Collaboration:College of William and Mary

Home < Collaboration:College of William and Mary

Abstract

This project has 3 goals:

Deliver guaranteed quality Image-to-Mesh (I2M) conversion tools for non-rigid registration of brain MRI. Specifically, we are interested to extend, for medical images, the traditional Delaunay-based mesh generation methods and develop a prototype software module for real-time I2M conversion that will simultaneous address four fundamental I2M conversion problems: image fidelity, sliver elimination, and guaranteed gradation and size optimality of the mesh as well termination of the FE-mesh process.
Develop real-time non-rigid registration of MRI images to meet the time constrains imposed by neurosurgery. Specifically use cooperative hardware architectures (based on multi-core and GPUs) that can be easily deployed in (or next to) the Operating Room without hindering routine surgery procedures to implement real-time non-rigid registration software which is accurate and robust.
Improve accuracy of non-rigid registration of brain MRI by utilizing the resources of the TeraGrid infrastructure. Our first objective is to perform a feasibility study to exploit the use distributed grid computing resources in order to provide computational platform for image processing during image-guided neurosurgery. In addition we want to demonstrate that we can utilize the vast resources of nation-wide platforms like the TeraGrid to facilitate large experimental studies of image processing algorithms to improve our understanding of their behavior under different inputs.

This project is a funded in part from NSF and John Simon Guggenheim Foundation

College of William & Mary: Nikos Chrisochoides, PI, Andrey Chernikov, Yixun Liu, Panagiotis Foteinos
NA-MIC: Andriy Fedorov

Chrisochoides N., Fedorov A., Kot A., Archip N., Black P.M., Clatz O., Golby A.J., Kikinis R., Warfield S.K. Toward Real-Time Image Guided Neurosurgery using Distributed and Grid Computing. Proceedings of the ACM/IEEE conference on Supercomputing 2006 Oct; 37-50.
Fedorov A., Chrisochoides N., Kikinis R., Warfield S.K. An Evaluation of Three Approaches to Tetrahedral Mesh Generation for Deformable Registration of Brain MR Images. Proc IEEE Int Symp Biomed Imaging. 2006 Apr; 658-61.
Fedorov A., Billet E., Prastawa M., Gerig G., Radmanesh A., Warfield S.K., Kikinis R., Chrisochoides N. Evaluation of Brain MRI Alignment with the Robust Hausdorff Distance Measures. The 4th International Symposium on Visual Computing 2008 Dec; LNCS 5358:594-603.
Fedorov A., Chrisochoides N. Tetrahedral Mesh Generation for Non-Rigid Registration of Brain MRI: Analysis of the Requirements and Evaluation of Solutions. 17th International Meshing Roundtable, 2008.
Y.Liu, A.Fedorov, R.Kikinis, N.Chrisochoides Real-time Non-rigid Registration of Medical Images on a Cooperative Parallel Architecture. IEEE International Conference on Bioinformatics & Biomedicine, 2009.
CRTC publications by year

A Novel Algorithmic Approach for Real-Time Image-to-Mesh Conversion of Brain MRI, 09/01/2009-08/31/2012, grant No. NSF CCF-0916526
A Multi-Layered Finite Element Application and Runtime System for Scalable High-End Computer Architectures, 09/01/2008-08/31/2011, grant No. NSF CCF-0833081
Software Environment for Real-Time Non-Rigid Registration using Commodity and Grid Computing, 09/01/2007-08/31/2010, grant No. NSF CNS-719929.
Three-Dimensional Generalized Parallel Delaunay Mesh Generation for the Numerical Solution of Partial Differential Equations, 09/15/2007-02/28/2009, grant No. NSF CCF-0750901.