A machine learning approach for potential Super-Agers identification using neuronal functional connectivity networks

Abstract

Aging is often associated with cognitive decline. Understanding neural factors that distinguish adults in midlife with superior cognitive abilities (Positive-Agers) may offer insight into how the aging brain achieves resilience. This study aims to (1) introduce an optimal labeling mechanism to distinguish between Positive-Agers and Cognitive Decliners, and (2) identify Positive-Agers using neuronal functional connectivity networks data and demographics. Principal component analysis initially created latent cognitive trajectories groups. A hybrid algorithm of machine learning and optimization was then designed to predict latent groups using neuronal functional connectivity networks derived from resting state functional MRI. The study achieved accurate Positive-Ager identification using rsfMRI and demographic data (AUC = 0.88), and found that the posterior default mode network has the highest importance.