A Pilot Study on Visually Stimulated Cognitive Tasks for EEG-Based Dementia Recognition

In the status quo, dementia is yet to be cured. Precise diagnosis prior onset of symptoms can prevent rapid progression emerging cognitive impairment. Recent progress has shown that Electroencephalography (EEG) promising and cost-effective test facilitate detection neurocognitive disorders. However, most existing works have been using only resting-state EEG. The efficiencies EEG signals from various tasks, for classification, thoroughly investigated. this study, we designed four tasks engage different performances: attention, working memory, executive function. We investigated these by statistical analysis on both time frequency domains three classes human subjects: Dementia (DEM), Mild Cognitive Impairment (MCI), Normal Control (NC). also further evaluated classification performances two features extraction methods: Principal Component Analysis (PCA) Filter Bank Common Spatial Pattern (FBCSP). found memory related yielded good recognition in cases PCA FBCSP. Moreover, FBCSP with combination revealed best sensitivity 0.87 specificity 0.80. To our knowledge, first work concurrently several scores. Our results essential information design aid conducting experimental early diagnose patients.