Classification of Radiographs in the 'Image Retrieval in Medical Applications' - System

In this paper we present a new approach to classifying radiographs, which is the first important task of the IRMA system. Given an image, we compute posterior probabilities for each image class, as this information is needed for further IRMA processing. Classification is done by using an extended version of Simard's tangent distance within a kernel density based classifier. We propose a new distortion model for radiographs and prove its effectiveness by applying the method to 1617 radiographs coming from daily routine. Although the distortion model alone yields very good results, the best recognition rates are obtained by combining it with tangent distance, i.e. by computing a `distorted tangent distance'.

• Type: Conference
• Year: 2000
• Where: RIAO
• Authors: Jörg Dahmen, Thomas Theiner, Daniel Keysers, Thomas Martin Lehmann, Berthold B. Wein
• Keywords: Distortion Model, RIAO 2000, RIAO 2007, Simard's Tangent Distance, Tangent Distance