Abstract This paper examines the question of how strongly the spectral properties of the EEG during microsleep differ between individuals. For this purpose, 3859 microsleep examples were compared with 4044 counterexamples in which drivers were very drowsy but were able to perform the driving task. Two types of signal features were compared: logarithmic power spectral densities and entropy measures of wavelets coefficient series. Discriminant analyses were performed with the following machine learning methods: support-vector machines, gradient boosting, learning vector quantization. To the best of our knowledge, this is the first time that results of the leave-one-subject-out cross-validation (LOSO CV) for the detection of microsleep are presented. Error rates lower than 5.0 % resulted in 17 subjects and lower than 13 % in another 11 subjects. In 3 individuals, EEG features could not be explained by the pool of EEG features of all other individuals; for them, detection errors were 15.1 %, 17.1 %, and 27.0 %. In comparison, cross validation by means of repeated random subsampling, in which individuality is not considered, yielded mean error rates of 5.0 ± 0.5 %. A subsequent inspection of raw EEG data showed that in two individuals a bad signal quality due to poor electrode attachment could be the cause and in one individual a very unusual behavior, a high and long-lasting eyelid activity which interfered the recorded EEG in all channels.
[1]
AUTOMATIC RECOGNITION OF MICROSLEEP EVENTS
,
2004
.
[2]
P. R. Davidson,et al.
Detecting Behavioral Microsleeps from EEG Power Spectra
,
2006,
2006 International Conference of the IEEE Engineering in Medicine and Biology Society.
[3]
D. Sommer,et al.
Evaluation of fatigue monitoring technologies
,
2010
.
[4]
Jianhua Z. Huang,et al.
Asymptotic optimality and efficient computation of the leave-subject-out cross-validation
,
2012,
1302.4607.
[5]
M. Golz,et al.
The role of expert evaluation for microsleep detection
,
2015
.
[6]
Tie-Yan Liu,et al.
LightGBM: A Highly Efficient Gradient Boosting Decision Tree
,
2017,
NIPS.