EEG-Based Classification of Stages of Alzheimer’s Disease (AD) and Mild Cognitive Impairment (MCI)

Alzheimer’s Disease (AD) accounts for 60-80% of dementia cases worldwide. According to the World Alzheimer Reports from Alzheimer’s Disease International, dementia is now the 7th leading cause of mortality around the world while also being one of the highest costs to society. Diagnosing Alzheimer’s Disease as early as possible is necessary to lessen the chances of the disease progressing to dementia and lessen the impacts - physical, physiological, social, and economic, on caregivers. No existing studies that use EEG as a modality for the analysis of changes in brain activity, have yet predicted the chances of patients with MCI symptoms eventually progressing to AD. Furthermore, none of these studies has characterized EEG signals of patients during the different stages of AD progression. In this study, we developed a machine learning model that can characterize electroencephalogram (EEG) signals and classify them accordingly for the detection and diagnosis of the different stages of AD. The proposed system was evaluated according to standard performance metrics. Upon performing cross-validation, our results show that the proposed system can accurately classify the stages of AD based on the patients’ recorded EEG signals. Future work can focus on testing the proposed system on a larger and more diverse population with varying demographics, genetic backgrounds, and disease sub-types to validate its effectiveness in the early detection of mild cognitive impairment (MCI) and Alzheimer’s disease (AD).