28-07-2014, 03:04 PM
USING DATA MINING TECHNIQUES IN HEART DISEASE DIAGNOSIS AND TREATMENT
DATA MINING.pdf (Size: 239.61 KB / Downloads: 52)
Abstract
The availability of huge amounts of medical data leads
to the need for powerful data analysis tools to extract useful
knowledge. Researchers have long been concerned with applying
statistical and data mining tools to improve data analysis on large
data sets. Disease diagnosis is one of the applications where data
mining tools are proving successful results. Heart disease is the
leading cause of death all over the world in the past ten years.
Several researchers are using statistical and data mining tools to
help health care professionals in the diagnosis of heart disease.
Using single data mining technique in the diagnosis of heart
disease has been comprehensively investigated showing
acceptable levels of accuracy. Recently, researchers have been
investigating the effect of hybridizing more than one technique
showing enhanced results in the diagnosis of heart disease.
However, using data mining techniques to identify a suitable
treatment for heart disease patients has received less attention.
This paper identifies gaps in the research on heart disease
diagnosis and treatment and proposes a model to systematically
close those gaps to discover if applying data mining techniques to
heart disease treatment data can provide as reliable performance
as that achieved in diagnosing heart disease.
INTRODUCTION
Data mining is the exploration of large datasets to extract
hidden and previously unknown patterns, relationships and
knowledge that are difficult to detect with traditional statistical
methods [1-5]. Data mining is rapidly growing successful in a
wide range of applications such as analysis of organic
compounds, financial forecasting, healthcare and weather
forecasting [6]. Data mining in healthcare is an emerging field
of high importance for providing prognosis and a deeper
understanding of medical data [7]. Data mining applications in
healthcare include analysis of health care centers for better
health policy-making and prevention of hospital errors, early
detection, prevention of diseases and preventable hospital
deaths, more value for money and cost savings, and detection
of fraudulent insurance claims [8]. Researchers are using data
mining techniques in the diagnosis of several diseases such as
diabetes [9], stroke [10], cancer [11], and heart disease [12].
Heart disease is the leading cause of death in the world over
the past 10 years [13]. The European Public Health Alliance
reported that heart attacks, strokes and other circulatory
diseases account for 41% of all deaths [14]. The Economical
and Social Commission of Asia and the Pacific reported that in
USING SINGLE AND HYBRID DATA MINING TECHNIQUES IN HEART DISEASE DIAGNOSIS
Statistical analyses have identified the risk factors
associated with heart disease to be age, blood pressure,
cholesterol, and smoking habit [27], total cholesterol [28],
diabetes [29], hypertension and family history of heart disease
[30], obesity, lack of physical activity, and high levels of
smoking [31]. Knowledge of the risk factors associated with
heart disease helps health care professionals to identify patients
at high risk of having heart disease. Heart disease professionals
store significant amounts of patients’ data. It is important to
analyze these datasets to extract useful knowledge. Data
mining is an effective tool for analysing data to extract useful
knowledge.
Different data mining techniques have been used to help
health care professionals in the diagnosis of heart disease.
Those most frequently used focus on classification: naïve
bayes, decision tree, and neural network. Other data mining
techniques are also used including kernel density,
automatically defined groups, bagging algorithm, and support
vector machine.
Table 1 shows a sample of different data mining techniques
used in the diagnosis of heart disease over different heart
disease datasets. The results of the different data mining
research cannot be compared because they have used different
datasets. However, over time a defacto benchmark data set has
arisen in the literature: the Cleveland Heart Disease Dataset
(CHDD http://archive.ics.uci.edu/ml/datasets/Heart+Disease).
Results of trials on this dataset do allow comparison.
TRENDS IN USING DATA MINING TECHNIQUES IN HEART DISEASE
Although applying data mining is beneficial to healthcare
[3, 23, 37], disease diagnosis [12, 21, 38], and treatment [39-
41], few researches have investigated producing treatment
plans for patients [40]. Accurate diagnosis and treatment given
to patients have been major issues highlighted in medical
services.
Recently, researchers started investigating using data
mining techniques to handle the error and complexity of
treatment processes for healthcare providers. Razali and Ali
(2009) investigated generating treatment plans for acute upper
respiratory infection disease patients using a decision tree. The
model recommended treatment through giving drugs to patients
showing accuracy of 94.73% [39]. Applying association rules
PROPOSED RESEARCH MODEL
We propose that applying data mining techniques in
identifying suitable treatments for heart disease patients is
fruitful and needs further investigation. To evaluate if applying
data mining techniques to heart disease treatment can provide
as reliable performance as achieved in heart disease diag
SUMMARY
Motivated by the world-wide increasing mortality of heart
disease patients each year and the availability of huge amounts
of data, researchers are using data mining techniques in the
diagnosis of heart disease. Although applying data mining
techniques to help health care professionals in the diagnosis of
heart disease is having some success, the use of data mining
techniques to identify a suitable treatment for heart disease
patients has received less attention. Also, applying hybrid data
mining techniques has shown promising results in the diagnosis
of heart disease, so applying hybrid data mining techniques in
selecting the suitable treatment for heart disease patients needs
further investigation. This paper identifies gaps in the research
on heart disease diagnosis and treatment and proposes a model
to systematically close those gaps to discover if applying data
mining techniques to heart disease treatment data can provide
as reliable performance as that achieved in diagnosing heart
disease patients