Medical Diagnosis Classification Using WEKA

Open Access
Conference Proceedings
Authors: José MachadoNicolas LoriAna CoimbraFilipe MirandaAntónio Abelha

Abstract: The use of data mining techniques is not new—commonly it is used in various other industries, such as financial services, marketing and manufacturing. The main goal of data mining is to find patterns in a large dataset that yield insight and expertise. Thus, in terms of healthcare, data mining methods have a wide range of uses, including diagnosing cancers, pattern recognition and prognosti-cating patient health outcomes. Each patient's diagnosis at the University of Porto Hospital (Centro Hospitalar Universitário Universitário do Porto) has an ICD-10-CM code. This data can be used to build a predictive model to classify diag-nosis using secondary diagnosis. Three datasets were then created to be tested using data mining techniques. As a result, the algorithm that had the best perfor-mance was the Random Tree (99.8% corrected classified instances) using the third dataset with the five main diagnoses of each patient as parameters

Keywords: Data Mining, ICD-10-CM, Classification, WEKA

DOI: 10.54941/ahfe100881

Cite this paper: