Early diagnosis of Alzheimer's disease and mild cognitive impairment based on electroencephalography: From the perspective of event related potentials and deep learning.

Alzheimer's disease (AD), a neurodegenerative disorder characterized by progressive cognitive decline, is generally prevalent in elderly people with significant disability and mortality. There is no effective treatment for AD currently, but the early diagnosis might be beneficial for delaying the disease progression. Apart from invasive laboratory tests and expensive neuroimaging examination, the electroencephalography (EEG) and event related potentials (ERPs) have emerged as promising approaches for the early detection of AD as well as mild cognitive impairment (MCI), due to its affordability, noninvasively, and superior temporal resolution. In addition, the recent advent of deep learning architectures further improves the accuracy of AD and MCI diagnosis. This article reviewed the application of EEG signal for the early diagnosis of AD and MCI, especially focusing on ERPs and deep learning. Furthermore, recommendation for further research to recruit the combination of ERP components and deep leaning models in diagnosing AD and MCI was proposed and highlighted.