ABSTRACT
Studies show that fuzzy logic has different approaches for enhancing personal health care delivery in the
health care sector. Currently,
breast cancer
rated as the second leading cause of death among women, according
to the World Health
Organisation. Previous studies relating to breast cancer prognosis using fuzzy logic approach were directed at reoccurrence of the disease as well as the survivability of individual. However, there is need for early identification of the predisposing risk
factors to breast cancer growth and its elimination. Consequently, this study focuses on developing an efficient Mobile-based Fuzzy Expert System (MFES) for initial breast cancer growth prognosis that can predict the individual risk level and thus,
reduce the high incidence rate.
Facts relating to the predisposing factors
of breast cancer were elicited from four domain experts through direct contact;
this was used to generate the appropriate fuzzy
rules. The fuzzy inference approach was
employed to formulate the membership
functions and fuzzy rules to design the MFES. Mamdani approach was
used for the fuzzification of input and
de-fuzzification of the output. The system
accommodates imprecision tolerance and uncertainty to
achieve tractability, robustness and least solution cost. Java
expert system shell running
on Android
operating system was used
to achieve the mobile technology aspect of the system. For the purpose
of system
evaluation, 2500 data were collected from two health care centers in Nigeria using
random sampling
technique. The data were stratified into
twenty-five different strata. Each stratum contained 100 dataset
and four individual data were selected at random.
The result indicated that the
fact elicited from the experts serves as range values for the 12 risk factors
used for the fuzzification of the input and thus, 36 rules were generated. The
rules were used as basis for the development of the MFES. The developed MFES for breast
cancer growth prognosis recorded 96% accuracy for all dataset picked from the 25
different strata.
The system showed that with
higher number of fuzzy
rules focusing on pre-tumour growth and detailed predisposing risk factors;
the prediction of risk level was reliable.
This work provided the resource for an individual to personally examine the breast cancer risk level, showing that the predisposing risk factors can
be reduced by personal health monitoring. Though, MFES employed higher number of fuzzy rules unlike the
existing systems with
less number of rules; it supports
pre-tumor growth instead of post-tumor growth which was incapable of handling
the high incidence rate. It is therefore recommended that MFES be
used to detect predisposing breast cancer
risk levels
early enough. The main
contribution of this work is the reduction of the incidence rate in contrast to
the existing methods currently applied in the diagnosis of breast cancer.
TABLE OF CONTENTS
Title Page
Abstract
Table of Contents
Appendices
CHAPTER ONE: INTRODUCTION
1.1 Background to the Study
1.2 Statement
of the Problem
1.3 Objective
of the Study
1.4 Methodology
1.5 Justification for the Study
1.6 Significance
of the Study
1.7 Scope of the Study
1.8 Operational Definition of Terms
1.9 Biblical
Implication of Disease
CHAPTER TWO: REVIEW OF LITERATURE
2.0 Introduction
2.1
Mobile Computing
2.1.1 Mobile Applications for Personal Health Care Monitoring
2.1.2 Benefits of Mobile
Apps for Personal Health Care Monitoring
2.1.3 Android Mobile Application
using Java Expert system shell
2.2 The Current
State of Art
of Digital Technologies in the Health and Care
Sector
2.3 Soft Computing
2.3.1 Computing
with words
2.4 Fuzzy logic
2.4.1 Fuzzy Logic Theory
2.4.2 Fuzzy Logic in Breast Cancer
2.5 Fuzzy Inference System
2.5.1 The Structure of Fuzzy
Inference Systems
2.6 Fuzzy Sets
2.6.1 Logical Operation on fuzzy set
2.7 Other
Relevant Issues in Fuzzy Inference
System
2.7.1 Linguistic
variables and Fuzzy Rules
2.7.2 Membership Function
2.8 Breast Cancer
2.8.1 Risk Factors
2.8.2 Facts and figures about Cancer
2.9 Related
Works
2.10 Summary
CHAPTER THREE: METHODOLOGY
3.0 Introduction
3.1 Mobile-based Fuzzy Expert System for
Breast Cancer Growth
Prognosis design
3.2 Breast Cancer Prognosis Features
