A NOVEL OPTIC DISC AND OPTIC CUP SEGMENTATION

TECHNIQUE TO DIAGNOSE GLAUCOMA USING DEEP

LEARNING ALGORITHMS
A PROJECT REPORT SUBMITTED IN FULFILMENT OF THE REQUIREMENTS FOR
THE AWARD OF THE DEGREE OF
BACHELOR OF TECHNOLOGY
IN
COMPUTER SCIENCE AND ENGINEERING

By

M. N. L. Soumya (18JG1A0566) N. Virija Soujanya(18JG1A0570)

I. Varalakshmi (19JG5A0503) N. Srilakshmi (18JG1A0565)

Under the guidance of

Dr. M. Bhanu Sridhar

Associate Professor
Department of Computer Science and Engineering

GAYATRI VIDYA PARISHAD COLLEGE OF ENGINEERING FOR WOMEN

(Affiliated to Jawaharlal Nehru Technological University, Kakinada)
2018-2022
 GAYATRI VIDYA PARISHAD COLLEGE OF ENGINNERING
FOR WOMEN
Department of Computer Science and Engineering

CERTIFICATE

This is to certify that the project report entitled “A Novel Optic Disc and Optic Cup Segmentation
Technique To Diagnose Glaucoma Using Deep Learning” is a bonafide work of following IV/IV
B.Tech students in the Department of Computer Science & Engineering of JNT University, Kakinada
during the academic year 2021-2022, in partial fulfillment of the requirement for the award of the
degree of Bachelor of Technology of this university.

M. N. L. Soumya (18JG1A0566) N. Virija Soujanya (18JG1A0570)

I. Varalakshmi (19JG5A0503) N. Srilakshmi (18JG1A0565)

PROJECT GUIDE HEAD OF THE DEPARTMENT

Dr. M. Bhanu Sridhar Dr. P. V. S Lakshmi Jagadamba
Associate Professor Professor
Department of CSE Department of CSE

EXTERNAL EXAMINER
 ACKNOWLEDGEMENT

The satisfaction that accompanies the successful completion of any task would be
incomplete without the mention of people who made it possible and whose constant
guidance and encouragement crown all the efforts with success.
We feel elated to extend our sincere gratitude to Dr. M. Bhanu Sridhar, Associate
Professor for encouragement all the way during analysis of the project. His annotations,
insinuations and criticisms are the key behind the successful completion of the thesis and
for providing us all the required facilities.
We express our deep sense of gratitude and thanks to Prof. Dr. P. V. S. Lakshmi
Jagadamba, Professor and Head of the Department of CSE for her guidance and for
expressing her valuable and graceful opinions in the project for its development and for
providing the lab sessions and extra hours to complete the project.
We would like to take this opportunity to express our profound sense of gratitude to
the revered Principal, Dr. R. K. Goswami for allowing us to utilize the college resources
there by facilitating the successful completion of our thesis.
We would like to take this opportunity to express our profound sense of gratitude to
the Vice Principal, Prof. Dr. G. Sudheer for allowing us to utilize the college resources
there by facilitating the successful completion of our thesis and not but the least we also
thankful to both teaching and non-teaching faculty of the Department of Computer
Science and Engineering, for giving valuable suggestions for our project.
We also take this opportunity to express our heartfelt credit to the teaching and non-
teaching staff of the department, for their perspective comments and suggestions.
 TABLE OF CONTENTS

TOPICS PAGE
NO.

Abstract i

List of Figures ii

List of Screens iii

List of Acronyms iv

1. INTRODUCTION 1
1.1 Motivation 1
1.2 Problem definition 1
1.3 Objective of Project 1
1.4 Limitations of Project 2
1.5 Organization of Documentation 2

2. LITERATURE SURVEY 3
2.1 Introduction 3
2.1.1 Features Extraction Technique 3
2.1.2 RNN algorithm 4
2.1.3 LSTM algorithm 4
2.1.4 CNN algorithm 4
2.2 Existing System 4
2.3 Disadvantages of Existing system 5
2.4 Proposed System 6
2.5 Conclusion 6

3. REQUIREMENT ANALYSIS 7
3.1 Introduction 7
3.2 Software Requirement Specification 7
3.2.1 User requirements 8
3.2.2 Hardware requirements 9
3.2.3 Software requirements 9
3.3 Content diagram of Project 9
3.4 Flowcharts 10
3.5 Conclusion 11

