Optimized feature subsets for epileptic seizure prediction studies.
Annu Int Conf IEEE Eng Med Biol Soc
; 2011: 1636-9, 2011.
Article
em En
| MEDLINE
| ID: mdl-22254637
ABSTRACT
The reduction of the number of EEG features to give as inputs to epilepsy seizure predictors is a needed step towards the development of a transportable device for real-time warning. This paper presents a comparative study of three feature selection methods, based on Support Vector Machines. Minimum-Redundancy Maximum-Relevance, Recursive Feature Elimination, Genetic Algorithms, show that, for three patients of the European Database on Epilepsy, the most important univariate features are related to spectral information and statistical moments.
Texto completo:
1
Coleções:
01-internacional
Base de dados:
MEDLINE
Assunto principal:
Convulsões
/
Algoritmos
/
Reconhecimento Automatizado de Padrão
/
Diagnóstico por Computador
/
Eletroencefalografia
/
Máquina de Vetores de Suporte
Tipo de estudo:
Diagnostic_studies
/
Evaluation_studies
/
Prognostic_studies
/
Risk_factors_studies
Limite:
Humans
Idioma:
En
Revista:
Annu Int Conf IEEE Eng Med Biol Soc
Ano de publicação:
2011
Tipo de documento:
Article
País de afiliação:
Portugal