Adaptable microsleep detection based on EOG signals: A feasibility study

The microsleeps (MS) cause many traffic and various other accidents. These states of extreme drowsiness could be prevented by automatic detection or prediction. With that in mind, the classifier of MS was designed in this study based on the EOG analysis. The algorithm was proposed to be able to adapt independently to each analysed EOG signal. Finally, it was tested on 39 MS episodes and compared with the method based on fix thresholding. We reached sensitivity 82 % and positive predictivity 67 % by using the presented approach. It is necessary to extend the dataset in a successive study.