Abstract:
In the advancement of the technology, humans have been used as a part in brain computing
interface field. BCI provides a communication pathway between wired brain and extraneous
devices using electroencephalography (EEG) signals. EEG is the signal technique in BCI field
which is used to track and record brain patterns from surface of the scalp using different
electrode locations which sends the signal to computer to record results. In the BCI research
field, Neuro-prosthetics applications are focused primarily, which purposes is to restoring
discredited hearing, visual perception and apparent movement of organs. Brain Machine
Interface has turned to a great research field that consists of many challenges in neurobiology.
Patients in a locked in syndrome (LIS) on account of wicked neurological disorders involve
unseamed emergency care by their caregivers or guardians. Nevertheless, it is a very hard job for
the guardians to endlessly monitor the patients’ state, particularly when there is no possibility of
direct communication. There have been existing research studies to enforce P300, Steady State
Visual Evoked Potential (SSVEP) and Hybrid speller applications which have not been validated
on patients with LIS diseases. Hence, their clinical value has not been validated. They worked on
spelling the characters, words, cursor movement, and spelling of numerical values.
In the present study, an Emergency Feedback System has been developed for such patients using
Hybrid approach which consists of combination of Steady State Visual Evoked Potential
(SSVEP) and P300 Event Related Potentials (ERPs). In order to affirm the usability of System
interface, three Healthy participants have been invited for experimentations. Further, three LIS
Patients have been invited to check the clinical feasibility of Emergency Feedback System. All
the Healthy participants and LIS Patients have succeeded in providing commands to our
Emergency Feedback System precisely. The average accuracy to perform a single session has
been reported 88.32% for all Healthy participants while the average accuracy to perform a single
session has been measured 85.66% for all LIS Patients. Beside these, the average time required
to execute a command has been measured 14 bpm for all Healthy participants while for the LIS
Patients, it has been reported 6 bpm per command.
