DSpace

King Saud University Repository >
King Saud University >
COLLEGES >
Science Colleges >
College of Computer and Information Sciences >
College of Computer and Information Sciences >

Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/15679

Title: Automatic PSO-Based Deformable Structures Markerless Tracking in Laparoscopic Cholecystectomy
Authors: Haroun Djaghloul
Mohamed Batouche
Jean-Pierre Jessel
Keywords: Markerless tracking - Minimally Invasive Surgery - Image-guided Surgery - Evolutionary algorithms and Swarm Intelligence
Issue Date: 2010
Abstract: An automatic and markerless tracking method of deformable structures (digestive organs) during laparoscopic cholecystectomy intervention that uses the (PSO) behavour and the preoperative a priori knowledge is presented. The associated shape to the global best particles of the population determines a coarse representation of the targeted organ (the gallbladder) in monocular laparoscopic colored images. The swarm behavour is directed by a new fitness function to be optimized to improve the detection and tracking performance. The function is defined by a linear combination of two terms, namely, the human a priori knowledge term (H) and the particle’s density term (D). Under the limits of standard (PSO) characteristics, experimental results on both synthetic and real data show the effectiveness and robustness of our method. Indeed, it outperforms existing methods without need of explicit initialization (such as active contours, deformable models and Gradient Vector Flow) on accuracy and convergence rate.
URI: http://hdl.handle.net/123456789/15679
Appears in Collections:College of Computer and Information Sciences

Files in This Item:

File Description SizeFormat
DrBatouche-conf-26.docx12.4 kBMicrosoft Word XMLView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

 

DSpace Software Copyright © 2002-2007 MIT and Hewlett-Packard - Feedback