4. DESIGN 12
 4.1 Introduction 12
4.1.1 UML 13
4.1.2 Goals of UML 14
4.1.3 Basic Building Blocks of UML 14
4.1.4 Components of the UML 15
4.2 UML diagrams 15
4.2.1 Use-Case Diagram 16
4.2.2 Class Diagram 16
4.2.3 Sequence Diagram 17
4.2.4 Collaboration Diagram 18
4.3 Conclusion 19

5. IMPLEMENTATION & RESULTS 20

5.1 Introduction 20
5.2 Anaconda 20
5.2.1 Anaconda Navigator Installations 20
5.3 Software Design 24
5.3.1 Frontend 24
5.3.2 Backend 36
5.4 Output Screens 59

6. TESTING & VALIDATION 69

6.1 Introduction 69
6.1.1 Scope 69
6.1.2 Defects and failures 70
6.1.3 Compatibility 70
6.1.4 Input combinations and preconditions 71
6.1.5 Static vs dynamic testing 71
6.1.6 Types of Testing 71
6.1.7 Software verification and validation 72
6.2 Design of test cases and scenarios 73
6.3 Validation 73
6.4 Conclusion 73

7. CONCLUSION 74

8. REFERENCES 75
 ABSTRACT
Glaucoma, the so-called silent thief of sight, is the leading cause of unpreventable and
irreversible blindness throughout the world. It occurs when circulation of an eye fluid is
irregular and results in increase of intraocular pressure (IOP) in aqueous humor of the human
eye. The increase of IOP leads to the damaging of the optic nerve of an eye and ends up in a
situation of complete or partial vision loss. There are no symptoms of glaucoma which are
clearly visible at an early stage and the only way to detect glaucoma at an initial stage is by
examining the structure of eye so that the internal changes raised can be identified. This
project mainly concentrates on detecting glaucoma at an early stage using Optic Disc (OD)
and Cup and Disc Ratio (CDR) from the fundus images of the eye through extracting the
optic disk and cup disc images. This data is used to calculate their respective areas to find the
ratio of cup area and disk area. Accordingly, the work also assesses the level of glaucoma in
patients, its extent, and if there is any chance of being a victim of glaucoma in near future.
Using Deep Learning algorithms, the proposed model is trained, the necessary cup and disk
images are extracted, and the remaining portion is cropped by applying pre- processing
techniques. The goal of this project is to apply the work on real-time patients, diagnose their
symptoms at an early stage and save their vision.

i
 LIST OF FIGURES

S.No. Figure No. Figure Name Page No.

1. Fig.1 Flow Graph of Glaucoma Detection 3

2. Fig.2 Existing Method 5
3. Fig.3 Content Diagram 9
4. Fig.4 Flowchart 10
5. Fig.5 System Diagram 13
6. Fig.6 Use Case Diagram 16
7. Fig.7 Class Diagram 17
8. Fig.8 Sequence Diagram 18
9. Fig.9 Collaboration Diagram 19
10. Fig.10 Retinal Fundus Image 59
11. Fig 11 Blood Vessels 60
12. Fig.12 Blood Vessel Representation 61
13. Fig.13 Preprocessed Green Channel 62
14. Fig.14 Smoothed Histogram Green Channel 63
15. Fig.15 Preprocessed Red Channel 64
16. Fig.16 Smoothed Histogram Red Channel 65
17. Fig.17 Optic Disc of Retinal Image 66
18. Fig.18 Optic Cup of Retinal Image 67
19. Fig.19 Results for RNN Model 68

ii
 LIST OF SCREENS

S.No. Screen No. Screen Name Page No.

1. Screen 1 Launch Anaconda Navigator 21

2. Screen 2 Install Jupyter Notebook 21

3. Screen 3 Installation Screen of Jupyter Notebook 22

4. Screen 4 Loading Packages 22

5. Screen 5 Final Screen after Installation 23

6. Screen 6 Launching Jupyter 23

7. Screen 7 Homepage of Jupyter 24

8. Screen 8 Frontend Output Screen 68

iii
 LIST OF ACRONYMS

CDR Cup to Disc Ratio

OD Optic Disc
OC Optic Cup
OCT Optical Coherence Tomography
RNN Recurrent Neural Network
UML Unified Modeling Language
IOP Intra-Ocular Pressure
LSTM Long short-term Memory
CNN Convolutional Neural Networks
ROI Region of Interest
ONH Optic Nerve Head

iv