes and any BIOPAC

biopotential amplifier or the GND terminal on the back of the UIM100C. Leads terminate in

standard 2 mm pin plug and connect to BIOPAC modules or to a Modular Extension Cable

(MEC series).

Table A.1: Lead Type Length Usage Note

Sr.

No.

Leads Name Type Length Procedure of Using Leads

1. LEAD110 Unshielded 1 m Works best as a ground electrode

2. LEAD110A Unshielded 3 m Works best with ground or reference electrodes

3. LEAD110S-

R

Shielded 1 m Use with recording electrodes for minimal noise

interference. The white lead plug is for the

electrode contact; the black lead pin plug is for

the lead shield.

4. LEAD110S-

W

Shielded 1 m Use with recording electrodes for minimal noise

interference. The white lead plug is for the

electrode contact; the black lead pin plug is for

the lead shield.

IMPORTANT SAFETY NOTES

1. MEC series cables are not to be used on humans when they are undergoing

electrosurgery or defibrillation. In fact, no BIOPAC equipment should be connected

to human subjects during the course of defibrillation or electrosurgery.

2. When MEC series cables are used, be careful to preserve the isolation of MP system

during defibrillation. No external lab equipment should be connected directly to the

UIM100C, IPS100C or any included amplifier module. To preserve MP system

isolation, all connections of this type should be made using INISO or OUTISO with

the HLT100C. To verify that the isolation of the recording system is intact, use a

multimeter to measure resistance from subject ground (on biopotential amplifier) to

mains ground; there should be no DC conductivity.

xv

Abstract

3. Do not connect the electrode leads attached to the MEC series cables directly to

defibrillator paddles. When using MEC cables, electrode leads should be connected to

the subject directly and not via the defibrillator paddles

A.1.a Common Extensions

MEC100C 100C-series Transducer amplifiers to Touchproof inputs

MEC110C 100C-series Biopotential amplifiers to Touchproof inputs

MEC111C 100C-series Biopotential amplifiers to Touchproof inputs—Protected

A.1.b Less Common Extensions

MEC100 DA100C or 100B-series Biopotential or Transducer amplifiers to 2mm socket

inputs

MEC101 100B-Series Biopotential amplifiers to 2mm socket inputs – Protected

MEC110 100B-series Biopotential or Transducer amplifiers to Touchproof inputs

MEC111 100B-series Biopotential amplifiers to Touchproof inputs—Protected

A.2 TSD201The TSD201 is a strain gauge transducer designed to measure respiratory-induced

changes in thoracic or abdominal circumference, and can therefore be used to record

respiratory effort. The TSD201 is essentially a resistive transducer and responds in a linear

fashion to changes in elongation through its length, with resistance increasing as length

increases.

The transducer is ideal for a variety of applications because it presents minimal

resistance to movement and is extremely unobtrusive. Due to its unique construction, the

TSD201 can measure extremely slow respiration patterns with no loss in signal amplitude

while maintaining excellent linearity and minimal hysteresis.

The TSD201 plugs directly into the RSP100C amplifier module (page 114). It

includes a fully adjustable nylon strap to accommodate a large range of circumferences (9 cm

to 130 cm). To attach the nylon belt to the respiration transducer, thread the nylon strap

through the corresponding slots so the strap clamps into place when tightened. Place the

transducer around the body at the level of maximum respiratory expansion. This location will

vary from the erect to supine positions (generally about 5 cm below the armpits).

Correct tension adjustment of the respiration transducer is important. For best

sensitivity, the transducer must be just slightly tight at the point of minimum circumference

(maximum expiration). To obtain proper tension, stretch the belt around the body and have

xv

Abstract

the subject exhale. At maximum expiration, adjust the nylon strap so there is slight tension to

hold the strap around the chest.

The transducer has three 2 mm pin plugs to connect to the amplifier. Insert the two

blue lead transducer pin plugs into the two RSP100C inputs labelled XDCR. Either blue lead

can be connected to either XDCR input. Insert the single black transducer lead into the GND

input of the RSP100C. The respiration transducer is ready for measurement.

A.2.a TSD201 Calibration

The TSD201 does not require calibration.

Table A.2: TSD201 Specifications

Sr.No

.

