Please use this identifier to cite or link to this item:
Title: Cepstrum Feature Selection for the Classification of Sleep Apnea-Hypopnea Syndrome based on Heart Rate Variability
Authors: Ravelo-Garcia, A. G. 
Navarro Mesa, J. L. 
Hernadez-Perez, E.
Martin-Gonzalez, S.I. 
Quintana-Morales, P. 
Guerra-Moreno, I. 
Julia-Serda, G.
UNESCO Clasification: 3314 Tecnología médica
Keywords: electrocardiography
medical signal processing
medical disorders
statistical analysis
Issue Date: 2013
Journal: Computers in Cardiology 
Conference: 40th Annual Meeting on Computing in Cardiology (CinC) 
Abstract: Cepstrum Coefficients are analyzed in order to study its performance in Sleep Apnea Hypopnea Syndrome (SAHS) screening. A forward feature selection technique is applied in order to know for one thing, what cepstrum parameters can extract better information about the influence of breath sleep disorder on the heart rhythm, and on the other hand, trying to detect apneas based on the RR series obtained from the electrocardiogram (EKG).70 ECG recordings from Computers in Cardiology Challenge 2000 are divided into a learning set and a test set of equal size. Each set consists of 35 recordings, containing a single ECG signal. Each recording includes a set of reference annotations, one for each minute, which indicates the presence or absence of apnea during that minute. These reference annotations were made by human experts on the basis of a complete polysomnography.Statistical classification methods based on Linear Discriminant Analysis (LDA) and Quadratic Discriminant Analysis (QDA) are applied to the classification of sleep apnea epochs. LDA presents a sensitivity of 64.1% and specificity of 90.2% (auc=0.87). QDA presents a sensitivity of 73.7% and specificity of 85.9% (auc=0.89). In both cases, contribution of the 4th coefficient related to Respiratory Sinus Arrhythmia plays an important role in SAHS detection.
ISSN: 0276-6574
Source: 2013 Computing In Cardiology Conference (Cinc)[ISSN 0276-6574],v. 40, p. 959-962, (2013)
Appears in Collections:Actas de congresos
Adobe PDF (1,15 MB)
Show full item record


checked on Feb 28, 2021

Page view(s)

checked on Feb 27, 2021


checked on Feb 27, 2021

Google ScholarTM



Export metadata

Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.