Medical image segmentation via coupled curve evolution equations with global constraints

In this work we modify the couple d curve evolution approach to snakes presente dby the authors in pre-vious work for bimodal and trimodal imagery through the intr oduction of glob al constr aints. The key idea, as b efor e, is to derive curve evolution equations which \pull apart " the values of one or more statistics within the image. However, by imposing a new constraint on the evolution of these statistics, we are able to seg-ment a larger class of medical imagery for which our original model would fail.