Classification of dementia type using the brain-computer interface

Abstract This paper addresses the development of a dementia screening tool using character-input-type brain–computer Interface (BCI). A blinking letter board is presented to subject for each matrix by BCI, and keeping an eye on one character, character-gazing estimated based event-related potential P300 subject. In this experiment, instructed specify subsequently watch task character. Four sets are made, consisting five or six letters per The subjects include 53 elderly people in their 60 s 90 who were diagnosed with specific symptoms dementia. types Alzheimer’s type (AD), Lewy body dementia, as well mild cognitive impairment (MCI). relationship between four BCI features explained Kruskal–Wallis test multiple comparisons. Also, classified that closely related type. results obtained inputs classifier three outputs. classification rate groups was about 60%. Since (DLB) low, performed two groups, MCI AD. Furthermore, 80% confirmed.