Predict the type of hearing aid of audiology patients using data mining techniques

Abstract

Our research transacts with a great various audiology data from National Health System (NHS) facility, including audiograms, structured data such as age, gender, and diagnosis, and a text of specific information about each patient, i.e., clinical reports. This research examines factors related to audiology patients depends on various data by using the mining and analysis of this data. This paper looks for factors affecting the choice between two prevalent hearing aid kinds: BTE (Behind The Ear) or ITE (In The Ear). This choice often done by audiology technicians working in specific clinics for this purpose, based on audiograms results and patient consultation. In many situations, there is an obvious choice, but sometimes the technicians need for the second opinion via an automatic system includes clarification of how to obtain that second opinion. The research deals with diversified specifics and more significant factors for choosing of confirmed hearing aid related to those specifics. We depend on the earlier study data (Bareiss, E. Ray, & Porter, Bruce (1987)). Protos: An Exemplar-Based Learning Apprentice. In the Proceedings of the 4th International Workshop on Machine Learning, 12-23, Irvine, California, which illustrates the database analysis for 180,000 records, for 23,000 patients, by the hearing aid clinic at James Cook University Hospital in Middlesbrough, UK. This data mined to find which factors contribute to the deduction to fit a BTE hearing aid as opposed to an ITE hearing aid. Here we conduct some enhancements on this database and analyze the data depends on medical information to create a new class then we use some intelligent Data Mining (DM) techniques to guess the most correct illness that could be associated with patient's information. Based on the result (according to the patients' diagnosis details), we can obtain right predictions of which type of Hearing Aid (HA) they should use.