Quantitative evaluation of free‐form deformation registration for dynamic contrast‐enhanced MR mammography

Abstract

In this paper, we present an evaluation study of a set of registration strategies for the alignment of sequences of 3D dynamic contrast-enhanced magnetic resonance breast images. The accuracy of the optimal registration strategies was determined on unseen data. The evaluation is based on the simulation of physically plausible breast deformations using finite element methods and on contrast-enhanced image pairs without visually detectable motion artifacts. The configuration of the finite element model was chosen according to its ability to predict in vivo breast deformations for two volunteers. We computed transformations for ten patients with 12 simulated deformations each. These deformations were applied to the postcontrast image to model patient motion occurring between pre- and postcontrast image acquisition. The original precontrast images were registered to the corresponding deformed postcontrast images. The performance of several registration configurations (rigid, affine, B-spline based nonrigid, single-resolution, multi-resolution, and volume-preserving) was optimized for five of the ten patients. The images were most accurately aligned with volume-preserving single-resolution nonrigid registration employing 40 or 20 mm control point spacing. When tested on the remaining five patients the optimal configurations reduced the average mean registration error from 1.40 to 0.45 mm for the whole breast tissue and from 1.20 to 0.32 mm for the enhancing lesion. These results were obtained on average within 26 (81) min for 40 (20) mm control point spacing. The visual appearance of the difference images from 30 patients was significantly improved after 20 mm volume-preserving single-resolution nonrigid registration in comparison to no registration or rigid registration. No substantial volume changes within the region of the enhancing lesions were introduced by this nonrigid registration.

Measuring sparse temporal-variation for accurate registration of dynamic contrast-enhanced breast MR images

“…This improper assumption was proven to give rise to the volume shrinking/expanding effect of tumor when registering images using Eq. (4) [10]. Considering the fact that obvious intensity changes between pre-contrast and post-contrast images happen in the tumor area, the intensity differences between I and J can be assumed to be sparsely distributed in the image.…”

“…The registration accuracies rely significantly on the results of the extra processing. To achieve more accurate estimation on the nonrigid motion, different registration methods [7][8][9][10] have been proposed, which leverage a more efficient similarity measure, a better model of transformation, or a more advanced optimization strategy, etc. However, most optimization strategies are gradient descent based, with which only a local minimum can be guaranteed and the results are highly dependent on the initialization.…”

Adaptive Grid Generation Based Non-rigid Image Registration using Mutual Information for Breast MRI

“…In [22], a finite element model is applied on the images to generate visually detectable motion to produce pre-contrast and post-contrast images. It is tested under different configurations of registrations.…”

Computerized method for nonrigid MR-to-PET breast-image registration

“…A modification of this approach that assured preservation of the volume was reported by Rohfing et al [34]. Validation studies of the aforementioned methods were performed by Tanner et al [35]. Normalized mutual information was used as a similarity measure.…”