International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 3, Issue 1, January 2014
ISSN 2319 - 4847
Hybridization Of Neural Network Using
Genetic Algorthim For Heart Disease
Detection
Midhuna.R1, Shilpa Mehta2
1
Third Sem M Tech (Signal Processing) ,Department Of Electronics And Communication Engineering, Reva Institute Of
Technology and Management, Bangalore, Karnataka, India.
2
Sr. Associative Professor, Department Of Electronics And Communication Engineering, Reva Institute Of Technology And
Management, Bangalore, Karnataka, India.
Abstract
The study is aimed at developing assistive supportive system for practionars. This paper describes which a few system are being
used for heart disease prediction based on feed forward neural network architecture and genetic algorithm. Hybridization has
applied to train the neural network using Genetic algorithm and it has been validated, hybrid learning is more stable compare to
back propagation. Learning detailed analysis has been done with respect to genetic algorithm behavior and its relationship with
learning performance of neural network, and results are compared.
Keywords: Neural Network Architecture, Genetic Algorithm, Heart Disease Prediction, Hybrid.
1. Introduction
According to the World Health Organization (WHO) heart disease and stroke kill some 17 million people a year, By
2020, heart disease and stroke will become the leading cause of both death and disability worldwide[1].
“No matter what advances there are in high-technology medicine, the fundamental message is
that any major reduction in deaths and disability from heart disease and stroke will come primarily from prevention, not
just cure. We believe that early cardiac physical examination (screening) will become a fundamental element of any
prevention program.” Quoted by Dr. Judith Mackay, co-author of the WHO Atlas of heart disease and stroke [1, 2].
The term Heart disease encompasses the diverse diseases that affect the heart. Heart disease was the major cause of
casualties in the United States, England, Canada and Wales as in 2007. Heart disease kills one person every 34 seconds in
the United States. Coronary heart disease, Cardiomyopathy and Cardiovascular disease are some categories of heart
diseases. The term “cardiovascular disease” includes a wide range of conditions that affect the heart and the blood vessels
and the manner in which blood is pumped and circulated through the body. Cardiovascular disease (CVD) results in
severe illness, disability, and death.
Narrowing of the coronary arteries results in the reduction of blood and oxygen supply to the heart and leads to the
Coronary heart disease (CHD). Myocardial infarctions, generally known as a heart attacks, and angina pectoris, or chest
pain are encompassed in the CHD. A sudden blockage of a coronary artery, generally due to a blood clot results in a heart
attack. Chest pains arise when the blood received by the heart muscles is inadequate. High blood pressure, coronary artery
disease, valvular heart disease, stroke, or rheumatic fever/rheumatic heart disease are the various forms of cardiovascular
disease.
Objective
• The main objective of this project deals with the design and implementation of expert system for medical diagnosis.
• Development of a system by using ANN technology & Genetic algorithm for the prediction of heart disease with high
accuracy. To achieve high accuracy genetic algorithm will be involved with neural technology.
2. Related Works
2.1. Block diagram
Volume 3, Issue 1, January 2014
Page 482
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 3, Issue 1, January 2014
ISSN 2319 - 4847
2.2 Problem Formulation
There are being decision support system for predicting the existence or absence of heart disease. Wrong diagnosis or poor
clinical decisions leads to mortality, All clinicians are not equally good in predicting the heart disease in which diagnosis
plays a very important role. In the case of heart disease “time is a precious”, proper diagnosis at the right time saves life
of many patients. The system can be considered assisting the doctor to come to decision making.
2.3 Human Experts Vs. Machine Expert In Diagnosis
Almost all the physicians are confronted during their conformation by the task of learning to diagnose. Here, they have to
solve the problem of deducing certain diseases or formulating a treatment based on more or less specified observations
and knowledge[1]. The diagnosis of disease is a significant and tedious task in medicine. The detection of heart disease
from various factors or Symptoms is a multi-layered issue which is not free from false presumptions often accompanied by
unpredictable effects [2].
With the complexity of information available from health care domain, we need better method to aid and assist physician
alone is not good enough to ascertain the proper diagnosis. The problems associated with people involve in the diagnosis
can considered broadly as (1) not having very high accuracy in decision (ii) shortage of expertise (iii) difficulties in
knowledge up gradation (iv) time dependent performance .
