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. © 2010 Springer-Verlag.
CITATION STYLE
Djaghloul, H., Batouche, M., & Jessel, J. P. (2010). Automatic PSO-based deformable structures markerless tracking in laparoscopic cholecystectomy. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6076 LNAI, pp. 48–55). https://doi.org/10.1007/978-3-642-13769-3_6
Mendeley helps you to discover research relevant for your work.