Background: Alzheimer's disease (AD) dementia and mild cognitive impairment (MCI) are currently underdiagnosed in the community, and early detection of cognitive deficits is crucial for timely intervention. FACEmemory®, the first completely self-administered online memory test with voice recognition, has been launched as an accessible tool to detect such deficits. This study aims to investigate the neuropsychological associations between FACEmemory subscores and cognitive composites derived from traditional paper-and-pencil neuropsychological tests and to develop an optimal algorithm using FACEmemory data and demographics to discriminate cognitively healthy (CH) individuals from those with MCI. Methods: A total of 669 participants (266 CH, 206 non-amnestic MCI [naMCI], and 197 amnestic MCI [aMCI]) were included. Multiple linear regression analyses were conducted using a cognitive composite as the dependent variable and FACEmemory subscores and demographic data (age, sex, and schooling) as independent variables. Machine learning models were compared to identify an optimal algorithm for distinguishing between CH and MCI (whole MCI, aMCI, and naMCI). Results: Multiple regression analyses showed associations between FACEmemory scores and the domains of memory (ρ = 0.67), executive functions (ρ = 0.63), visuospatial/visuoperceptual abilities (ρ = 0.55), language (ρ = 0.43), praxis (ρ = 0.52), and attention (ρ = 0.31). An optimal algorithm distinguished between CH and aMCI, achieving a FACEmemory cutoff score of 44.5, with sensitivity and specificity values of 0.81 and 0.72, respectively. Conclusions: FACEmemory is a promising online tool for identifying early cognitive impairment, particularly aMCI. It may contribute to addressing the underdiagnosis of MCI and dementia in the community and in promoting preventive strategies.
Keywords: Alzheimer’s disease; digital biomarkers; early detection; memory; mild cognitive impairment; new technologies.