Apnea detection based on hidden Markov model kernel

Título:	Apnea detection based on hidden Markov model kernel
Autores/as:	Travieso, Carlos M. Alonso, Jesús B. Ticay-Rivas, Jaime R. Del Pozo-Baños, Marcos
Clasificación UNESCO:	3307 Tecnología electrónica
Palabras clave:	Apnea Detection, Hidden Markov Model, Kernel Building, Pattern Recognition, Non-linear Processing
Fecha de publicación:	2011
Editor/a:	0302-9743
Publicación seriada:	Lecture Notes in Computer Science
Conferencia:	5th International Conference on Nonlinear Speech Processing (NOLISP 2011) 5th International Conference on Nonlinear Speech Processing, NOLISP 2011
Resumen:	This work presents a new system to diagnose the syndrome of obstructive sleep apnea (OSA) that includes a specific block for the removal of Electrocardiogram (ECG) artifacts and the R wave detection. The system is modeled by ECG cepstral coefficients. The final decision is done with two different approaches. The first one is based on Hidden Markov Model (HMM), as classifier. On the other hand, another classification system is based on Support Vector Machines, being the parameterization based on the transformation of HMM by a kernel. Our results reached up to 98.67%.
URI:	http://hdl.handle.net/10553/44041
ISBN:	9783642250194
ISSN:	0302-9743
DOI:	10.1007/978-3-642-25020-0_10
Fuente:	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)[ISSN 0302-9743],v. 7015 LNAI, p. 71-79
Colección:	Actas de congresos

Vista completa