EVolution: an edge-based variational method for non-rigid multi-modal image registration

B Denis De Senneville,C Moonen,M Ries,C Zachiu

doi:10.1088/0031-9155/61/20/7377

Abstract

Image registration is part of a large variety of medical applications including diagnosis, monitoring disease progression and/or treatment effectiveness and, more recently, therapy guidance. Such applications usually involve several imaging modalities such as ultrasound, computed tomography, positron emission tomography, x-ray or magnetic resonance imaging, either separately or combined.In the current work, we propose a non-rigid multi-modal registration method (namely EVolution: an edge-based variational method for non-rigid multi-modal image registration) that aims at maximizing edge alignment between the images being registered. The proposed algorithm requires only contrasts between physiological tissues, preferably present in both image modalities, and assumes deformable/elastic tissues. Given both is shown to be well suitable for non-rigid co-registration across different image types/contrasts (T1/T2) as well as different modalities (CT/MRI). This is achieved using a variational scheme that provides a fast algorithm with a low number of control parameters. Results obtained on an annotated CT data set were comparable to the ones provided by state-of-the-art multi-modal image registration algorithms, for all tested experimental conditions (image pre-filtering, image intensity variation, noise perturbation). Moreover, we demonstrate that, compared to existing approaches, our method possesses increased robustness to transient structures (i.e. that are only present in some of the images).

Highlights

Image registration is the process of aligning two or more images of the same scene acquired at different time instants, using different sensors and/or from a different point-of-view
Mono-modal registration algorithms typically rely on the assumption that an anatomical structure is present in all images included in the registration process and that the structures preserve, to a certain extent, their gray-level intensities
An image registration tool box including complementary multi-modal algorithms is a necessary prerequisite in the field of medical imaging, with respect to the ever growing need to align images of a same scene acquired at different time instants, using different sensors and/or from a different point-of-view

Summary

Introduction

Image registration is the process of aligning two or more images of the same scene acquired at different time instants, using different sensors and/or from a different point-of-view. Image registration is part of a large variety of medical applications including diagnosis, monitoring disease progression and/or treatment effectiveness and, more recently, therapy guidance (Mani and Arivazhagan 2013) Such applications usually involve several imaging modalities such as ultrasound (US), computed tomography (CT), positron emission tomography (PET), x-ray or magnetic resonance imaging (MRI), either separately or combined. Mono-modal registration algorithms typically rely on the assumption that an anatomical structure is present in all images included in the registration process and that the structures preserve, to a certain extent, their gray-level intensities. Such algorithms are well suited, for example, for tumor growth monitoring and intervention verification (Maintz and Viergever 1998). Compared to mono-modal algorithms, multi-modal methods are usually more complex and require additional computational resources

Methods

Results

Discussion

Conclusion