EEGNET for the Classification of Mild Cognitive Impairment

Mild cognitive impairment (MCI) denotes a stage of cognitive decline that could indicate an increased vulnerability to developing Alzheimer’s disease or other forms of dementia. Timely identification of MCI is crucial for intervention and monitoring, potentially delaying or preventing further cognitive decline. This study aims to devise a method for early MCI detection using electroencephalogram (EEG) signals. The EEGNet model, a specialized neural network architecture tailored for EEG data analysis, is employed. The EEG data undergoes preprocessing, involving the application of a bandpass filter to isolate relevant frequency components, and segmentation into shorter epochs. These epochs are subsequently classified as either HC or MCI using the EEGNet model. The model achieved a remarkable accuracy rate of 99%, underscoring the potential of leveraging EEG data to enhance diagnostic capabilities and enable early interventions for individuals at risk of cognitive decline.