Statistical features and voxel-based morphometry for Alzheimer's disease classification

Abstract

Alzheimer's disease causes progressive decline in the mental abilities that usually starts with memory loss and ends with cognitive and behavioral disorders. Studying the biomarkers found in structural MRI can help detecting early changes in the brains of people at high risk for developing alzheimer's disease. This work presents an image analysis technique for the prediction of alzheimer's disease. The technique combines texture features extracted from gray level co-occurrence matrix and voxel-based morphometry neuroimaging analysis to classify alzheimer's disease patients by the means of support vector machine classifier. Feature selection using entropy is applied to overcome the curse of dimensionality. The proposed technique is applied on gray matter tissues, and managed successfully to achieve accuracy of 88% for differentiating between alzheimer's disease patients and normal controls.