The white matter network in cognitive normal elderly predict the rate of cognitive decline

Weijie Huang,Ni Shu

DOI: https://doi.org/10.1101/2020.04.16.044479

2020-01-01

Abstract:White matter degradation has been proposed as one possible explanation for age-related cognitive decline. The human brain is, however, a network and it may be more appropriate to relate cognitive functions to properties of the network rather than specific brain regions. Cognitive domains were measured annually (mean follow-up = 1.25 ± 0.61 years), including processing speed, memory, language, visuospatial, and executive functions. Diffusion tensor imaging was performed at baseline in 90 clinically normal older adults (aged 54–86). We report on graph theory-based analyses of diffusion tensor imaging tract-derived connectivity. The machine learning approach was used to predict the rate of cognitive decline from white matter connectivity data. The reduced efficacy of white matter networks could predict the performance of these cognitive domains except memory. The predicted scores were significantly correlated with the real scores. For the local regions for predicting the cognitive changes, the right precuneus, left inferior parietal lobe and cuneus are the most important regions for predicting monthly change of executive function; some left partial and occipital regions are the most important for the changed of attention; the right frontal and temporal regions are the most important for the changed of language. Our findings suggested that the global white matter connectivity characteristics are the valuable predictive index for the longitudinal cognitive decline. For the first time, topological efficiency of white matter connectivity maps which related to special domains of cognitive decline in the elderly are identified.