Create and analyze new audiology data set and using data mining techniques to predict hearing aid factors for audiology patients(field of bioinformatics and healthcare system)

Abstract

Clustering algorithm has been used in many studies to find patterns and relationships in audiology dataset. Employs clustering algorithm in grouping distinctive audiometric profiles of hearing-impaired patients. However, association analysis with Frequent Pattern Growth (FPGrowth) algorithm is more efficient in extracting related items in a dataset. Majority of these techniques are depending on components of conventional software that not mostly advanced for the field of audiology like SAS and SPSS. Despite the recent works upon audiology records and medical information in general, now a days actually no accessible of discovery appliance of hybrid knowledge enable to transact with aggregate of structured data, unstructured information, and audiograms as it found within our records of audiology data. In this research, we will present an approach of integrating supervised techniques and unsupervised clustering of audiology patient’s data with the identification of textual keywords associated with each cluster, in more precisely, those related to text comment, diagnosis and Hearing Aid (HA) type. This research deals with new specific data set which collects and analyze depends on audiology information and patient's diagnosis. It consists of 72 fields distributed on 71 fields for data details and one further for class. All fields are categorical with some of missing values. It was subjected to a very accurate analysis (before cleaning) based on the correct medical diagnosis and comprehensive information of the most important points that directly affect the selection of appropriate HA for Audiology Patient (AP), and via applying Data Mining (DM) techniques, we obtained a prediction of 100% for HA selection, and 98% to determine which power type of HA that those patients should use. To reach our goal, we examined DM techniques utilizing Python for coding and modeling.