Because of these problems there is necessity to develop the expert system to provide the assistance mechanism in
diagnosis procedure [3].
The conclusion is clear: humans may make error complex data. Intelligent systems are increasing being deployed in
medicine and healthcare, to practically aid the busy clinician and to improve the quality of patient care. The need for an
objective methodology for evaluating such systems is widely recognized .In medicine and healthcare where safety is
critical, this is important if techniques such as medical expert systems and neural systems are to be widely accepted in
clinical practice [2,3].
2.4 Computational Network
Artificial neural networks are computational paradigms based on mathematical models that unlike traditional computing
have a structure and operation that resembles that of the mammal brain. Artificial neural networks or neural networks for
short are also called connectionist systems or parallel distributed systems or adaptive systems, because they are composed
by a series of interconnected processing elements that operate in parallel [1].
Neural networks lack centralized control in the classical sense, since all the interconnected processing elements change
or “adapt” Simultaneously with the flow of information and adaptive rules. Neural networks are being applied to an
increasing large number of real world problems. Their primary advantage is that they can solve problems that are too
complex for conventional technologies; problems that do not have an algorithmic solution or for which an algorithmic
solution is too complex to be defined.
In algorithmic approach, the computer follows a set of instructions in order to solve a problem. Unless the specific steps
that the computer needs to follow are known, the computer cannot solve the problem. That restricts the problem solving
capability of conventional computers to problems that we are already understood and knows how to solve. These problems
include pattern recognition and forecasting – which requires the recognition of trends in data. In case of neural network,
Volume 3, Issue 1, January 2014
Page 483
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 3, Issue 1, January 2014
ISSN 2319 - 4847
even for imprecise inputs the network is able to retrieve the desired output or the data that is closest to the desired output
[2].
Considering the successful applicability of neural networks in many areas, an endeavour to assess their performance by
providing the assistance in heart disease is the basis for this work. Artificial Neural Networks (ANNs) [5] ,have been
widely used as tool for solving many decisions modelling problems. The various capabilities and properties of ANN like
Nonparametric, on-linearity, Input-Output Mapping, Adaptively make it a better alternative for solving massively parallel
distributed structure and complex task in comparison of statistical techniques, where rigid assumption are made about the
model. Artificial Neural Network being non-parametric, makes no assumption about the distribution of the data and thus
capable of “letting the data speak for itself”. As a result, they are a natural choice for modelling complex medical
problems when large databases of relevant medical information are available.
3. Algorithm
Basic ideas on Genetic algorithms were first developed by John Holland, and are mainly used as search and optimization
methods. Given a large solution space, one would like to pick out the point which optimizes an object function while still
fulfilling a set of constraints. In network planning, a solution point could be a specific link topology, a routing path
structure, or a detailed capacity assignment with minimum costs [4].
Genetic algorithms are based on the idea of natural selection. In nature, the properties of an organism are determined by
its genes. Starting from a random first generation with all kinds of possible gene structures, natural selection suggests that
over the time, individuals with "good" genes survive whereas "bad" ones are rejected. Genetic algorithms try to copy this
principle by coding the possible solution alternatives of a problem as a genetic string.
The genes can be bits, integers, or any other type from which a specific solution can be deduced. It is required that all
solution points can be represented by at least one string. On the other hand, a specific gene string leads to exactly one
solution. A set of a constant number of gene strings, each characterizing one individual, is called a generation. Since the
different strings have to be evaluated and compared to each other, the notion of fitness is introduced. The fitness value
correlates to the quality of a particular solution.
Instead of working with the actual solution itself, genetic algorithms operate on the respective string representation. The
following three basic operators are applied
(i) Reproduction
(ii)Crossover
(iii) Mutation.
The reproduction process creates a new generation, starting from an existing generation; strings are reproduced with a
probability respective to their fitness value. Strings which represent solutions with good properties have a higher chance
to survive than strings depicting solution points with bad characteristics. This principle is also known as "survival of the
fittest”. The crossover operator exchanges genetic information between two strings. The strings of two randomly selected
solutions are broken up at randomly chosen position, and parts of the strings are exchanged. One hopes that two solutions
with good properties create an even better one. New genetic material is introduced by the mutation operator.
The values of individual genes are changed and hence, new solutions are chosen. Mutation becomes important when after
some generations the number of different strings decreases because strong individuals start dominating. In a situation of
strong dominance of few strings, the crossover operator alone would not bring any changes and the search for an optimal
solution would be ended. To partially shift the search to new locations in the solution space, the mutation operator
randomly alters genes [3, 4].
4. Discussion
We propose an original idea to design an expert system, any clinician irrespective of their expertise can utilize this
project. Artificial Neural Network (ANN) is a technology that helps to develop decision supportive assist system, which
will standardize the procedure of predicting the heart decision. Mat lab is used to develop this decision supportive system.
Currently we are working through the input attributes parts for training neural network.
5. Conclusion
This study helps us to understand various approaches for predicting heart disease; further work will be done in developing
better and more accurate and fast prediction methodologies. This will help patients to avoid heart disease problems.
Volume 3, Issue 1, January 2014
Page 484
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 3, Issue 1, January 2014
ISSN 2319 - 4847
References
[1]. Shen, Z. Dept. of Electr. Eng., Brunel Univ., Uxbridge, UK Clarke, M. ; Jones, R. ; Alberta,
T. “A Neural
Network Approach to the Detection of Coronary Artery Disease’’, Computers in Cardiology 1993, Proceedings.
Page(s):221 – 224.
[2]. Mehdi, Bushra Electronic Engineering Sir Syed University of Engineering & Technology, Karachi, Pakistan Khan,
Tahmina ; Ali, Zain Anwar,”Cardiac Sudden Death Risk Detection Using Hybrid Neuronal-Fuzzy Networks”,
Electrical and Electronics Engineering, 2006 3rd International Conference on,6-8 Sept. 2006, Page(s):1 – 4.
[3]. Jadhav, S.M. Dept. of Inf. Technol., Dr. Babasaheb Ambedkar Technol. Univ., Lonere, India Nalbalwar, S.L. ;
Ghatol, A.A.’’Data Fusion for Heart Diseases Classification Using Multi-Layer Feed Forward Neural Network
‘’Computer Engineering & Systems, 2008. ICCES 2008. International Conference on25-27 Nov. 2008 Page(s):67 70.
[4]. Valavanis,I.K.,School
Of
Electrical
And
Computer
Engineering,
National
Technical
UniversityOfAthens,9heroonPolytechneiouStr,15780Zographou,Geece,Mougiakakou,S.G;Grimaldi,Ka,Nikita Ks,”.
Analysis Of Postprandial Lipemia As A Cardiovascular Disease Risk factor Using Genetic And Clinical Information:
An Artificial Neural network Perspective”, Engineering In Medicine And Biology Society 2008.Embs 2008.30th
Annual International IEEE Embs Conference Vancouver, British Columbia, Canada, August 20-24, 2008 Pages
4609-4612.
[5]. Kumaravel, N. Coll. of Eng., Anna Univ., Madras, India Sridhar, K.S. ; Nithiyanandam, N.’’Artificial Neural
Network Based Cardiac Arrhythmia Disease Diagnosis’’,Process Automation, Control and Computing (PACC), 2011
International Conference on 20-22 July 2011,Page(s):1 – 6.
AUTHORS PROFILE
Midhuna R. , She majored in Medical Electronics Engineering and received the B.E. degree in 2011 from the
VTU ,Belgaum. Currently pursuing her M Tech in Signal Processing degree from the VTU Belgaum. She
worked previously as faculty in Sri Krishna Institute Of Technology dated 2011-2012. She has presented and
published more than 10 conference papers at national and one in international conferences. Her current
research interests include image processing, video processing, artificial neural networks, and medical
imaging.
Shilpa Mehta completed her B.E. in Electronics from Vikram University, Ujjain (MP) in 1991, being the
gold medalist of her batch, and her M. E. in Digital Techniques from DAVV Indore (MP) in 1997. She
started her teaching career in the year 1992. She has a total teaching experience of about 20 years. She has
presented more than 10 conference papers at national and international conferences. She is presently
pursuing her PhD from JNTU Anantapur. Presently, she is working as an Associative Professor at Reva ITM, Bangalore.
Volume 3, Issue 1, January 2014
Page 485