Specification
3.3 Design
stages of the Mobile-based Fuzzy
Expert System
3.3.1 Knowledge representation in JESS
3.3.2 Input Variables
3.3.3 Fuzzy Rule-based System Specification
3.3.4 Methods
for obtaining fuzzy rules
3.3.5 Rules
for the Mobile-base Fuzzy Epert System for breast cancer growth
Prognosis
3.3.6 Structure
of the Mobile-base Fuzzy Epert System for breast cancer
growth prognosis
3.3.7 The
Aggregation of the outputs rule
3.4 Data Collection
3.5 Unified Modeling Language
3.5.1 Use case Diagram
3.5.2 Class Diagram
3.5.3 Sequence
Diagram
3.6 Algorithm for the MFES
3.7 Ethical consideration
CHAPTER FOUR: DATA ANALYSIS, RESULTS AND
DISCUSSION
OF FINDINGS
4.0 Introduction
4.1 Development
Tools
4.2 Designed
System Specification
4.3 Design
Considerations
4.3.1 Fuzzy Modelling
4.3.2 Evolving
fuzzy systems for the Breast Cancer Prognosis
4.3.3 Fuzzy system parameters
4.4 System
Implementation
4.5 Implementation Details for the designed MFES for breast cancer
Prognosis
4.5.1 The MFES User Interface
4.6 Experimental Result
4.7 Discussion
4.7.1 Performance Evaluation of the designed system
CHAPTER FIVE: SUMMARY, CONCLUSION AND RECOMMENDATIONS
5.1 Summary
5.2
Conclusion
5.3 Recommendations
5.4 Contributions to Knowledge
5.5 Suggestion
for Further Studies
References
Appendix I: Breast
Cancer Dataset
Appendix II: Questions asked
to collect data from healthy individuals
Appendix
III: Codes for the designed Mobile Based Fuzzy Expert System
CHAPTER ONE
INTRODUCTION
1.1 Background to the Study
Information and Communication Technology (ICT), specifically mobile
health (mHealth), can play a key role in enhancing and enabling health care systems, when linked to
specific needs. The initiation of various types of mobile portable computer
devices – smartphones, private digitally
powered assistants, and tablet systems has influenced an appreciated positive
impact in many works of life which includes the health sector. This has been influenced
by the increasing excellence and availability of application software in the
health sector, (Aungst,
2013). These softwares are set of
instructions that have been written in a particular programming language to run
on a moveable portable aid or on a computer system to achieve a particular
purpose, (Wallace, Clark & White,
2012). In recent
development faster processors and improved memory in the
analysis of complex data in the
health sector have paved the way for diverse medical mobile expert systems.
These systems are either individualised or used by medical expert (Ozdalga, Ozdalga & Ahuja, 2012). These
portable application systems are designed to supplement the experts work in order to deliver a
resource that will advance the results for private health monitoring and at the
point of care (Aungst, 2013). There are existing medical expert system models and health calculators which include Breast Cancer Surveillance Consortium (BCSC) Risk Calculator
(for breast cancer risk calculation), the Breast Cancer Risk Assessment
Tool (the Gail model) often used by health care providers to estimate risk,
MedCalc. These models did
not explore detail risk factors for breast cancer growth, and detail fuzzy
rules were not explored as well. Most of the mobile health calculators
for breast cancer prognosis are
not user friendly. They are not
readily available for personal use and are majorly used by the medical
professionals in the the
health care sectors.
World Health Organisation (WHO) in 2012
described cancer as a leading cause global deaths. In, 2008 cancer accounted
for about 13% (7.6 million) deaths (WHO, 2017). There are divergent views on the exact cause of breast cancer. Though, knowing an individual risk
factors and preventing the growth of the malignant (breast cancer) could be a preferred approach to tackling
this disease because most research works that have developed models for prognosis and
diagnosis have not actually reduced the death rate (Global Cancer Facts &
Figures, 2015). This is because reviewed literatures have shown that the existing systems focused on
diagnosing/prognosing
the survivability and recurrence of the disease. By the time patients report at the
hospital for diagnosis, the tumour has grown to the metastatic stage where
survival is almost impossible.
