An automatic screening approach for obstructive sleep apnea from photoplethysmograph using machine learning techniques

Smily Jeya Jothi E., Anitha J., D. Jude Hemanth

Abstract


Obstructive sleep apnea (OSA), a very common sleep disorder remains as an underdiagnosed root cause for several cardiovascular and cerebrovascular diseases. In this paper, we propose an efficient and accurate system that utilizes a single sensor for effective screening of OSA using machine learning algorithms. The automatic screening system involves a photoplethysmogram (PPG) signal, a novel algorithm to detect and remove the corrupted part of the signal, a feature extraction module to extract several features from the PPG waveform and a classifier module which helps in screening for OSA. The elemental idea behind this work is that there is a characteristic relationship between the shape of the PPG waveform and the oxygen desaturation in the apnea patients. The method as described was tested on 285 subjects, inclusive of both normal and apnea patients, and the results were obtained after 10-fold-cross validation of the different machine learning techniques viz., univariate regression, multivariate regression, support vector machine and random forest. The best results in screening OSA were obtained from random forest algorithm with the highest performance (Acc:98.0%, Sen:98.6%, Spec:99.3%) for all the combined features. The proposed work is an effective system for automatic screening of OSA from a single PPG sensor, thereby reducing the need for a very expensive and overnight polysomnography sleep study.

Keywords


multivariate regression; obstructive sleep apnea; photoplethysmogram; random forest; support vector machine; univariate regression;

Full Text:

PDF


DOI: http://dx.doi.org/10.12928/telkomnika.v19i4.19371

Article Metrics

Abstract view : 0 times
PDF - 0 times

Refbacks



Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

TELKOMNIKA Telecommunication, Computing, Electronics and Control
ISSN: 1693-6930, e-ISSN: 2302-9293
Universitas Ahmad Dahlan, 4th Campus
Jl. Ringroad Selatan, Kragilan, Tamanan, Banguntapan, Bantul, Yogyakarta, Indonesia 55191
Phone: +62 (274) 563515, 511830, 379418, 371120
Fax: +62 274 564604

View TELKOMNIKA Stats