Parametric reconstruction of boundaries and surfaces for medical imaging

The design of a reliable, high-quality method for finding the boundary contour of a two-dimensional object, or for finding the surface of a three dimensional object in the presence of random noise poses difficulties but is essential to medical imaging applications as when the image is supporting evidence for a decision or when its analysis guides a medical procedure. Stochastic function recovery is a recently developed mathematical technology and we show that its application to this task allows for the recovery of these boundaries, even when the noise is quite excessive. The technique should be considered as one of several that is able to improve the visualization of medical images. © 2007 IEEE