DIAGNOSIS OF ALZHEIMER'S DISEASE IN EARLY STAGE USING LINEAR TEXTURE ANALYSIS IN MRI IMAGES

The purpose of this study was to evaluate the potential of linear discriminant
analysis (LDA) and principal component analysis (PCA) in discriminating atrophy of Alzheimer's
disease in early stage and atrophy of aging using MRI images.
Materials & Methods: In general, 26 MRI images (13 Alzheimer and 13 elderly) were analyzed under
applied options and two texture features analysis methods: principal component analysis (PCA), linear
discriminant analysis (LDA) using MaZda software. The K-NN (K=1) classifier was used for features
resulting from PCA and LDA. The confusion matrix and Receiver operating characteristic (ROC) curve
were also calculated.
Results: Computer aim diagnosis is able to discriminate atrophy of Alzheimer's disease from atrophy
of normal aging.
Discussion: Our results indicated that texture analysis can be an auxiliary tool in diagnosing Alzheimer's
disease in early stages.