possible scalable solution for Early detection of prodromal phase of dementia based on EEG complexity

Abstract Mild cognitive impairment (MCI) is recognized as a predementia stage and important risk factor for Alzheimer's disease (AD). The electroencephaloram (EEG) signal reflects activity of the brain cortex cells, which is very complex, as well as brain structure and organization. Contrary to the classical (standard) Fourier-based analysis that presumes the signals' stationarity, the fractal and nonlinear analysis might be a better suited for EEG analysis allowing to early detect changes in such a complex dynamical system as the brain. The application of complex systems dynamics theory in physiology (physiological complexity) is connected to the stereotypy of disease. Certain levels of decomplexification are confirmed in healthy human ageing, but the levels of complexity characteristic for disease are distinct and can serve as a marker. In early detection of dementia risk nonlinear measures extracted from EEG as a proxy of the brain as a complex system are promising as accessible, accurate and potentially clinically useful biomarkers of dementia. Together with the use of wearables for health, this approach to early detection can be done out of the clinical setting improving the chances of increasing the quality of life in seniors.