Study on real-time detection of alertness based on EEG

Bi Luzheng*, Zhang Ran, Chen Zhilong

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

14 Citations (Scopus)

Abstract

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.

Original languageEnglish
Title of host publication2007 IEEE/ICME International Conference on Complex Medical Engineering, CME 2007
Pages1490-1493
Number of pages4
DOIs
Publication statusPublished - 2007
Event2007 IEEE/ICME International Conference on Complex Medical Engineering, CME 2007 - Beijing, China
Duration: 23 May 200727 May 2007

Publication series

Name2007 IEEE/ICME International Conference on Complex Medical Engineering, CME 2007

Conference

Conference2007 IEEE/ICME International Conference on Complex Medical Engineering, CME 2007
Country/TerritoryChina
CityBeijing
Period23/05/0727/05/07

Fingerprint

Dive into the research topics of 'Study on real-time detection of alertness based on EEG'. Together they form a unique fingerprint.

Cite this