Nonrigid Registration of Brain Tumor Resection MR Images Based on Joint Saliency Map and Keypoint Clustering

Abstract

This paper proposes a novel global-to-local nonrigid brain MR image registration to compensate for the brain shift and the unmatchable outliers caused by the tumor resection. The mutual information between the corresponding salient structures, which are enhanced by the joint saliency map (JSM), is maximized to achieve a global rigid registration of the two images. Being detected and clustered at the paired contiguous matching areas in the globally registered images, the paired pools of DoG keypoints in combination with the JSM provide a useful cluster-to-cluster correspondence to guide the local control-point correspondence detection and the outlier keypoint rejection. Lastly, a quasi-inverse consistent deformation is smoothly approximated to locally register brain images through the mapping the clustered control points by compact support radial basis functions. The 2D implementation of the method can model the brain shift in brain tumor resection MR images, though the theory holds for the 3D case.

Intraoperative Imaging Modalities and Compensation for Brain Shift in Tumor Resection Surgery

“…The first is to use registration techniques based on intraoperative image data. Modalities such as iMR and iUS but also Laser Range Scanner (LRS) and Stereo Vision are used to acquire intraoperative image data [ 49 – 52 , 56 , 59 – 70 , 77 , 79 , 80 , 87 ]. Usually, LRS and Stereo Vision are used to acquire the intraoperative cortical surface deformation which is then integrated into a precomputed patient specific biomedical model in order to estimate the volumetric deformation.…”

“…The tissue deformation is then performed as Free Form Deformation using B -spline interpolation between the control points. Another approach to address the registration of pre- and postresection MR images has been proposed in [ 56 ]. In this work, the pre- and postresection MR images are first aligned rigidly to each other by maximizing the mutual information of salient structures enhanced by the joint saliency map.…”

A learning based fiducial-driven registration scheme for evaluating laser ablation changes in neurological disorders

“…Rechberg et al [24] presented an automatic approach for identifying landmark candidates by computing and choosing high distinctiveness values for the voxels within a region of interest. Gu and Qin [25] designed a global-to-local nonrigid brain image registration scheme in which the keypoints (i.e. landmarks) are selected based on a computed joint salient map.…”

Real-time visual saliency by Division of Gaussians

“…Subsequently, it formed the basis of several computer vision applications, such as in automatic object detection [2,3,4,5], medical imaging [6] and robotics [7]. Different saliency definitions exist, however, in this paper a generalised version of the definition by Achanta et al [8] is used: "Visual saliency is the perceptual quality that makes a group of pixels stand out relative to its neighbours".…”