EEG-based biomarker for the identification of mild cognitive impairment via coherency analysis and iterative dynamic time warping.

Mild cognitive impairment (MCI) is an early stage of dementia, and its timely diagnosis is vital for preventing or delaying disease progression, which strongly impacts older adults' health and quality of life. Non-intrusive MCI screening can combine cognitive assessments with physiological biomarkers. However, studies on practical, non-invasive biomarkers suitable for daily life remain limited, highlighting the need for new indicators. This study proposes a novel biomarker derived from event-related potentials (ERPs). Electroencephalography was recorded from the prefrontal cortex during an auditory oddball task. ERPs were segmented into epochs based on standard and target stimuli. Iterative dynamic time warping was applied to reduce noise and correct epoch variability, including non-linear timing distortions. The imaginary part of beta-band coherency was then extracted from ERP epochs to form the biomarker. Data from 1,619 participants (1,142 cognitively normal and 477 with MCI) were analyzed. Results demonstrated significant differences (p < 0.001) between control and MCI groups. Additionally, the biomarker showed low correlation with previously studied biomarkers, indicating it may capture new aspects of MCI-related physiological changes. These findings suggest the biomarker could improve non-intrusive MCI screening and provide insights into underlying neural alterations, supporting the development of more accurate and convenient diagnostic tools.