Research On Brain-Computer Interface For Clinical Application

Brian is the most important and complex organ of human.The patient with brain injury suffers from the disorder of consciousness and motor dysfunction.Some stroke patients suffered the motor dysfunction of body in contralateral of damaged brain.Patients with severe brain injury suffered from disorder of consciousness(DOC).Therefore,brain computer interface can be applied in the diagnosis and rehabilitation of patients with brain injury in the future.The main work of this thesis is about the research on using brain computer interface to assist clinical diagnosis of DOC and rehabilitation of motor function.Currently,the traditional way to evaluate consciousness in DOC patients is behavioral assessment,which highly dependent on the motor abilities.However,motor disability poses a significant challenge for evaluating of DOC patients.Brain computer interfaces(BCIs),which directly detect the brain response to the external stimulus,may provide a potential solution to complement classical behavioral or peripheral physiological observations.Therefore,three novel BCI systems for evaluating the auditory startle,visual fixation and visual pursuit were developed for DOC patients.Firstly,we explored the application of BCIs in assisting auditory startle assessment of DOC patients.An auditory EEG-based BCI system with an oddball paradigm was proposed to facilitate the evaluation of auditory startle in the Coma Recovery Scale-Revised(CRS-R).Nineteen DOC patients participated in the CRS-R and BCI assessments,of which sixteen patients' results were consistent in both assessments,and more importantly,other three patients who did not show behavioral response in the CRS-R assessment were responsive in the BCI assessment.These results revealed that the proposed BCI may provide more sensitive results than the CRS-R.Secondly,the proposed visual BCI was designed to assess visual fixation in DOC patients.Four buttons(brightly colored ball)were arranged at the edge of the graphic user interface(GUI)in four directions and one of them was pseudo-randomly chosen as the target.After the experiment started,a moving button in the center moved to the target location for guiding the patient.When the moving ball achieved and overlapped the target button,the four buttons flashed from the foreground to the background in a random order to elicit the P300.The results obtained from fifteen patients indicated that four out of fifteen DOC patients had a reliable visual fixation in BCI system,one of the four was not observed any visual fixation behavior by CRS-R.The other eleven patients did not show visual fixation in BCI assessment,including two patients showed higher visual function behavior in CRS-R.This phenomenon have discussed in the previous studies,many DOC patients failed to show visual fixating behavior in CRS-R,while higher cognitive functions existed.The BCI system needs to improve for better performance in detecting visual fixation.Next,a BCI system was designed to simulate the visual pursuit detection in CRS-R.The GUI included four buttons,the foreground of the buttons was produced by a realtime camera,whereas the backgrounds were pseudo-randomly chosen from 10 unfamiliar face images.One button was pseudo-randomly chosen as the target and moved from the center to the edge of the GUI,while the other three did not.Four buttons flashed in a random order when the target moving.The patients were prompted to follow the moving button.Based on the collected electroencephalography(EEG)data,the algorithm determined whether the patient focused on the moving target.Among the 14 DOC patients participated in the assessments based on the BCI system and the CRS-R,four patients exhibited visual pursuit,and three were nonresponsive in both assessments.More importantly,seven patients who did not exhibit visual pursuit in CRS-R were detected to be responsive to the moving target in the BCI assessment.Furthermore,five out of seven recovered consciousness to some degree and showed visual pursuit in the second CRS-R assessment.The proposed BCI system is better able to detect visual pursuit than the behavioral scale-based assessment.Finally,we designed the motor imagery-based brain computer interface for motor function training.A pilot study showed that the motor imagery classified accuracies of the stroke patients can not achieved the requirement of the brain computer interface control.Then,a hybrid BCI paradigm combining motor imagery(MI)and steady-state visually evoked potentials(SSVEPs)has been proposed to provide effective continuous feedback for motor imagery training.During the initial training sessions,subjects must focus on flickering buttons to evoke SSVEPs as they perform motor imagery tasks.The feedback of the hybrid BCI is based on hybrid features consisting of motor imageryand SSVEP-related brain signals.As the training progresses,the subjects can gradually decrease their visual attention to the flickering buttons,and the feedback is still effective.In this case,the feedback is mainly based on motor imagery.