Analysis of Alzheimer’s Disease Progression in Structural Magnetic Resonance Images

Structural magnetic resonance imaging (MRI) of the brain is an increasingly useful tool in the study of neurodegenerative diseases. MRI is currently the fastest developing medical imaging modality which is applied to an increasingly number of different medical diagnostic situations. Serial acquisition structural images of a subject’s brain are acquired over time offers opportunities to monitor the progression of tissue volume changes in fine detail at all anatomical locations. As a result, the analysis of structural MRI data has been an active area of image analysis research for many years especially in early diagnosis, tracking of disease progression, which makes it possible to investigate how, for instance, a patient responds to treatment. The aim of the paper is to investigate and analyze brain tissue changes in Alzheimer’s disease using nonrigid medical image registration and statistical analysis techniques. In our proposed approach, first the source and the target images are affinely registered to correct global differences between the images and then non-rigid registration based on free-form deformation using B-spline approximation is performed on the images. The resulting displacement values from the non-rigid registration are further investigated for deformation in the entire brain to detect typical deformation patterns. Finally, statistical method namely t-test is performed for analysis of the results. The initial results indicate that the tissue volume change in the brain occurs predominantly in the hippocampus of the brain. Key-Words: Magnetic resonance imaging, Image registration, Free-form deformation, Alzheimer’s disease, Statistical analysis, T-test.