Evaluation of Classification Algorithms vs Knowledge-Based Methods for Differential Diagnosis of Asthma in Iranian Patients

Evaluation of Classification Algorithms vs Knowledge-Based Methods for Differential Diagnosis of Asthma in Iranian Patients

Reza Safdari (Department of Health Information Technology, Tehran University of Medical Sciences, Tehran, Iran), Peyman Rezaei-Hachesu (Department of Health Information Technology, Tabriz University of Medical Sciences, Tabriz, Iran), Marjan GhaziSaeedi (Department of Health Information Technology, Tehran University of Medical Sciences, Tehran, Iran), Taha Samad-Soltani (Department of Health Information Technology, Tabriz University of Medical Sciences, Tabriz, Iran) and Maryam Zolnoori (National Library of Medicine, Bethesda, USA)
Copyright: © 2018 |Pages: 14
DOI: 10.4018/IJISSS.2018040102

Abstract

Medical data mining intends to solve real-world problems in the diagnosis and treatment of diseases. This process applies various techniques and algorithms which have different levels of accuracy and precision. The purpose of this article is to apply data mining techniques to the diagnosis of asthma. Sensitivity, specificity and accuracy of K-nearest neighbor, Support Vector Machine, naive Bayes, Artificial Neural Network, classification tree, CN2 algorithms, and related similar studies were evaluated. ROC curves were plotted to show the performance of the authors' approach. Support vector machine (SVM) algorithms achieved the highest accuracy at 98.59% with a sensitivity of 98.59% and a specificity of 98.61% for class 1. Other algorithms had a range of accuracy greater than 87%. The results show that the authors can accurately diagnose asthma approximately 98% of the time based on demographics and clinical data. The study also has a higher sensitivity when compared to expert and knowledge-based systems.
Article Preview

Introduction

Asthma is a common chronic inflammatory disease of the airways that is characterized by aggravated symptoms, reversible airway obstruction, and bronchial spasms (Louis et al., 2000), and the incidence of this chronic disease is estimated to range from 1.4% to 27.1% in different areas of the world (Zolnoori, Zarandi, Moin, Heidarnezhad, & Kazemnejad, 2012b). The prevalence of asthma has increased dramatically since the 1970s, and in 2014 alone, 334 million people were afflicted with asthma worldwide (Network, 2014), and it caused 250,000 deaths (Bousquet et al., 2010). Asthma dramatically affects patients and shows an increasing rate in developing countries (Beasley, Crane, Lai, & Pearce, 2000; Zolnoori, Zarandi, & Moin, 2012). Asthma diminishes the patient’s quality of life. Despite understanding the mechanism of asthma pathology that laboratory facilities have promoted, the treatment process involves several critical problems, including lack of proper diagnosis and incorrect estimation of severity (Zolnoori, Zarandi, & Moin, 2012). Similarly, insufficient experience of general practitioners (GPs) increases the difficulty of producing a correct asthma diagnosis. Physicians, in the diagnosis of asthma, make decisions based on symptoms and patient history (Zarandi, Zolnoori, Moin, & Heidarnejad, 2010).

Data mining is defined as a process for finding patterns and relationships along complex data in a database to build predictive models for decision-making (Kincade, 1998). Moreover, it is the process of selecting, exploring, and building models using mass stored data to identify pre-existing patterns (U. M. Fayyad, Piatetsky-Shapiro, Smyth, & Uthurusamy, 1996; Kavita, 2017). Data mining is used to identify new, accurate, understandable, and potentially useful patterns and relationships within the data by using a combination of sophisticated mining models to apply to human problems (Koh & Tan, 2011; Zhongxian, Ruiliang, Qiyang, & Ruben, 2010). Data mining in health care is used for proper diagnosis and for gaining a deeper understanding of medical data. Medical practitioners use it to solve real-world problems in the diagnosis and treatment of diseases (Liao & Lee, 2002; Soltany, Langarizadeh, & Shanbezadeh, 2013; Samad-Soltanim Ghanei & Langarizadeh,2015). Researchers use it for classification and diagnosis of various diseases; therefore, they apply various techniques and algorithms that possess different levels of accuracy and precision (Shouman, Turner, & Stocker, 2012). Several studies have reported the diagnosis of asthma using data mining algorithms (Alizadeh, Safdari, Zolnoori, & Bashiri, 2015; Chakraborty, Mitra, Mukherjee, & Ray, 2009; Choi et al., 2007; Fazel Zarandi, Zou, Moein, & Heydarnezhad, 2010; Prasadl, Prasad, & Sagar, 2011; Samad Soltani, Langarizadeh, & Zolnoori, 2015; Tyagi & Singh, 2014; Zolnoori, Zarandi, Moin, Heidarnezhad, & Kazemnejad, 2012a).

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 11: 4 Issues (2019): 1 Released, 3 Forthcoming
Volume 10: 4 Issues (2018)
Volume 9: 4 Issues (2017)
Volume 8: 4 Issues (2016)
Volume 7: 4 Issues (2015)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing