Epileptic seizure endorsement technique using DWT power spectrum

• Tipo: Artigo em Revista Científica Internacional
• Data: 2022-04

Autores

Anand Ghuli (Autor) Outra; Damodar Reddy Edla (Autor) Outra; João Manuel R. S. Tavares (Autor) FEUP

Revista

Título: Journal of Supercomputing

Vol. 78 Nº 6

Páginas: 8604-8624

ISSN: 0920-8542

Editora: Springer Nature

Indexação

ISI Web of Knowledge - 0 Citações

ISI Web of Science

Scopus - 0 Citações

Clarivate Analytics

Classificação Científica

CORDIS: Ciências Tecnológicas

FOS: Ciências médicas e da saúde

Outras Informações

ID Authenticus: P-00V-ZFB

DOI: 10.1007/s11227-021-04196-3

Abstract (EN): EEG signals play significant role in the study of mental disorders. Epilepsy is one of the major mental disorders and need significant technological support in the treatment. A method proposed here is an endorsement technique for epileptic seizures using electroencephalogram (EEG) signals captured using non-invasive method. The method uses power spectrum density and discrete wavelet transformation (DWT). The impact of power spectral analysis along with the usage of EEG characteristics in endorsement of epilepsy is addressed here. A publicly available EEG epileptic dataset is processed using FIR filters along with DWT. The power spectrum density and its average were compared with specific spectrum to get the results and were compared against the standard EEG signal frequency range. It is found that the usage of DWT is more accurate and reliable to process and classify the EEG data for epilepsy endorsement.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 21