Early Detection of Alzheimer's Disease through Analysis of EEG Responses to Word Recognition

Abstract

Early detection is crucial in addressing Alzheimer's disease (AD). Although regular testing has proven effective, the invasiveness and associated costs pose challenges for many. Therefore, we developed a classification method which utilizes electroencephalography (EEG). This study examined 20 patients with subjective cognitive decline (SCD) and 10 with AD. During an immediate word retrieval task, we recorded EEG signals and extracted defining features from the presentation of old words. Our findings reveal an 86% test accuracy and 91% sensitivity in the classification of patients with SCD and AD, utilizing features extracted from events involving old words. In the case of features extracted from events involving new words, we achieved an 57% test accuracy and 73% sensitivity. This study contributes to the early detection of AD, potentially facilitating the initiation of treatment.