Abstract
This work aims to detect Alzheimer’s disease (AD) through non-linear features of speech signals. The features are extracted from signal subbands, which are obtained through the wavelet transform, and some of its descriptive statistics are used as input for several classifiers. Accuracies of 100, 77.8 and 85.2% are obtained in detecting AD among women, men and all, respectively, using logistic regression classifiers.
Original language | Portuguese |
---|---|
Title of host publication | Anais do XLI Simpósio Brasileiro de Telecomunicações e Processamento de Sinais |
Publisher | Sociedade Brasileira de Telecomunicações |
Pages | 1-5 |
Number of pages | 5 |
DOIs | |
Publication status | Published - 2023 |
Event | XLI Simpósio Brasileiro de Telecomunicações e Processamento de Sinais: SBrT 2023 - São José dos Campos, Brazil Duration: 8 Oct 2023 → 11 Oct 2023 |
Conference
Conference | XLI Simpósio Brasileiro de Telecomunicações e Processamento de Sinais |
---|---|
Country/Territory | Brazil |
City | São José dos Campos |
Period | 8/10/23 → 11/10/23 |
Keywords
- Alzheimer’s disease
- Speech signal
- Non-linear analysis
- Wavelet
- Classifiers