Recognition of radiopaque markers in X-ray images using a neural network as nonlinear filter

Neural networks are developed to recognise radiopaque markers in biplane ci neangiographic video-images with a background composed of different objects . Our connectionist approach is compared theoretically as well as experimen tally with linear template matching. Theoretically, neural networks are lik ely to give better recognition results as they can implement nonlinear disc riminants. Experiments confirm that the networks result in better marker re cognitions than template matching. (C) 1999 Elsevier Science B.V. All right s reserved.