Title: Computer Aided Diagnosis for Chronic Kidney Disease Using Data Mining

Issue Number: Vol. 8, No. 3
Year of Publication: Sep - 2019
Page Numbers: 188-193
Authors: Mohammad Ashraf Ottom, Khalid M. Nahar
Journal Name: International Journal of Cyber-Security and Digital Forensics (IJCSDF)
- Hong Kong

Abstract:


Chronic Kidney Disease ( CKD) is a gradual disorder of kidney tasks over time. CKD early detection could reduce the impacts and harms by offering the necessary treatment. Nowadays, data mining can assist doctors to diagnose and predict diseases. In this paper, we shown that data mining techniques is a successful tool for diagnosing CKD using well know classification techniques such as naïve bays, and positively enhanced the diagnostic process. Since digital medical records could have redundant and unnecessary features, the paper also utilized features selection techniques to identify the most useful features that improves the diagnosis. The experiments showed that data mining techniques are capable of predicting and diagnosing CKD. In addition, features selection techniques such as CorrelationAttributeEval and CfsSubsetEval can assist to achieve better prediction accuracy. The experiments also showed that naïve bays classification technique performed better in collaboration with features selection techniques.