International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 04 | Apr-2016 www.irjet.net p-ISSN: 2395-0072
© 2016, IRJET | Impact Factor value: 4.45 | ISO 9001:2008 Certified Journal | Page 840
Review on Tuberculosis Detection Using Various Data Mining
Techniques
Rupali Zakhmi
1
, Jyoti Arora
2
1
Research Scholar, Dept. Computer Science and Engineering, Desh Bhagat University, Punjab, India
2
Assistant Professor, Dept. Computer Science and Engineering, Desh Bhagat University, Punjab, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Data Mining is one of the most important and
inspiring area of research with the goal of discovering
purposeful information from massive data sets. Data mining
plays an important role in healthcare field to detect causation
of various diseases, their treatment methods .Tuberculosis is
one of the well-known disorders among all the persons in the
nation including India. Tuberculosis is a virus which strikes the
immune system of an individual, usually transmits through air.
It is primarily occurs in lungs. It is the typical cause of necrosis.
This paper discussed about various data mining techniques to
detect tuberculosis such as Classification, Clustering and
Association. There are some parameters that are also useful to
detect Tuberculosis like Age, Cough, Fever, Chest pain and
Weight Loss etc.
Key Words: Tuberculosis, Classification, Clustering,
Association, Data Mining
1. INTRODUCTION
Tuberculosis is a serious problem and transmits
through bacteria known as Mycobacterium Tuberculosis. It
is prime mover of demise. If right treatment is not given to
the patient at proper time then it is very difficult to cure
from TB. This disease is found on cattle, birds as well as in
human being. The organs such as lungs are badly influenced
by tuberculosis in most of the tuberculosis cases. TB attacks
both grown-ups and kids. In early days, many different
techniques were used such as sputum smear microscopy,
chest radiography etc. These methods have several
drawbacks, such as these methods require expertise to
operate citified tools. These methods are suitable to obtain
better results on time. Sometimes some symptoms of
tuberculosis are same with other diseases, it leads to death.
Incomplete information given by the patient or patient’s
family can stand in the way to find right treatment.
To control over these problems, some researcher use
images, sounds or variables as inputs parameters. In this
research, we will take some variables as input parameters to
detect and identify tuberculosis. Most commonly data mining
techniques have discussed in this research [2].
The aim of using data mining is to find significant
information from vast data sets. It is also useful in the field of
healthcare where unforeseen and relevant information are
identified. Medical sector uses data mining techniques to
know about various diseases, their causes and treatments.
Techniques of data mining are very effective at the time of
decision regarding patient heath. Medical data contains all
about number of patients, cost of treatment, medical
facilities etc. Analyzing this data, healthcare introduced
powerful tool which extract important information that is
necessary for patient’s recovery. It also verifies how much
time is taken by patients for diagnosis. Identification of
tuberculosis at right time is very important. To enhance the
performance of patient’s treatment - Classification,
Clustering and Association approaches have been
introduced. Results of using data mining approaches provide
benefits to healthcare domain by grouping the patients
having same types of health issues [4].
Fig -1: Variables taken as input parameter to detect TB
In Fig-1,Eight variables are used to identify and detect
tuberculosis that are Age, Gender, Fever, Chest Pain, Weight
Loss, Cough, Night sweats and Hemoglobin. Brain H. Tracey
[8] uses cough detection algorithm to recover patient from
pulmonary tuberculosis. Author proposed classification
technique to decrease the cough count of patient.