Brain portion peeling from T2 axial MRI head scans using clustering and morphological operation

Automatic brain segmentation methods are useful but are computationally intensive tools in medical image processing. In this study, we propose a brain segmentation method based on clustering and morphological operation. Initially, from the middle slice of the whole volume, the brain portion is segmented and using this as a reference, brain portion in the remaining slices are extracted. Experiments using the proposed method on two T2 weighted data set show that the proposed method gives satisfactory results.