Mobile Based Image Analysis System For Cervical Cancer Detection
ABSTRACT
Cervical cancer is the third major killer disease in developed and developing countries.
Whereas screening and other preventive measures reduce the mortality rate in developed
countries, mortality rates still remain very high in developing countries. This project focuses
on the analysis of a digital image of the cervix; captured with a low-level camera, under a
contrast agent: the visual inspection with acetic acid (VIA) is known as one of the reference
methods to detect cervical cancer. Gaussian and mean filter techniques were used to remove
the speckles. A segmentation algorithm was used to isolate the region of interest (ROI) from
the image. Additionally a canny edge detection algorithm was used to find edges.
Furthermore, quantification and classification of the images were done. An Android
application was used to integrate all the above. This allows usage in rural settings. The results
obtained were quite satisfactory (Specificity 79% and Sensitivity of 83%).
Â
TABLE OF CONTENTS
ABSTRACT………………………………………………………………………………………………………………4
ACKNOWLEDGEMENTS………………………………………………………………………………………..5
DEDICATION…………………………………………………………………………………………………………..6
LIST OF FIGURES……………………………………………………………………………………………………8
LIST OF TABLES……………………………………………………………………………………………………..9
LIST OF ABBREVIATIONS……………………………………………………………………………………10
CHAPTER ONE
INTRODUCTION……………………………………………………………………………………………………15
1.0 Background of Study…………………………………………………………………………………………15
1.1Problem Statement…………………………………………………………………………………………….15
1.2Motivation and Purpose………………………………………………………………………………………15
1.3Research Contribution………………………………………………………………………………………..15
1.4Research Objective and Scope…………………………………………………………………………….15
1.5 Target Platform…………………………………………………………………………………………………15
1.6 Thesis Overview……………………………………………………………………………………………….16
1.7 Chapter Summary……………………………………………………………………………………………..16
CHAPTER TWO
LITERATURE REVIEW…………………………………………………………………………………………17
2.0 Related Works………………………………………………………………………………………………….17
2.1 Digital Images………………………………………………………………………………………………….18
2.1.1 Image Acquisition……………………………………………………………………………………….18
2.1.2 Image Pre-processing………………………………………………………………………………….18
2.1.3 Image Segmentation……………………………………………………………………………………19
2.1.4 Image Classifications………………………………………………………………………………….20
2.2 Android Programming………………………………………………………………………………………20
2.2.1 Introduction……………………………………………………………………………………………….20
2.2.2 The Android Software Stack………………………………………………………………………..21
2.2.3 Android Building Blocks……………………………………………………………………………..21
2.3 Chapter Summary……………………………………………………………………………………………..24
CHAPTER THREE
CERVICAL CANCER……………………………………………………………………………………………..25
3.0 Introduction……………………………………………………………………………………………………..25
3.1 Anatomy of Human Cervix………………………………………………………………………………..25
3.2 Development of Precancer and Cancer………………………………………………………………..25
3.3 Causes and Symptoms……………………………………………………………………………………….25
3.4 Treatment and Diagnosis……………………………………………………………………………………26
3.5 Colposcopy………………………………………………………………………………………………………26
8
3.6 Cervical Cancer Imaging……………………………………………………………………………………27
3.7 Chapter Summary……………………………………………………………………………………………..28
CHAPTER FOUR
APPROACH AND METHODS………………………………………………………………………………..29
4.1 Design Methodology…………………………………………………………………………………………29
4.1.1 User Interface Design………………………………………………………………………………….29
4.1.2 Image Enhancement……………………………………………………………………………………29
4.1.3 Conversion to greyscale……………………………………………………………………………..30
4.1.4 Image Segmentation……………………………………………………………………………………30
4.1.5 Edge detection…………………………………………………………………………………………..31
4.1.6 Image Classification……………………………………………………………………………………31
4.1 Conceptual Diagram………………………………………………………………………………………….31
4.2 System Design………………………………………………………………………………………………….31
4.2.1 Activity Diagram………………………………………………………………………………………..31
4.2.2 Class diagram…………………………………………………………………………………………….33
4.3 Chapter Summary……………………………………………………………………………………………..33
CHAPTER FIVE
RESULT AND DISCUSSIONS…………………………………………………………………………………34
5.1 Data Set…………………………………………………………………………………………………………..34
5.3 Performance and Measurement…………………………………………………………………………..34
5.3.1 Sensitivity………………………………………………………………………………………………….34
5.3.2 Specificity………………………………………………………………………………………………….34
5.4 Results…………………………………………………………………………………………………………….35
5.4 Discussion……………………………………………………………………………………………………….37
5.5 Chapter Summary……………………………………………………………………………………………..37
CHAPTER SIX
SUMMARY, CONCLUSION AND FUTURE WORK……………………………………………….38
6.1 Summary…………………………………………………………………………………………………………38
6.2 Conclusion……………………………………………………………………………………………………….38
6.3 Future Work……………………………………………………………………………………………………..38
REFERENCES………………………………………………………………………………………………………..39
APPENDIX……………………………………………………………………………………………………………..41
9
CHAPTER ONE
INTRODUCTION
1.0 Background of Study
Cervical cancer is one of the curable types of cancers in women if detected early. Most cases
of cervical cancer are caused as a result of infection with certain types of Human
Papillomavirus (HPV) [4, 6]. Although women who have early exposure to sexual
relationships and those with multiple sexual partners are at high risk of contracting HPV and
eventually, cervical cancer, it is however possible for a woman to be infected with HPV even
if she has had only one sexual partner. In the developed nations, women above the age of 30,
who are at high risk of HPV infection, are given HPV vaccines, to reduce the chances of
having the disease [4].
