A0184
Title: Statistical modeling issues in brain age prediction
Authors: Fengqing Zhang - Drexel University (United States) [presenting]
Abstract: Brain age prediction based on neuroimaging data and machine learning models has emerged as a promising approach for characterizing typical brain development and neuropsychiatric disorders. However, few studies examine multi-modal imaging features derived from MRI, DTI as well as rs-fMRI for brain age prediction. In addition, several studies report that the predicted brain age is underestimated for older subjects and overestimated for younger subjects. We examine this systematic bias and propose different approaches to correct for the bias. We also compare different machine learning models to integrate different combinations of multi-modal imaging features. Current methods of brain age prediction provide a single value representing the whole brains average developmental or ageing status. This approach might miss the divergent patterns of change in various brain structures. We, therefore, propose a novel multidimensional brain-age index. The proposed methods are evaluated using large neuroimaging data sets.
