Segmentation of the left venctricle from MRimages via snake models incorporating shape similarities

an novel shape similarity energy is constructed for segmentation of LV from MRIs

when extracting the epicardium of the Left Ventricle from MRIs, I assume that the epicardium is similar to the endocardium in shape. in order to make use of this similarity, I construct an energy to characterize this similarity, using the endocardium result as reference. in this code, I implement this shape similarity energy using the
initial contour, i.e., the endocardium result, as reference. in fact, this shape similarity can be used for arbitrary shape, and one
can employ it for other shapes and other applications.

this work was first presented in ICIP2006, and recently published in
CVIU2013, as follows,
Yuanquan Wang(*) and Yunde Jia. Segmentation of the left venctricle from MRimages via snake models incorporating shape similarities. ICIP’2006,pp. 213-216, Oct. 8-11,2006, USA
Yuwei Wu, Yuanquan Wang(*), Yunde Jia, Segmentation of the left ventricle in cardiac cine MRI using a shape-constrained snake model, Computer Vision and Image Understanding (CVIU),Vol.117,No.9,2013,Pp. 990–1003

Also, we tried shape constraints for LV segmentation in the following
paper, the shape is just a cirlce, no shape similarity.
Yuanquan Wang(*) and Yunde Jia. Segmentation of the left ventricle from cardiac MR Images based on degenerated minimal surface diffusion and shape priors. ICPR’2006, vol.4, pp.671-674

I further extend this similarity energy recently, but I am not preparing a paper
since I am too busy. if any one is interested, pls write to me,
we can collabrate on this new work, corresponding author is enough to me!