ASSESSMENT OF ROBUST MRI BRAIN SEGMENTATION METHODS TOWARD DIAGNOSING OF ALZHEIMER'S DISEASE

Whole brain segmentation from MR image is a key process toward higher level processing required by computerized Alzheimer's diagnosis. Delineated structures serve as the domain for subsequent hippocampus and other tissue segmentation. This paper proposed a robust method which involves less intervention from the user, compared to the conventional 2D geometric figure, based on a Level Set. Both methods were benchmarked on the T1 MR images obtained from the Alzheimer's disease Neuroimaging Initiative (ADNI) database. The experimental results reported herein indicated that the proposed method can extract the whole brain structure more accurately than its counterpart