Using novel method: Real Cepstral Discrete Cosine Transform, for detecting Parkinson from multiple system atrophy, other neurological diseases and healthy cases using voice analysis
| Autor: | Achraf Benba, Ahmed Hammouch, Abdelilah Jilbab, Imane Laaqira |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Linguistics and Language
Computer science Speech recognition Early detection medicine.disease Language and Linguistics nervous system diseases Voice analysis Human-Computer Interaction Svm classifier Atrophy Cepstrum medicine Discrete cosine transform Computer Vision and Pattern Recognition Mel-frequency cepstrum Software |
| Zdroj: | International Journal of Speech Technology. 25:163-172 |
| ISSN: | 1572-8110 1381-2416 |
| DOI: | 10.1007/s10772-021-09896-y |
| Popis: | This paper seeks to detect Parkinson’s disease among healthy cases, several neurological diseases, in particular those which are very similar, that is to say representing a parkinsonian syndrome such as multi system atrophy. Early detection based on the phonatory symptoms will offer a possibility to the treatments proposed by the doctors to act effectively on the patient hence the interest of our project. We used several algorithms such as MFCC, PLP and RASTA PLP in addition to our new technique called Real Cepstral Discrete Cosine Transform with an SVM classifier with its different kernels to guarantee a better detection. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink