Ioannis Kosmidis

The paper introduces penalized generalized estimating equations for relative risk regression to address instability in modeling brain lesion data, showing improved convergence and interpretation.

The paper compares voxel-wise and spatial modelling approaches for binary lesion masks using a simulation framework and finds bias-reduced voxel-wise GLMs are accurate and efficient.

This study investigates the spatial distribution of white matter hyperintensities (WMHs) related to cerebrovascular risk factors and their impact on cognitive function in a large cohort.