Home // INFOCOMP 2013, The Third International Conference on Advanced Communications and Computation // View article

Adaptive Active Contours for the 3D Segmentation of Computed Tomography Images

Authors:
Miguel Alemán-Flores
Luis Alvarez

Keywords: segmentation; computed tomography; active contours

Abstract:
Computed Tomography is one of the most valuable modalities of medical imaging and is used in the diagnosis of a wide range of pathologies. In many cases, it is necessary to obtain a precise segmentation and a 3D visualization of certain organs, tissues, nodules or regions in the tomography. However, manual delimitation is a very time-consuming process and advanced applications are needed to perform this process automatically. Active contours intend to fit an initial approximation to the most relevant edges around the starting contour. Most approaches of active contours assume that all points in the initial approximation are close to the final solution and that similar conditions can be set to all the sections of the contour, which is frequently a false assumption. For that reason, we have developed a new approach, in which different terms are used and adapted according to the evolution of the contour. Balloon, regularizing and attraction terms are combined and extended to 3D, avoiding the need for a close initial approximation and reducing human intervention. Furthermore, a multiscale framework is added to tackle the heterogeneity of the images and the variability of the approximations. The combination of these terms allows obtaining a more precise segmentation of the tomography.

Pages: 66 to 71

Copyright: Copyright (c) IARIA, 2013

Publication date: November 17, 2013

Published in: conference

ISSN: 2308-3484

ISBN: 978-1-61208-310-0

Location: Lisbon, Portugal

Dates: from November 17, 2013 to November 21, 2013