Abstract #1597

Deformable Registration of In Vivo Human Spine Magnetic Resonance Imaging Using Scale Invariant Feature Transform (SIFT) Algorithm

Dong-Hoon Lee^1,2, Do-Wan Lee^3,4,5, Yong Hyun Chung², and Bong-Soo Han²

¹Brain and Mind Centre, University of Sydney, Sydney, Australia, ²Department of Radiological Science, Yonsei University, Wonju, Korea, Republic of, ³Ewha Brain Institute, Ewha Womans University, Seoul, Korea, Republic of, ⁴Department of Biomedical Engineering, The Catholic University of Korea College of Medicine, Seoul, Korea, Republic of, ⁵Research Institute of Biomedical Engineering, The Catholic University of Korea, Seoul, Korea, Republic of

Deformable registration process for in vivo human spine MR images can provide the crucial information to investigate the diagnostic performance and treatment effects of pathologies. Here, based on the scale invariant feature transform (SIFT) algorithm, we attempted to evaluate the deformable registration process for spine images, and compared with commercially mounted algorithm in MRI system. The results qualitatively and quantitatively showed fine results, and clearly showed the reproducibility of the use of SIFT algorithm compared with commercially mounted stitching algorithm on MRI system. Our approach can be helpful for the extension of other medical imaging modalities for image deformable registration.

This abstract and the presentation materials are available to members only; a login is required.

Join Here