Classifying Volition for Comatose Patients: a BCI perspective

Traditional clinical methods for diagnosis of disorders of consciousness are highly subjective and dependant on the patient's ability to move or speak. Objective tools mainly neuroimaging based validation of imagery paradigms akin to those used in brain computer interface applications have partially addressed this problem. But there is now widespread consensus that long term benefits to quality of life for this patient group can be better achieved at the bedside. This work is aimed to develop an objective portable tool at the bedside to establish the ability to respond to command without moving or speaking. A volitional paradigm with a classic motor imagery flavour but specifically designed to utilize the stimuli previously used in fMRI paradigm is developed for this purpose. The neural signatures computed using classifier performance indicate similarity to the areas of brain activation as revealed in the neuroimaging studies. The results show a consistent classifiability of the volitional tasks: 'imagine playing tennis' (IPT) and 'imagine spatial navigation' (ISN) with classification accuracies of more than 90%. The paradigms using these volitional tasks can be further used and implemented on selected patients for objectively detecting awareness.