Study on real-time detection of alertness based on EEG

Human alertness varies in tasks requiring sustained attention. This can lead to severe consequences in occupations like car traffic operation, air traffic control and nuclear power plant monitoring and so on. EEG that is the electrical activity of brain can reflect the state of alertness. In this paper, the present work aims at real-time estimation of alertness from EEG signals. We made subjects perform a test of variables of attention (TOVA) and recorded their response time that was selected as a metric to quantify the subject's performance. Synchronously, we acquired the EEG signals of the subjects during the whole test. We studied the correlation between EEG power spectrum and response time, and used the power spectrum to construct the models of estimation alertness for single subjects by means of support vector machines method. The experimental results show the possibility of using EEG signals to real-time estimation alertness. We conclude that the method is helpful for the construct of the practical real-time detection system to alertness.