Please use this identifier to cite or link to this item: http://scholarbank.nus.edu.sg/handle/10635/35582
Title: OBSTRUCTIVE SLEEP APNEA DIAGNOSIS WITH APNEA EVENT DETECTION IN SNORING SOUND USING A CONDITIONAL RANDOM FIELD
Authors: HE LIAN
Keywords: obstructive sleep apnea, snoring sound, conditional random field, PSG, apnea, hypopnea
Issue Date: 30-Jul-2012
Source: HE LIAN (2012-07-30). OBSTRUCTIVE SLEEP APNEA DIAGNOSIS WITH APNEA EVENT DETECTION IN SNORING SOUND USING A CONDITIONAL RANDOM FIELD. ScholarBank@NUS Repository.
Abstract: Obstructive Sleep Apnea (OSA) has become increasingly prevalent throughout the world in recent decades, but its proper diagnosis is severely constrained by the limited accessibility of polysomnography (PSG) facilities. To resolve this problem, researchers investigated the potential of OSA diagnosis by using snore-related sounds. However, most existing approaches to OSA diagnosis analyze snore episodes or silence episodes individually. In this thesis, we propose a method to identify apnea events by incorporating ISPJ and F1 lables and learning the relation among these sequential acoustic signal components using a conditional random field. Compared with three existing methods, the proposed method exhibits the best performance by achieving a sensitivity of 92.31% and a specificity of 80% under the threshold of apnea index set to 5. Moreover, the number of apnea events detected by our approach effectively approximates the actual one reported by PSG, which makes the proposed method a potential alternative for manual annotation. Based on the proposed method, a prototype named Mobile Obstructive Sleep Apnea Diagnosis is implemented on a mobile device. Validation results demonstrate the prototype¿s effectiveness and efficiency. The efficacy and portability of our system illustrate its promising potential for OSA screening in a home environment.
URI: http://scholarbank.nus.edu.sg/handle/10635/35582
Appears in Collections:Master's Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Helian_MasterThesis.pdf1.56 MBAdobe PDF

OPEN

NoneView/Download

Page view(s)

248
checked on Dec 11, 2017

Download(s)

52
checked on Dec 11, 2017

Google ScholarTM

Check


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