Do multiple sclerosis lesions affect the outcome of magnetic resonance image registration?

Paulo Guilherme de Lima Freire, Ricardo José Ferrari

Abstract


Multiple sclerosis (MS) is a chronic inflammatory demyelinating disease of the central nervous system manifested morphologically by inflammation, demyelination, axonal loss and gliosis. The Magnetic Resonance (MR) imaging, which presents high resolution and good differentiation between brain tissues, is considered the gold standard for the detection and evolution assessment of MS lesions. Fairly recently, a number of automatic image processing systems have been proposed in the literature to help radiologists detect and compute the volume of MS lesions in MR images. Among most of these approaches, registration of multi-contrast MR clinical images (T1/T2/PD/FLAIR) as well as the registration of patient’s data to anatomical atlases have proved to be essential steps in the neuroimaging processing pipeline to successfully detect and quantitatively assess lesion load in MS. However, despite its importance and common use in automatic image processing systems, the effect of MS lesions on the final results of image registration has not been thoroughly investigated. In this work, image registration techniques using both affine and deformable transformations were used to assess if MS lesions (stratified in mild, moderate or severe) had any affect on the final results of image registration. Based on quantitative results obtained using metrics such as Dice and volume overlap, it was verified that MS lesions do not significantly affect the registration process. In the severe lesions case, for example, the Dice metric gave results of 0.999 and 0.963 for affine and deformable transformation, respectively. Also, it was verified that the sole use of affine transformation is very reasonable to correctly align images even if they do have lesions.

Full Text:

PDF


DOI: https://doi.org/10.22456/2175-2745.47589

Copyright (c) 2018 Paulo Guilherme de Lima Freire, Ricardo José Ferrari

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.