Drowsiness detection for single channel EEG by DWT best m-term
     approximation

Silveira, Tiago da; Kozakevicius, Alice de Jesus; Rodrigues, Cesar Ramos

doi:10.1590/2446-4740.0693

Acessibilidade / Reportar erro

Brasil

Research on Biomedical Engineering

Español English

Brasil

Español English

sumário « anterior atual seguinte »

Sumário

Original Articles • Res. Biomed. Eng. 31 (2) • June 2015 • https://doi.org/10.1590/2446-4740.0693 copy

Drowsiness detection for single channel EEG by DWT best m-term approximation

Authorship SCIMAGO INSTITUTIONS RANKINGS

Introduction

In this paper we propose a promising new technique for drowsiness detection. It consists of applying the best m-term approximation on a single-channel electroencephalography (EEG) signal preprocessed through a discrete wavelet transform.

Methods

In order to classify EEG epochs as awake or drowsy states, the most significant m terms from the wavelet expansion of an EEG signal are selected according to the magnitude of their coefficients related to the alpha and beta rhythms.

Results

By using a simple thresholding strategy it provides hit rates comparable to those using more complex techniques. It was tested on a set of 6 hours and 50 minutes EEG drowsiness signals from PhysioNet Sleep Database yielding an overall sensitivity (TPR) of 84.98% and 98.65% of precision (PPV).

Conclusion

The method has proved itself efficient at separating data from different brain rhythms, thus alleviating the requirement for complex post-processing classification algorithms.

Signal processing; Drowsiness detection; Wavelet transform; Best m-term approximation; Frequency bands; DB2 Wavelet

Scale	Level (J = 12)	Level (J =10)	$f_{α}$ [Hz]
2	$j_{11}$	$j_{9}$	33.34
4	$j_{10}$	$j_{8}$	16.67
8	$j_{9}$	$j_{7}$	8.33
16	$j_{8}$	$j_{6}$	4.17
32	$j_{7}$	$j_{5}$	2.08
64	$j_{6}$	$j_{4}$	1.04
128	$j_{5}$	$j_{3}$	0.52

Brain rhythm	DWT Level (J=12)	$f_{α}$ [Hz], DB2
Beta [13-30Hz]	$j_{β} = j_{11}$	33.34
Alpha [8-13Hz]	$j_{α} = j_{9}$	8.33
Theta [4-8Hz]	$j_{θ} < j_{8}$	4.17

Patient	TP (%)	TN (%)	FP (%)	FN (%)	PPV (%)	TPR (%)
Sc4002e0	100	100	0	0	100	100
Sc4012e0	100	78.69	21.31	0	82.43	100
Sc4022e0	70	100	0	30	100	70
Sc4032e0	40	100	0	60	100	40
Sc4042e0	90	100	0	10	100	90
Sc4052e0	100	100	0	0	100	100
Sc4062e0	100	100	0	0	100	100
Sc4072e0	80	100	0	20	100	80
Sc4082e0	90	100	0	10	100	90
Sc4092e0	70	100	0	30	100	70
Sc4102e0	81.82	100	0	18.18	100	81.82
Sc4112e0	82.86	100	0	17.14	100	82.86
Sc4122e0	100	100	0	0	100	100
Total	84.98	98.36	1.64	15.02	98.65	84.98

Results	Sensitivity (%)	Specificity (%)	Accuracy (%)	Precision (%)
( Lin et al., 2010 Lin C-T, Chang C-J, Lin B-S, Hung S-H, Chao C-F, Wang I-J. A real-time wireless brain-computer interface system for drowsiness detection. IEEE Biomedical Circuits and Systems. 2010; 4(4):214-22. http://dx.doi.org/10.1109/TBCAS.2010.2046415. PMid:23853367. http://dx.doi.org/10.1109/TBCAS.2010.204... )	88.7	-	-	76.9
( Hu et al., 2013 Hu S, Peters B, Zheng G. Driver fatigue detection from electroencephalogram spectrum after electrooculography artifact removal. IET Intelligent Transport Systems. 2013; 7(1):105-13. http://dx.doi.org/10.1049/iet-its.2012.0045. http://dx.doi.org/10.1049/iet-its.2012.0... )	86	64	75	-
( Lawhern et al., 2013 Lawhern V, Kerick S, Robbins KA. Detecting alpha spindle events in EEG time series using adaptive autoregressive models. BMC Neuroscience. 2013; 14(101):1. http://dx.doi.org/10.1186/1471-2202-14-101. PMid:24047117. http://dx.doi.org/10.1186/1471-2202-14-1... )	94.2	98.4	-	72.8
This work	99.9	97.5	98.7	97.4

Sociedade Brasileira de Engenharia Biomédica Centro de Tecnologia, bloco H, sala 327 - Cidade Universitária, 21941-914 Rio de Janeiro RJ Brasil, Tel./Fax: (55 21)2562-8591 - Rio de Janeiro - RJ - Brazil
E-mail: rbe@rbejournal.org

Acompanhe os números deste periódico no seu leitor de RSS

[1] *e-mail: cesar@ieee.org