Automatic segmentation of cardiac short axis slices in perfusion

Segmentation of the myocardium in dynamic contrast enhanced MR short axis images is an important step towards the estimation of semi-quantitative or quantitative parameters to determine the perfusion to the tissue regions. The perfusion indices of the tissue are obtained by dividing the tissue into regions of interest and estimating perfusion to each region. A fast automatic segmentation method based on level sets has been developed that makes use of the spatial and temporal information available in the dynamic images. The algorithm is validated on cardiac data qualitatively and quantitatively by comparing against regional flow indices from manually segmented tissue regions