Towards measuring neuroimage misalignment

Abstract

To enhance neuro-navigation, high quality pre-operative images must be registered onto intra-operative configuration of the brain. Therefore evaluation of the degree to which structures may remain misaligned after registration is critically important. We consider two Hausdorff Distance (HD)-based evaluation approaches: the edge-based HD (EBHD) metric and the Robust HD (RHD) metric as well as various commonly used intensity-based similarity metrics such as Mutual Information (MI), Normalized Mutual Information (NMI), Entropy Correlation Coefficient (ECC), Kullback-Leibler distance (KLD) and Correlation Ratio (CR). We conducted the evaluation by applying known deformations to simple sample images and real cases of brain shift. We conclude that the intensity-based similarity metrics such as MI, NMI, ECC, KLD and CR do not correlate well with actual alignment errors, and hence are not useful for assessing misalignment. On the contrary, the EBHD and the RHD metrics correlated well with actual alignment errors; however, they have been found to underestimate the actual misalignment. We also note that it is beneficial to present HD results as a percentile-HD curve rather than a single number such as the 95-percentile HD. Percentile-HD curves present full range of alignment errors and also facilitate the comparison of results obtained using different approaches. Furthermore, the qualities that should be possessed by an ideal evaluation metric were highlighted. Future studies could focus on developing such an evaluation metric.

Biomechanical modeling and computer simulation of the brain during neurosurgery

“…Next, to obtain a qualitative assessment of the degree of alignment after registration, we examine the overlap of corresponding anatomical features of the intraoperative and registered preoperative image. To ensure repeatability and objectivity of the analysis, we chose Canny edges [111] as features for comparison [112]. Edges are easily recognizable features and a large proportion of them correspond to boarders of real anatomical structures visible in the image.…”

“…For a quantitative evaluation of the accuracy of the displacement calculations we used the Edge-based Hausdorff distance. This methodology, based on pioneering work of Huttenlocher et al [113], is described in detail in Garlapati et al [112].…”

Brain-shift compensation using intraoperative ultrasound and constraint-based biomechanical simulation

“…For the validation of our compensation method, the warped preoperative MRI must be compared with actual intraoperative information. While objective methods exist in the literature (Garlapati et al, 2015), they require data of the same modality (intraoperative MRI in that case). In our context, US images are the only intraoperative information available.…”

Registration quality filtering improves robustness of voxel-wise analyses to the choice of brain template

“…For component 2), we investigate the sensitivity of locations of voxel clusters where FA is significantly associated with age (FA clusters) to the choice of template. Poorly-registered images are filtered out using a new morphism quality measure that utilizes an average of Haussdorf-like distances ( Garlapati et al, 2015 ) to compare FreeSurfer-generated atlases of candidate templates and subjects morphed onto them ( Fleysher et al, 2017 ). We hypothesize that this average inter-atlas distance can characterize the extent of subject-to-template registration errors and determine the suitability of a candidate template for voxel-wise analysis of a set of specific subjects.…”