Topology Adaptive Deformable Surfaces for Medical Image Volume Segmentation.pdfVIP

840 IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 18, NO. 10, OCTOBER 1999 Topology Adaptive Deformable Surfaces for Medical Image Volume Segmentation Tim McInerney,* Member, IEEE , and Demetri Terzopoulos, Member, IEEE Abstract— Deformable models, which include deformable con- inherent connectivity and smoothness that counteracts noise tours (the popular snakes) and deformable surfaces, are a power- and boundary irregularities, and the ability to incorporate a ful model-based medical image analysis technique. We develop a priori anatomic knowledge. Addressing the fact that few med- new class of deformable models by formulating deformable sur- faces in terms of an affine cell image decomposition (ACID). Our ical images lend themselves to fully automatic analysis with approach significantly extends standard deformable surfaces, satisfactory results, deformable models furthermore afford the while retaining their interactivity and other desirable properties. often times essential option of interactive control over the In particular, the ACID induces an efficient reparameterization segmentation process. They support interaction mechanisms mechanism that enables parametric deformable surfaces to evolve via an intuitive physics-based paradigm: users can push, pull, into complex geometries, even modifying their topology as nec- essary. We demonstrate that our new ACID-based deformable anchor, reposition, and otherwise manipulate a deformable surfaces, dubbed T-surfaces, can effectively segment complex model using a mouse or other input devices. anatomic structures from medical volume imag