Deep Grey Matter Index in Alzheimer's Brain MR Images using Minkowski Functionals

摘要

In this work, an attempt is made to analyse morphology changes in the selected ROI around the deep grey matter and the ventricle using Minkowski Functionals (MFs) in Alzheimer's brain MR images. The normal and abnormal images considered in this work are obtained from MIRIAD database. Initially, the ROI is selected and subjected to analysis using MFs such as area, perimeter and Euler number. The Deep grey matter MF2D index is formulated and is correlated with the Mini-Mental State Examination (MMSE) score. Results show that the MF2D index of deep grey matter and ventricle is distinct for both normal and varying Alzheimer Disease (AD) conditions. It is also observed that the deep grey matter MF2D index provide better discrimination of normal and abnormal subjects. Its correlation with MMSE is observed to be 0.79 for normal and 0.91 for severe Alzheimer subjects. Deep grey matter MF2D index which takes in account of memory related atrophy and the global atrophy of ventricle can be used for study of neurodegenerative disease progression.