Loading [a11y]/accessibility-menu.js
Effective and efficient interpolation for mutual information based multimodality elastic image registration | IEEE Conference Publication | IEEE Xplore

Effective and efficient interpolation for mutual information based multimodality elastic image registration


Abstract:

Mutual information (MI) is a popular similarity metric for multimodality image registration purpose. However, it is negatively influenced by artifacts due to interpolatio...Show More

Abstract:

Mutual information (MI) is a popular similarity metric for multimodality image registration purpose. However, it is negatively influenced by artifacts due to interpolation effects. As a result, registration algorithms performance could be affected. In this paper a novel interpolation scheme is presented. It is both effective and efficient. Effective because it limits the presence of local maxima in the mutual information curve, efficient because it is simple to compute being based on simple and optimized distance measures. The method is validated and compared against other techniques both from performance and time complexity perspectives. Differently from other reference works, which perform tests using rigid transformations, method performance was measured using non-rigid parametric transformation. Experimental results show that the proposed scheme represents a good tradeoff between performance and computing requirements, making it suitable for actual elastic image registration purposes.
Date of Conference: 27 September 2009 - 04 October 2009
Date Added to IEEE Xplore: 03 May 2010
ISBN Information:
Conference Location: Kyoto, Japan

References

References is not available for this document.