Development of sensitive markers that can detect Alzheimer’s Disease at its pre-dementia stages is of critical importance. We hypothesized that altered cytoarchitecture caused by AD pathology would lead to a subtle altered pattern of voxel intensities on MRI which could be detected by texture analysis. In a training set of 111 AD and 141 normal controls, we extracted a set of texture features from core regions affected by pathology – precuneus, posterior cingulate cortex, hippocampus – that could efficiently discriminate between the two groups(AUC=0.875, p<0.001) as well as predict MCI/AD conversion from the normal stage (AUC=0.716, p=0.031).
This abstract and the presentation materials are available to members only; a login is required.