Classification and diagnosis of the parkinson disease by stacked autoencoder

Parkinson disease is neuro-degenerative motion disorder this disease causes voice and speech dysfunctions. Therefore, disease can be diagnosed by determining dysphonia produced by a subject with parkinson diseases. In this paper we have introduced a new classification method of the parkinson disease...

Full description

Saved in:
Bibliographic Details
Published in2016 National Conference on Electrical, Electronics and Biomedical Engineering (ELECO) pp. 499 - 502
Main Authors Badem, Hasan, Caliskan, Abdullah, Basturk, Alper, Yuksel, Mehmet Emin
Format Conference Proceeding
LanguageEnglish
Published The Chamber of Turkish Electrical Engineers 01.12.2016
Subjects
Dogs
Machine learning
Neural networks
Parkinson's disease
Random access memory
Speech
Online AccessGet full text

Cover

More Information
Summary:Parkinson disease is neuro-degenerative motion disorder this disease causes voice and speech dysfunctions. Therefore, disease can be diagnosed by determining dysphonia produced by a subject with parkinson diseases. In this paper we have introduced a new classification method of the parkinson disease which is based on the stacked autoencoder. The method is tested with two data sets and results show that the performance of the stacked autoencoder is significantly better than the-state-art-methods.