A fast alignment method for breast MRI follow-up studies using automated breast segmentation and current-prior registration

Abstract

In breast cancer screening for high-risk women, follow-up magnetic resonance images (MRI) are acquired with a time interval ranging from several months up to a few years. Prior MRI studies may provide additional clinical value when examining the current one and thus have the potential to increase sensitivity and specificity of screening. To build a spatial correlation between suspicious findings in both current and prior studies, a reliable alignment method between follow-up studies is desirable. However, long time interval, different scanners and imaging protocols, and varying breast compression can result in a large deformation, which challenges the registration process. In this work, we present a fast and robust spatial alignment framework, which combines automated breast segmentation and current-prior registration techniques in a multi-level fashion. First, fully automatic breast segmentation is applied to extract the breast masks that are used to obtain an initial affine transform. Then, a non-rigid registration algorithm using normalized gradient fields as similarity measure together with curvature regularization is applied. A total of 29 subjects and 58 breast MR images were collected for performance assessment. To evaluate the global registration accuracy, the volume overlap and boundary surface distance metrics are calculated, resulting in an average Dice Similarity Coefficient (DSC) of 0.96 and root mean square distance (RMSD) of 1.64 mm. In addition, to measure local registration accuracy, for each subject a radiologist annotated 10 pairs of markers in the current and prior studies representing corresponding anatomical locations. The average distance error of marker pairs dropped from 67.37 mm to 10.86 mm after applying registration

RGB-Topography and X-rays Image Registration for Idiopathic Scoliosis Children Patient Follow-up

“…In [23] a rough affine transform using Sum of Squared Differences (SSD) similarity measure is applied to register landmarks of two 3D breast MRI. Edge-based non rigid deformation is then estimated using a Normalized Gradient Fields (NGF) similarity measure.…”

Deformable current-prior registration of DCE breast MR images on multi-site data

“…Since the evaluation of the prior MR image is crucial to rate the malignancy of breast lesions, numerous studies have already developed deformable image registration techniques to ease the comparison. [4][5][6][7][8][9] The registration also simplifies the assessment of the growth rate of a tumor in the 3D image.…”

“…A crucial first step is the registration of prior and current MRI studies, to extract structural changes. In this work, a deformable registration technique building on previous work 6 and a new breast segmentation 10 is evaluated.…”

“…The state-of-the-art evaluation for image registration is the comparison against expert landmarks in the source and target images. [4][5][6][7][8] Others propose to use the severity of motion artifacts in the subtraction images to show the improvement with registration. 9 Furthermore, the registration can be rated via Dice score of the overlap of breast segmentations 6 which quantifies the global breast alignment, but can not account for local tissue deformations.…”

“…Furthermore, we re-evaluate our approach to datasets from a previous study on current-prior breast MRI registration. 6…”