Electrooculogram-aided intelligent sensing and high-performance communication control system for massive ALS individuals

Abstract

Neurodegenerative disease was one of the progressive diseases that affect the brain’s neurons and cause muscle dysfunction. There is a need to focus on these challenges for interfaces to communicate with others. We analyze three categories of subjects from 15 subjects between different age groups from 20 to 55 to investigate the efficiency to promote the human–computer interface using eye activities. Observed signals were estimated with root mean square for 11 tasks and classified with probabilistic neural network. In the experimental analysis, it determines that subjects belonging to young-aged group performance were appreciated and obtained the average mean classification accuracy of 94.00%, the subjects belonging to old-aged group performance were medium and obtained the average mean classification accuracy of 93.27%, and subjects belonging to amyotrophic lateral sclerosis (ALS) performance were low and obtained the average mean classification accuracy of 90.37%. The study examined that young-aged subjects’ performance was marginally high compared with the other two categories of subjects. Our study concluded that first maximum performances were obtained for young-aged subjects and secondary performance was attained for old-aged subjects, and ALS subjects obtained minimum performance compared to other subjects who participated in this analysis. Finally, our study concluded that designing intelligent sensing and high-performance communication control systems for massive ALS-affected subjects was minimal. They also need more training than other categories of subjects in this experiment.