Rough set analysis of electrostimulation test database for the prediction of post-operative profits in cochlear implanted patients

Abstract

A new method of examining the hearing nerve in deaf people has been developed at the Institute of Physiology and Pathology of Hearing in Warsaw. It consists in testing deaf people by speech signal delivered through a ball shaped microelectrode connected to the modulated current source and attached to the promontory area. The electric current delivered to the ball shaped electrode is modulated with real speech signal which is transposed downwards the frequency scale. A computer database of patients’ data and electrostimulation test results has been created. This database was analyzed using the rough set method in order to find rules allowing prediction of hearing recovery of cochlear implantation candidates. The Rough Set Class Library (RSCL) has been developed in order to implement data mining procedures to the engineered database of electrostimulation test results. The RSC Library supports symbolic approach to data processing. Additionally, the library is equipped with a set of data quantization methods that may be a part of an interface between external data environment and the rough set-based kernel of the system. The results of studies in the domain of prediction of post-operative profits of deaf patients based on the rough set analysis of electrostimulation test database are presented and discussed in the paper.