Please use this identifier to cite or link to this item:
http://hdl.handle.net/10553/52824
Title: | Breast nodule ultrasound segmentation through texture-based active contours | Authors: | Aleman-Flores, Miguel Alvarez, Luis Caselles, Vicent |
UNESCO Clasification: | 220990 Tratamiento digital. Imágenes 32 Ciencias médicas 120601 Construcción de algoritmos 120602 Ecuaciones diferenciales 120326 Simulación |
Issue Date: | 2008 | Publisher: | 1612-3956 | Journal: | Progress In Industrial Mathematics At Ecmi 2006 | Conference: | 14th European Conference for Mathematics in Industry | Abstract: | This work presents a method for the segmentation of breast nodules in ultrasonography. Speckle noise is reduced using an anisotropic filter for which texture is described using Gabor filters. Afterwards, an initial segmentation is extracted using a front propagation scheme. Finally, the initial segmentation is refined using active contours. In order to delimit the nodules not only by means of image intensity, but also by texture pattern, we introduce certain terms in the classical active contours equations. | URI: | http://hdl.handle.net/10553/52824 | ISBN: | 978-3-540-71991-5 | ISSN: | 1612-3956 | Source: | Progress In Industrial Mathematics At Ecmi 2006[ISSN 1612-3956],v. 12, p. 858-+ |
Appears in Collections: | Actas de congresos |
Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.