EMD and PCA for the prediction of sleep apnoea: a comparative study

H.J. Robertson, J.J. Soraghan, C. Idzikowski, B.A. Conway

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

3 Citations (Scopus)

Abstract

A sleep apnoea episode prediction system is presented that is based exclusively on the airflow signal. Detection of obstructive sleep apnoea (OSA) is generally carried out using polysomnography, with the data being analysed and a diagnosis formed. Being able to predict when a sleep apnoea episode is going to occur will allow for treatment to be applied before the episode becomes detrimental to the patient. Airflow signals were extracted from polysomnographic data and processed using three techniques: epoching of the flow signal, principle component analysis (PCA) and empirical mode decomposition (EMD). These processed signals were then classified using three distance functions: Euclidean, Hamming and Spearman distance. Classification of the airflow signal preceding an apnoea by Hamming distance produced the best results, with sensitivity of 81% and specificity of 76%. Reliability statistics were increase when classifying apnoea and hypopnoea episodes, with sensitivity of 95% and specificity of 100%, using Hamming distance and the empirical mode decomposition. In conclusion, classification of inspiratory airflow signal before an apnoea and hypopnoea is possible with high reliability statistics.

LanguageEnglish
Title of host publication2007 IEEE International Symposium on Signal Processing and Information Technology
Place of PublicationPiscataway, NJ
PublisherIEEE
Pages419-424
Number of pages6
ISBN (Print)9781424418343
DOIs
Publication statusPublished - 26 Feb 2008
Event2007 IEEE International Symposium on Signal Processing and Information Technology - Cairo, Egypt
Duration: 15 Dec 200718 Dec 2007

Conference

Conference2007 IEEE International Symposium on Signal Processing and Information Technology
Abbreviated titleISSPIT 2007
CountryEgypt
CityCairo
Period15/12/0718/12/07

Fingerprint

Hamming distance
Decomposition
Statistics
Sleep

Keywords

  • empirical mode decomposition
  • hypopnoea
  • inspiratory flow
  • obstructive sleep apnoea
  • prediction
  • principle component analysis

Cite this

Robertson, H. J., Soraghan, J. J., Idzikowski, C., & Conway, B. A. (2008). EMD and PCA for the prediction of sleep apnoea: a comparative study. In 2007 IEEE International Symposium on Signal Processing and Information Technology (pp. 419-424). Piscataway, NJ: IEEE. https://doi.org/10.1109/ISSPIT.2007.4458166
Robertson, H.J. ; Soraghan, J.J. ; Idzikowski, C. ; Conway, B.A. / EMD and PCA for the prediction of sleep apnoea : a comparative study. 2007 IEEE International Symposium on Signal Processing and Information Technology. Piscataway, NJ : IEEE, 2008. pp. 419-424
@inproceedings{dce8e1c5f58b48b7a3003526c04fd7e1,
title = "EMD and PCA for the prediction of sleep apnoea: a comparative study",
abstract = "A sleep apnoea episode prediction system is presented that is based exclusively on the airflow signal. Detection of obstructive sleep apnoea (OSA) is generally carried out using polysomnography, with the data being analysed and a diagnosis formed. Being able to predict when a sleep apnoea episode is going to occur will allow for treatment to be applied before the episode becomes detrimental to the patient. Airflow signals were extracted from polysomnographic data and processed using three techniques: epoching of the flow signal, principle component analysis (PCA) and empirical mode decomposition (EMD). These processed signals were then classified using three distance functions: Euclidean, Hamming and Spearman distance. Classification of the airflow signal preceding an apnoea by Hamming distance produced the best results, with sensitivity of 81{\%} and specificity of 76{\%}. Reliability statistics were increase when classifying apnoea and hypopnoea episodes, with sensitivity of 95{\%} and specificity of 100{\%}, using Hamming distance and the empirical mode decomposition. In conclusion, classification of inspiratory airflow signal before an apnoea and hypopnoea is possible with high reliability statistics.",
keywords = "empirical mode decomposition, hypopnoea, inspiratory flow, obstructive sleep apnoea, prediction, principle component analysis",
author = "H.J. Robertson and J.J. Soraghan and C. Idzikowski and B.A. Conway",
year = "2008",
month = "2",
day = "26",
doi = "10.1109/ISSPIT.2007.4458166",
language = "English",
isbn = "9781424418343",
pages = "419--424",
booktitle = "2007 IEEE International Symposium on Signal Processing and Information Technology",
publisher = "IEEE",