Parameters Sensor Parameter Range Values

1. True DC Response Yes

2. Variable Resistance Output 5-125 KΩ (increases as length increases)

3. Circumference Range 15 cm x 150 cm (can be increased with a longer strap)

4. Attachment Velcro® strap (adjustable length)

5. Sterilizable Yes

6. Sensor Weight 18 gs

7. Sensor Dimensions 66 mm (long), 40 mm (wide), 15 mm (thick)

8. Cable Length 3 m

CBLCALC Calibration Cable for 100C-series Biopotential Amplifiers

CBLCAL Calibration Cable 100-B series Biopotential Amplifiers

Use CBLCAL/C to verify the calibration of the any of the Biopotential amplifiers.

The cable (1.8m) connects between the amplifier input and the UIM100C D/A output 0 or 1.

To verify the amplifier’s frequency response and gain settings, create a stimulus signal using

AcqKnowledge and monitor the output of the amplifier connected to the Calibration Cable.

The Calibration Cable incorporates a precision 1/1000 signal attenuator.

Amplifier specification tests are performed at the factory before shipping, but a

Calibration Cable can ensure users peace of mind by permitting precise frequency response

and gain calibrations for exact measurements.

CBLCAL/C Calibration

A.2.b Hardware Setup

1. Connect the MP150/100, UIM100C and biopotential amplifiers as normal.

xv

Abstract

2. Connect the CBLCAL/C between the selected amplifier and the UIM100C, inserting

the single 3.5mm plug into the Analog Output “0” port on the UIM100C.

3. Connect the end containing several 2mm pins into the corresponding holes on the face

of the biopotential amplifier.

4. Select a Gain setting of 1,000 for DA, ECG, EGG, EMG, and EOG, or 5,000 for EEG

and ERS.

5. Turn all filters to the desired position.

6. Select an appropriate channel on the top of the amplifier being tested (usually channel

one, as this is the default setup in the software).

A.2.c Software Setup

1. Under Channel Setup, insure that the default is set to analog channel one (A1).

2. Under Acquisition Setup

a) Choose a sampling rate of 2000Hz (or higher).

b) Choose an acquisition period of at least 5 seconds.

c) Choose Record Last mode.

3. Under Stimulator Setup

a) Select the sine wave for the shape of the output signal.

b) Set the “Seg. #1 Width’ to zero. This means that the signal will be transmitted

continuously starting at time-point zero.

c) Set “Seg. #2 Width” to 1,000 msec (one second). This is the length of the output

signal.

d) Select “Analog Output: 0.”

e) Select “Output continuously.”

f) The most important settings are the signal magnitude and frequency. Set the

magnitude to 5 Volts (i.e., 10V p-p) if the module gain setting is 1,000. If the

lowest module gain setting available is 5,000, choose 1 Volt.

g) Set the frequency to 10Hz to check the gain calibration (on a sinusoidal signal,

this setting is appropriate for all biopotential amplifiers).

A.2.d Calibration Procedure

AcqKnowledge is now set-up to check for the proper calibration of biopotential amplifiers.

1. Start the acquisition. Theoretically, since you are in record last mode and are

outputting a signal continuously, AcqKnowledge could acquire data forever.

2. Stop the acquisition when the waveform has stabilized.

3. Use the “I-beam” cursor to select the latter part of the record.

xv

Abstract

4. Perform all your calibration measurements on the latter part of the collected record.

(A) Internet Resources Used :

(a) Database Website Names: www.physio.net

www.qdheart.com

http://www.bsignetics.com/databases.htm

(b) Multimedia Website Names: http://coursewareobjects.elsevier.com/objects/hao/anim/13-010ap.htm

http://www.nhlbi.nih.gov/health/dci/Diseases/hhw/hhw_pumping.html

http://www.blaufuss.org/tutonline.html#

http://www.cardionics.com/video/classroomstudy/index.html

http://library.med.utah.edu/kw/pharm/hyper_heart1.html

http://sprojects.mmi.mcgill.ca/mvs/WHAT.HTM

http://sprojects.mmi.mcgill.ca/mvs/SHOCK/HRTSPLIT.HTM

(c) Tutorials Website Names http://www.bsignetics.com/case_studies_productPM.htm

www.mathworks.com

www.biopac.com

www.techonline.com

www.newwaveinstruments.com

http://easycalculation.com/statistics/learn-correlation.php

http://stattrek.com/AP-Statistics-1/Correlation.aspx

http://en.wikipedia.org/wiki/

http://gmrt.ncra.tifr.res.in/gmrt_hpage/Users/doc/WEBLF/LFRA/node71.html

(B) Commercial Manufacturers:

BIOPAC System Inc.

xv