Majority of the models applied in
medical field are naturally unclear. As a result of the unclear (fuzzy) nature
of medical data and models as well as the relationships that exist in
the models, fuzzy logic technique is suitable for medical applications. Fuzzy logic
(an aspect of soft computing) proposes approaches of result production that
have the capability of estimated representation of decisions. As a result of
the difficulty in medical exercise, the old-style numerical study methods are not satisfactory and
may not be suitable. The utmost causes of ambiguity are as follows:
- Incomplete data about an individual: either from patient or family members
- Often time, the patient’s state of health account is provided by the individual, or by the
family member. These data to a large extent are subjective and ambiguous.
- The well being check-up: Often time, medical practitioners get impartial
facts
- Laboratory test and prognosis results may also be subject to various mistakes.
- The delinquency of patient’s preceding health status check-up can also cause error in the test report.
- Symptoms might be faked or overstated more/fewer than they truly appear.
- Patients are likely to neglect some of the symptoms.
- Some symptoms might be indescribable by patients
Hence, fuzzy logic a
soft computing methodology has the capability to
reduce uncertainty in decision making in medical field.
1.2 Statement of the Problem
The most recurrent and second leading cause of
death in women is breast cancer. The inadequacies of the existing methods, such as Mammography, Magnetic
Resonance Imaging (MRI), Self-examination and others, account for the breast cancer high mortality. The shortcomings of the existing models
include:
- Late discovery of the cancerous germs - these methods only detect breast cancer at the metastatic stage. (the tumour has grown and spread to
other parts of the body);
- Existing models cause patients pains and related inconvenience which dissuade
women from voluntary screening. Thus, most people do not report cases of breast cancer until it has got
to the third stage and stack the odd of survival against the patient.
- Imprecise diagnosis
because it involves several layers of uncertainty. These shortcomings make
the traditional approaches inappropriate.
Thousands
of people fall victim to breast cancer every year due to limitation of medical services and the inability to use the existing services effectively. Late presentation of
cases at advanced stages when little or no benefit can be derived from any form
of therapy is the hallmark of breast cancer among Nigerian women. The available
breast cancer calculators are only focused on survivability and re-occurence and also not safe because individuals do not know where their personal
data is being saved. To curtail
the worsening incidence of breast cancer deaths, a Mobile-based
Fuzzy Expert System (MFES) for breast cancer pre-growth prognosis that would obviate the inadequacies of the
existing models, encourage voluntary personal screening and more importantly,
detect the risk of developing breast cancer is designed. Pre
growth prognosis
of a disease like breast cancer is very crucial to a successful
reduction of death rate caused by the disease. This research weaved its
solution/prognosis intervention around a nature motivated method that is biologically
inspired. This method would be able to detect the risk of early developments
and proffered likely solutions thereby reducing the consequence of
ignorance which may lead to death.
1.3 Objective of the Study
The general objective of this
study
work is to design and implement a Mobile-based Fuzzy Expert System (MFES) for breast
cancer pre-growth prognosis. The
fuzzy
expert system would be capable of capturing ambiguous and imprecise information prevalent in breast cancer prognosis. The specific objectives are to:
- determine the range values
for the Membership Function (Breast Cancer Risk
factors) using experts rating for the indicators for fuzzification;
- formulate
the membership functions using information in (1);
- design a MFES
for breast cancer pre-growth prognosis and
- implement and carry out performance evaluation of the developed mobile based fuzzy expert system using in comparison existing fuzzy logic models.
1.4 Methodology
In order to
achieve the stated objectives, the following approaches were considered:
1. Upper and lower
values were determined from the values (facts) collected from the domain
experts to determine the membership functions.
2. Membership
functions for all the risk factors were formulated, using the values in (1).
3. The rules for
all the risk factors were formulated.
4.
Java expert system shell (JESS) was used to
develop the MFES, using the
informations in (1), (2) and (3) and this runs on Android systems.
5.
The MFES performance evaluation
was carried out
using data from healthy people and
those already diagnosed with the disease and also in comparison with existing fuzzy logic models
1.5 Justification for the
Study
The inadequacy
of existing methods to identify breast cancer at the early stage and late
presentation is the hallmark in breast cancer issue. Based on the nature of the
disease (breast cancer), a nature inspired methodology became the best approach
to handle the short comings of existing
models. Further review of existing literatures on softcomputing approaches,
revealed that researchers anchored their solutions on survivability and
reoccurence of the disease. These approaches did not in no way reduce the death
issues associated with the disease. Hence, this reasearch work proposes and designed a
MFES for breast cancer pre growth prognosis. This system (MFES) using detailed
risk factors and judiciously formulated membership functions and fuzzy rules
could be capabale to pick the smallest information that initiates the growth of
the disease. Recommendations are made to individuals on appropriate action(s)
to take by the system.
