AWERProcedia
Information Technology
&
Computer Science
1 (2012) 1074-1081
2
nd
World Conference on Information Technology (WCIT-2011)
Headache diagnosis with K-Means algorithm
Nilufer Yurtay
*
, Ufuk Celik , Yuksel Yurtay
Sakarya University, Computer and Informatic Engineering Department, Esentepe Campus 5418, Sakarya TURKEY
Abstract
Proposing an accurate headache diagnosis for a patient can be a really difficult task considering all types of headaches.
According to International Headache Classification there are 272 headache types. A computer aided diagnosis system
provides great facility of achieving information and fast ways for the doctors. Headache diagnosis is a classification problem
and K-means algorithm is one of the most widely used clustering methods for classification. In this study, K-Means which is
one of the data mining clustering algorithms has been used in order to develop a medical diagnosis for migraine without aura
and probable migraine without aura.
Keywords: Headache, Migraine, Clustering algorithms, Data mining;
Selection and peer review under responsibility of Prof. Dr. Hafize Keser.
©2012 Academic World Education & Research Center. All rights reserved.
1. Introduction
Medical decision support systems studies have been greatly expanded in recent years and used in wide range
of design and it will be inevitable in the future. Although it is difficult to develop a clinical decision support
system for gynecological diseases, a decision support system developed using Turbo Prolog with artificial neural
network systems helps a gynecologist to select the medicine of a particular gynecological disease quickly, easily
and precisely.[1] A clinical decision support system named DXplain in Minesota Mayo Clinic was implemented
and used to improve assess of general practitioners and interns.[2] An application named SIBAS developed using
ORACLE provided a decision support system for blood bank staffs specified various operational rules for donor
screening, donation evaluation, sample testing, blood processing and product dissemination with special
concerns on blood donation and transfusion service.[3] Inappropriate medication among elderly people
* ADDRESS FOR CORRESPONDENCE:
a b c
a
Nilufer, Yurtay, Sakarya University, Computer and Informatic Engineering Department, Esentepe Campus
5418, Sakarya TURKEY
E-mail address: ucelik001@gmail.com