Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

THIS WEBPAGE IS DEPRECATED - please visit our new website

Invited Talk by Ali Khamene PhD (Siemens Corporate Research (SCR), Princeton, USA)

Ali Khamene, <noautolink>PhD Ali Khamene, PhD (Siemens Corporate Research (SCR), Princeton) will be giving an invited talk on Friday, Feb. 27, 10:30-11:30, in room MI 03.13.010


"How to make 2d-3d registration 'robust'? Mean shift optimization on the SE(3) manifold"

Abstract: In this talk I introduce a new optimization method for 2D-3D image registration problem. The main problem in conventional optimization techniques used for 2D-3D registration is the limited capture range due to the convergence to local optima. The central idea in the proposed method is to use a robust clustering algorithm, such as mean shift, to compartmentalize the search manifold and find a robust optimum. In this approach, the distance metric used within the clustering algorithm, is chosen to be conformant to the topology of the search manifold. I show examples of how this method outperforms more conventional ones for various clinical problems.

Biography: Ali Khamene received his BSc in Electrical and MSc in Biomedical Engineering from Sharif Univ of Tech. He received his PhD in Electrical and Computer Engineering from the University of Miami. He has been employed at SCR since 2000, and currently spearheads the oncology imaging group, focusing on innovations for both diagnostic and interventional applications.

Title: Invited Talk by Ali Khamene PhD (Siemens Corporate Research (SCR), Princeton)
Date: 27 February 2009
Location: MI 03.13.010
Abstract: In this invited talk, Ali Khamene will introduce his work on a new optimization method for the 2D-3D image registration problem.
Imageurl: Ali.jpg
Type: News

Edit | Attach | Refresh | Diffs | More | Revision r1.4 - 19 Aug 2010 - 09:49 - DarkoZikic

Lehrstuhl für Computer Aided Medical Procedures & Augmented Reality    rss.gif