Detection of event-related hemodynamic response to neuroactivation by dynamic modeling of brain activity

► A dynamic model of event-related hemodynamic activity is proposed. ► A simple means of generating hemodynamic predictor (basis) function is proposed. ► The method is computationally efficient in predicting brain tasks. ► It incorporates event-related intra- and inter-activation-dynamics of the tasks. ► It avoids delay and enhances the event-related features for peer BCI applications.