2012 | OriginalPaper | Buchkapitel
Femur Localization Using the Discriminative Generalized Hough Transform
verfasst von : Francesco Boero, Heike Ruppertshofen, Hauke Schramm
Erschienen in: Bildverarbeitung für die Medizin 2012
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
The Discriminative Generalized Hough Transform, a method for object localization and training of suitable models, is employed here for the localization of the femur in long-leg radiographs. The method is extended to improve the robustness of the procedure by integrating anatomical knowledge. This is achieved by combining the femur localization with the more robust knee localization. To this end, a region of interest is extracted from the original image, based on the result of the knee localization, in which the femur is initially localized. In addition, the procedure focuses on the target by extracting smaller regions of interest around the previously found coordinates and repeating the localization with smaller models. By this means, the precision of the first coarse localization is refined. The presented method is tested here on a large set of long-leg radiographs, where it achieves a localization rate of 94% on femurs not showing pathological conditions.