Using In-Situ Statistics and a Spatially-Aware Kernel for Longitudinal Neuroimaging Analysis

Abstract

This paper introduces a framework for analyzing longitudinal neuroimaging datasets. We address the problem of detecting subtle, short-term changes in neural structure that are indicative of cognitive decline and correlate with risk factors for Alzheimer's disease. Previous approaches have focused on separating populations with different risk factors based on gross changes, such as decreasing gray matter volume. In contrast, we introduce a new spatially-sensitive kernel that allows us to characterize individuals, as opposed to populations. We use this for both classification and regression, e.g., to predict changes in a subject's cognitive test scores from neuroimaging data alone. In doing so, this paper presents the first evidence demonstrating that very small changes in white matter structure over a two year period can predict change in cognitive function in healthy adults.