1.6 Significance of the Study
Despite
the enomous research work on breast cancer, using fuzzy logic approaches death
rate is still on the increase. These research works focus were majorly on
survivabilty and recurrence of the disease which has not recorded
reasonable decrease in the global death
rate caused by the disease. Hence, there is the strong reason to tackle this
contemporary issue in a different approach that would reduce/eliminate the life
threatening disease. The findings of this study would benefit the global
society considering the global death rate caused by the disease (Breast
Cancer). Also considerating the great
efficiency in the use of mobile system in recent time, this study is timely and
relevant because breast cancer has become a serious contemporary issue. Global
use of the designed MFES for breast cancer growth prognosis by females would
reduce/eliminate this life threatening disease (breast cancer). The end users
of the designed MFES ranges from teenagers to adults who have not been diagnosed of
the disease in the global society. It is designed to run on widely used and
existing personal mobile devices which runs on android operating system. This
implies that it would be easily accessible at no extra cost.
1.7
Scope of the Study
Many research works have been carried out in
breast cancer prognosis, but the incidence rate of breast cancer is still on
the increase. The key issue here is on how to prevent the growth of breast cancer (tumour) and a user friendly system.
Having identified the gap (stated in chapter two) in other
research work, this study covers pre
growth prognosis of breast cancer using breast cancer risk factors to design a mobile
based pre breast cancer growth prognosis - Fuzzy Expert System approach. The system is capable of capturing early enough ambiguous
and imprecise information prevalent in the risk of developing breast cancer.
The Mamdani fuzzy inference Model was adopted because this research work
involves a number of
fuzzy if-then rules, each of which describes the local behaviour of the mapping
of each risk indicators
(breast cancer risk factors). Java Expert System Shell (JESS) was used in the coding of the Mobile-based
expert system which runs on Android personal devices. The Direct Rating method was used to elicit data
(facts) from medical experts for the fuzzification stage of the model. Tumour sample data was not collected from the body of the individuals. Data was
collected from both healthy people and those already diagnosed with the disease
for the performance evaluation of the Mobile Expert System.
1.8 Operational Definition of Terms
Crisp Set:
A
collection of objects taken from the universe of objects.
Fuzzy: Refers
to linguistic uncertainty.
Fuzzy Sets: Allow
objects to have membership in more than one set.
Fuzzy Statement: Describes
the grade of a fuzzy variable with an expression
Fuzzy Logic Rule: Uses membership functions as
variables.
Statement: A sentence which
unambiguously either holds true or holds false.
Membership
Function: Defines a fuzzy
set by mapping crisp values from its domain to the sets associated degree of
membership.
Universe of Discourse: Range of all possible values, or concepts, applicable to a system
variable.
1.9 Biblical
implication of disease
Behind every health issue and every
emotional or spiritual problem resides the "Spirit of Fear." The
Spirit of Fear is the Devil's faith working in people by using lies to control
them. And if we dwell on those lies long enough, we will begin to believe them,
thus resulting in responding to them which can
lead to all kinds of problems. We need to discover the root cause behind our problems. The things that we experience may be a manifestation of a
root. But thank God, we have an advocate, Jesus Christ, who is our Saviour and
the Truth that sets us free (Romans 7:24-25, 8:2). Many of the diseases are
being used by God to bring to surface what is in our hearts. God does
not give disease. The disease is the manifestation from a heart
condition. Just as a fever is an indication that something is amiss in our
bodies. These are warning signs. And we can either cooperate with God or deal
with what was exposed. God
already knows our heart; it's no surprise what is there. We are the ones who are surprised by what is there. But
as they come to the surface, don't run, don't
hide in the bushes like Adam did Deal with
it.(The Holy Bible, King James Version)
================================================================
Item Type: Postgraduate Material | Attribute: 110 pages | Chapters: 1-5
Format: MS Word | Price: N3,000 | Delivery: Within 30Mins.
================================================================
No comments:
Post a Comment