Abstract
Background
Early detection of mild cognitive impairment (MCI) is essential for initiating timely intervention and delaying progression to dementia. However, current diagnostic approaches often lack scalability and efficiency, limiting their utility in primary care and community settings.
Objective
This study developed and validated the Integrated Cognitive Screening Platform (ICSP), a self-administered, tablet-based tool designed for rapid, multidomain screening of MCI.
Methods
ICSP comprises five cognitive tasks involving immediate and delayed memory, attention, sensory perception, and executive function. Both accuracy scores and reaction times (RTs) were recorded and processed within five minutes. A total of 126 participants (76 with MCI, 50 cognitively normal controls) completed standard neuropsychological assessments and the ICSP battery. A multivariate logistic regression model was developed using 60% of the data as a training set and evaluated on the remaining 40% as a validation set.
Results
RTs and accuracy scores in sensory perception and executive function tasks, along with educational attainment, were identified as significant predictors of MCI. The model achieved high classification performance (AUC: 0.915; p < 0.001), with robust validation performance (AUC: 0.821; p < 0.001).
Conclusions
ICSP is an accurate and scalable digital screening tool capable of identifying MCI with high specificity. Its multimodal design and automated analysis make it well-suited for clinical and community-level early detection of cognitive impairments.
Trial registration
This study was registered with the Chinese Clinical Trial Registry (Registration No. ChiCTR2400082429) on March 28, 2024.
Keywords
Get full access to this article
View all access options for this article.
