Interactive Volume Reconstruction and Measurement on the Grid

Summary

Objectives: To prove the advantages of integrating grid computing within medical image analysis software, and to discuss the technological, sociological and health care-related issues.

Methods: Presentation of an instant volume reconstruction and measurement tool (PTM3D) used in clinical practice, including percutaneous nephrolithotomy examples; description of a parallel implementation of volume reconstruction, evaluation of this implementation on lung and body reconstruction, presentation of the technical limitations for clinical use and description and discussion of a prototype grid implementation.

Results: Volume reconstruction can broaden its medical scope and use by accessing high-performance computing systems; interactive exploration of medical images can co-exist with the usual batch workload of grid systems; the EGEE grid middleware offers some of the required core services; a fully adequate computing environment needs further evolution to integrate real-time constraints.

Conclusions: Clinical experiments of a grid-enabled PTM3D become possible. Widespread adoption of grid technology in the medical images analysis field will benefit from this “early user” project. Convergences appear between two broadly different fields, high energy physics and medical image, towards the need of a smooth integration of the new resources offered by grid systems into the everyday tools of their respective end-users. It can be expected that the convergence will mature towards truly interactive grids, able to serve the needs of the medical community.

Keywords

Radiology - computing grids - visualization

• References

• 1 Osorio A, Valette P-J, Mihalcea A, Atif J, Ripoche X. A new PC based software to perform semiautomatic hepatic segmentation using CT and MR images. InfoRAD-RSNA. 2002
• 2 Traxer O, Merran S, Osorio A, Atif J, Ripoche X, Thibault P. 3D reconstruction and instant volume measurement of complex renal calculi: a treatment preparation tool for percutaneous nephrolithotomy. InfoRAD-RSNA. 2003
• 3 Osorio A, Traxer O, Merran S, Atif J, Ripoche X, Tligui M. An augmented reality system for percutaneous nephrolithotomy. InfoRAD-RSNA. 2003
• 4 Beaumont O, Legrand A, Robert Y. Optimal algorithms for scheduling divisible workloads on heterogeneous systems. Procs 12th IEEE Heterogeneous Computing Workshop. 2003
• 5 Germain C, Osorio A, Texier R. A case study in medical imaging and the grid. 1st Health grid Conference. EC-IST 2003; 110-8.
• 6 Breton V, Medina R, Montagnat J. Datagrid, Prototype of a BioMedical grid. Methods Inf Med 2003; 42 (02) 143-8.
• 7 Raman R, Livny M, Solomon M. Resource Management through Multilateral Matchmaking. In: Procs 9th IEEE Symp on High Performance Distributed Computing (HPDC9). 2000: 290-1.
• 8 Stefanescu R, Pennec X, Ayache N. Grid Enabled Non-Rigid Registration with a Dense Transformation and a priori Information. In: Proc of MICCAI’03, LNCS. Springer Verlag: 2003
• 9 Berry D, Hill D, Knopp D, Pieper S, Saltz J, Germain C. Report of the IMAGE03: Images, Medical Analysis and Grid Environments workshop. UK e-Science Technical Report Series. http://www.nesc.ac.uk/technical_papers/.
• 10 Estrella F, McClatchey R, Rogulina D, Amendolia R, Solomonides T. A Service-Based Approach for Managing Mammography Data. 11th World Congress on Medical Informatics (MedInfo’04). San Francisco, USA: 2004