}

Robertson, HJ, Soraghan, JJ, Idzikowski, C & Conway, BA 2008, EMD and PCA for the prediction of sleep apnoea: a comparative study. in 2007 IEEE International Symposium on Signal Processing and Information Technology. IEEE, Piscataway, NJ, pp. 419-424, 2007 IEEE International Symposium on Signal Processing and Information Technology, Cairo, Egypt, 15/12/07. https://doi.org/10.1109/ISSPIT.2007.4458166

EMD and PCA for the prediction of sleep apnoea : a comparative study. / Robertson, H.J.; Soraghan, J.J.; Idzikowski, C.; Conway, B.A.

2007 IEEE International Symposium on Signal Processing and Information Technology. Piscataway, NJ : IEEE, 2008. p. 419-424.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - EMD and PCA for the prediction of sleep apnoea

T2 - a comparative study

AU - Robertson, H.J.

AU - Soraghan, J.J.

AU - Idzikowski, C.

AU - Conway, B.A.

PY - 2008/2/26

Y1 - 2008/2/26

N2 - A sleep apnoea episode prediction system is presented that is based exclusively on the airflow signal. Detection of obstructive sleep apnoea (OSA) is generally carried out using polysomnography, with the data being analysed and a diagnosis formed. Being able to predict when a sleep apnoea episode is going to occur will allow for treatment to be applied before the episode becomes detrimental to the patient. Airflow signals were extracted from polysomnographic data and processed using three techniques: epoching of the flow signal, principle component analysis (PCA) and empirical mode decomposition (EMD). These processed signals were then classified using three distance functions: Euclidean, Hamming and Spearman distance. Classification of the airflow signal preceding an apnoea by Hamming distance produced the best results, with sensitivity of 81% and specificity of 76%. Reliability statistics were increase when classifying apnoea and hypopnoea episodes, with sensitivity of 95% and specificity of 100%, using Hamming distance and the empirical mode decomposition. In conclusion, classification of inspiratory airflow signal before an apnoea and hypopnoea is possible with high reliability statistics.

AB - A sleep apnoea episode prediction system is presented that is based exclusively on the airflow signal. Detection of obstructive sleep apnoea (OSA) is generally carried out using polysomnography, with the data being analysed and a diagnosis formed. Being able to predict when a sleep apnoea episode is going to occur will allow for treatment to be applied before the episode becomes detrimental to the patient. Airflow signals were extracted from polysomnographic data and processed using three techniques: epoching of the flow signal, principle component analysis (PCA) and empirical mode decomposition (EMD). These processed signals were then classified using three distance functions: Euclidean, Hamming and Spearman distance. Classification of the airflow signal preceding an apnoea by Hamming distance produced the best results, with sensitivity of 81% and specificity of 76%. Reliability statistics were increase when classifying apnoea and hypopnoea episodes, with sensitivity of 95% and specificity of 100%, using Hamming distance and the empirical mode decomposition. In conclusion, classification of inspiratory airflow signal before an apnoea and hypopnoea is possible with high reliability statistics.

KW - empirical mode decomposition

KW - hypopnoea

KW - inspiratory flow

KW - obstructive sleep apnoea

KW - prediction

KW - principle component analysis

UR - http://www.scopus.com/inward/record.url?scp=71549161198&partnerID=8YFLogxK

UR - https://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=4455349

U2 - 10.1109/ISSPIT.2007.4458166

DO - 10.1109/ISSPIT.2007.4458166

M3 - Conference contribution book

SN - 9781424418343

SP - 419

EP - 424

BT - 2007 IEEE International Symposium on Signal Processing and Information Technology

PB - IEEE

CY - Piscataway, NJ

ER -

Robertson HJ, Soraghan JJ, Idzikowski C, Conway BA. EMD and PCA for the prediction of sleep apnoea: a comparative study. In 2007 IEEE International Symposium on Signal Processing and Information Technology. Piscataway, NJ: IEEE. 2008. p. 419-424 https://doi.org/10.1109/ISSPIT.2007.4458166