A Fast and Stable Method for Detecting and Tracking Medical Organs in MRI Sequences
Daewoong Kang · 1999
Introduction The analysis of medical organization through Magnetic Resonance Imaging is one of the recent noninvasive techniques developed to understand and model the human head or the heart structure. Clinical MRI studies include analysis for a large number of images from which qualitative information can be obtained by viewing the images in some sequences. Many applications in medical image processing rely on the reliable definition of object contours[1]. If done manually, this can be tedious and time-consuming. A fast and robust automated system is required for the contour extraction. This letter presents a central element of an image processing system, capable of detecting and tracking deformable contours in large MRI image sequences. Solutions should be presented to avoid undesirable deformation effects, like shrinking and vertex clustering, which are common in existing active contour models. Eviatar et al. [2] proposed an equilibrium term to prevent the snake from collapsing to a single point. The internal force enforces