Volumetric analysis on MRI and PET images for the early diagnosis of Alzheimer's disease

In this paper we present the development of a software for the extraction of the hippocampus and surrounding medial-temporal-lobe (MTL) regions from T 1-weighted magnetic resonance (MR) and from Positron Emission Tomography (PET) images with no interactive input from the user. With this software we introduce a novel statistical index computed on the intensities in the automatically extracted MTL regions. This index is a measure of gray-matter (G M) atrophy and allows to: distinguish between (a) patients with Alzheimer's disease (A D), patients with amnestic mild cognitive impairment (aMCI), (b) patients with amnestic mild cognitive impairment who will later develop A D in a time frame of 2 years (aM-CIconv), and (c) a set of age-matched elderly controls. Once refined, this method could be used to infer about the clinical outcome of aMCI patients.