Traditionally, optical tests such as VIA, cervicography and colposcopy that employ direct
visual examination of the cervix, are becoming popular as a diagnostic tool. Healthcare
professionals study the cervix at about one minute after applying the 5% acetic acid to the
cervix area. Acetowhite region (AW), which is the suspected region of cervix, and other
vascular abnormalities such as mosaicism, punctuation and vasculature may appear [4].
Cervical cancer is second only to breast cancer as the highest cause of cancer-related death of
women in the world [1]. In 2012, it was the fourth leading cause of cancer death in women
worldwide with an estimate of about 65,700 deaths. Unfortunately, up to 90% of these deaths
occurred in the developing nations of the world, especially in sub-Saharan Africa; 60,100
deaths in Africa, 28,600 in Latin America and the Caribbean, and 144,400 in Asia. India, the
second most populous country in the world, accounted for 28,600. Latin America and the
Caribbean, accounted for 25% (67,500) of cervical cancer deaths [3].
The main reason for this discrepancy is the lack of organized, population-wide, screening
programmes and medical personnel to administer and translate various test involved [6].
This thesis work seeks to address the problem of the lack of medical personnel to administer
and translate various test by coming up with the automation of cervical cancer detection using
digital images of low resolution, deployed on Android mobile devices.
14
1.1 Problem Statement
More than 80% of the total deaths recorded due to cervical cancer in women have occurred in
sub-Saharan Africa. The absence of both the resources for screening to aid early detection,
and most importantly, the scarcity of medical personnel to interpret these results is the major
contributing factor. Hence, the need to design an Android application, which performs; the
analysis and classifies cervix digital image. This will be very useful in the areas with limited
or no medical experts.
1.2 Motivation and Purpose
The development in mobile computing in the recent year has made it a great tool in
healthcare [5]. These devices are gaining more acceptability due to their portability and
reduced price. We are not aware of any existing work done in the deployment of an
automated cervical cancer detection tool on mobile platform.
1.3 Research Contribution
A lot of research work has been carried out on digital images for cervical cancer detection.
This thesis seeks to contribute to the body of knowledge by using digital images captured
with a low-resolution mobile device’s camera and also the development of an Android
demonstrator as a mobile tool in the automation of cervical cancer detection.
1.4 Research Objective and Scope
This research work aims to design a mobile based digital image analysis system for cervical
cancer detection. We seek to achieve the following objectives;
To critical study several algorithms involved in digital image processing and analysis;
To select an appropriate approach suitable for cervical cancer image analysis;
To develop an Android application that serves as a demonstrator for the designed
approach; and
To integrate the applications and tools on an Android-based mobile device
1.5 Target Platform
The Android operating system was selected as our platform of choice due to the numerous
advantages it offers. It has the largest user base of all the smartphone operating systems. Also,
15
it is completely open source. Moreover, Android development is relatively easy; most of its
syntax and structure are built around the Java programming language and Extensible Markup
Language (XML).
Fig 1.0: Smartphone OS market share
1.6 Thesis Overview
Chapter one – introduces the research work, and highlights the research contribution,
objectives and scope and the methodology to be employed.
Chapter two – critically looks into the review of literature. Some of the related works are
briefly reviewed and basic concepts in digital imaging are discussed.
Chapter three – discusses cervical cancer in detail: the anatomy of the cervix, precancerous
and cancerous lesions, causes and preventions etc.
Chapter four – extensively highlights the implemented approach & methods.
Chapter five – presents the result of findings, performance measurement and overview
discussions.
Chapter six – concludes the thesis work by summarizing the thesis and suggesting future
directions.
16
1.7 Chapter Summary
This chapter introduces the main thesis work. It highlights the research contribution,
objectives and scope, the motivation and purpose, and finally the methodology to be
employed.
In the next chapter, we will explore some of the related works that has been done in this
realm of research. Some important concepts that relate to this thesis work will also be
discussed.
IF YOU CAN'T FIND YOUR TOPIC, CLICK HERE TO HIRE A WRITER»
