Title: Non-Rigid Registration for brain MRI: faster and cheaper

Authors: Yixun Liu, Andriy Fedorov, Ron Kikinis, Nikos Chrisochoides

Addresses: Department of Computer Science, College of William and Mary, Williamsburg, VA 23185, USA. ' Surgical Planning Laboratory, Brigham and Women's Hospital, Boston, MA 02115, USA. ' Surgical Planning Laboratory, Brigham and Women's Hospital, Boston, MA 02115, USA. ' Department of Computer Science, College of William and Mary, Williamsburg, VA 23185, USA

Abstract: We study the problem of Non-Rigid Registration (NRR) for intra-operative recovery of brain shift during image-guided neurosurgery. Time-critical nature of the tumour resection procedure presents a major obstacle to the routine clinical use of many available NRR approaches. In this paper, we utilise the resources of a single multicore workstation with an advanced graphics card to parallelise and evaluate an end-to-end implementation of a clinically validated NRR method. The results on clinical brain MRI data show the parallel NRR can reach real-time clinical requirement.

Keywords: non-rigid registration; GPU; multicore workstation; real-time clinical requirements; brain shift; image-guided neurosurgery; brain MRI data; magnetic resonance imaging; tumour resection.

DOI: 10.1504/IJFIPM.2010.033245

International Journal of Functional Informatics and Personalised Medicine, 2010 Vol.3 No.1, pp.48 - 57

Received: 06 Mar 2010
Accepted: 23 Mar 2010
Published online: 14 May 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article