A pre-processing scheme for real-time registration of dynamic contrast-enhanced magnetic resonance images

Abstract

The paper proposes a fully automatic pre-processing scheme for real-time registration of the breast dynamic contrast-enhanced magnetic resonance images (DCE-MRIs). It consists of an intensity inhomogeneity correction followed by a breast region of interest segmentation. To correct the intensity inhomogeneity in the breast images, a combination of a greyscale morphological closing with the Gaussian filtering is used. The breast region of interest segmentation is based on fitting the human chest by an ellipse. The main advantages of the proposed pre-processing scheme are a low computational complexity and robustness to the breast density. The scheme was evaluated on 50 T1-weighted DCE-MRIs demonstrating a significant decrease in the time taken for registration (almost a factor 2 for the affine registration and almost a factor 9 for a non-linear B-spline free-form deformation).