Detection of Alzheimer Disease Using Grow Cut Method and Brain Volume Estimation

The loss of neurons in the brain is one major indication of Alzheimer"s disease(AD). Voxel based MRI studies have demonstrated that reduction in brain volume shows the atropic changes in mild cognitive impairment (MCI) preceding the onset of AD type dementia. In many cases, medical experts use the magnetic resonance image(MRI) to quantitatively measure the loss of neurons by shrinkage of structure of interest in brain. Moreover, atrophy in any particular structure will directly lead to reduction in overall size of brain. This paper proposes a new method for the early detection of AD algorithm based on the volume estimation of images. Training processes are performed to generate two training classes, healthy control(HC) and AD patients. The volume of brain using segmentation of brain MRIs are estimated. Finally, this paper utilizes the advantages of grow cut method to extract the brain volume from MRI accurately. The proposed method is simple yet efficient and the experimental results have shown its superiority.