Diagnosis Rule Extraction from Patient Data for Chronic Kidney Disease Using Machine Learning

Diagnosis Rule Extraction from Patient Data for Chronic Kidney Disease Using Machine Learning

Alexander Arman Serpen (Southview High School, Sylvania, OH, USA)
Copyright: © 2016 |Pages: 9
DOI: 10.4018/IJBCE.2016070105
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

This research study employed a machine learning algorithm on actual patient data to extract decision making rules that can be used to diagnose chronic kidney disease. The patient data set entails a number of health-related attributes or indicators and contains 250 patients positive for chronic kidney disease. The C4.5 decision tree algorithm was applied to the patient data to formulate a set of diagnosis rules for chronic kidney disease. The C4.5 algorithm utilizing 3-fold cross validation achieved 98.25% prediction accuracy and thus correctly classified 393 instances and incorrectly classified 7 instances for a total patient count of 400. The extracted rule set highlighted the need to monitor serum creatinine levels in patients as the primary indicator for the presence of disease. Secondary indicators were pedal edema, hemoglobin, diabetes mellitus and specific gravity. The set of rules provides a preliminary screening tool towards conclusive diagnosis of the chronic kidney disease by nephrologists following timely referral by the primary care providers or decision-making algorithms.
Article Preview

Introduction

Chronic kidney disease (CKD), also known as chronic renal disease, is a medical condition in which kidney function is lost over a long period ranging from months to years (NKF, 2002). There is no specific set of detectable symptoms for the disease but possible symptoms may include feeling generally unwell or having a reduced appetite. Therefore, chronic kidney disease is most often detected in individuals who are at high risk through advanced screening processes that determine health attributes associated with chronic kidney disease. Such attributes include high blood pressure, diabetes, or having a blood type strongly associated with the presence of chronic kidney disease. The presence of this disease can also be identified from a blood test for creatinine, a breakdown product of muscle metabolism. Chronic kidney disease progresses through multiple stages, with each having its own unique health complications. What differentiates chronic kidney disease from its counterpart, acute kidney disease, is that the reduction in kidney function develops and strengthens over the course of at least three months in CKD. Acute kidney disease occurs rapidly over the course of a few hours or days and is easily reversible (Mayo Clinic, n.d.). The attributes or indicators associated with CKD are the subject of this study as they were used to assess the presence of disease (NKF, 2002).

Often chronic kidney disease is discovered in patients at a later stage in which kidney transplants are necessary and mortality from cardiovascular disease or other related conditions is highly likely (Naghavi et al., 2015). It is important to detect the disease in individuals at an early stage so that treatments that delay the progression of chronic kidney disease can be applied. Additionally, if an underlying cause of chronic kidney disease is discovered, such as obstructive nephropathy, then that specific cause can be treated to slow the progression of CKD as well. In doing so, the overall mortality risk from this disease can be limited (Naghavi et al, 2015).

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 6: 2 Issues (2017)
Volume 5: 2 Issues (2016)
Volume 4: 2 Issues (2015)
Volume 3: 2 Issues (2014)
Volume 2: 2 Issues (2013)
Volume 1: 2 Issues (2012)
View Complete Journal Contents Listing