Diagnosis of Parkinson disease using the wavelet transform and MFCC and SVM classifier
| Autor: | Belhoussine Drissi Taoufiq, Ammoumou Abdelkrim, Benayad Nsiri, Zayrit Soumaya |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
business.industry
Computer science information science Wavelet transform Pattern recognition 02 engineering and technology Support vector machine Daubechies wavelet 03 medical and health sciences Svm classifier 0302 clinical medicine Wavelet Compression (functional analysis) 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Radial basis function Mel-frequency cepstrum Artificial intelligence business 030217 neurology & neurosurgery |
| Zdroj: | 2019 4th World Conference on Complex Systems (WCCS). |
| DOI: | 10.1109/icocs.2019.8930802 |
| Popis: | Purpose of this paper is to assess the performance of method centered on support vector machine (SVM) categorization of vocal recoding to distinguish between patients with Parkinson disease and healthy patients. We studied the state of 18 healthy patients and 20 affected patients and we proceeded this way: the compression of the vocal recording using the Daubechies wavelet transform (WT) and we extract the cepstral coefficients of the Mel Frequency Cepstral Coefficients (MFCC), then we use the SVM linear and Radial Basis Function RBF kernels. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|