Multimodal Surface Matching: Fast and Generalisable Cortical Registration Using Discrete Optimisation

Abstract

Group neuroimaging studies of the cerebral cortex benefit from accurate, surface-based, cross-subject alignment for investigating brain architecture, function and connectivity. There is an increasing amount of high quality data available. However, establishing how different modalities correlate across groups remains an open research question. One reason for this is that the current methods for registration, based on cortical folding, provide sub-optimal alignment of some functional subregions of the brain. A more flexible framework is needed that will allow robust alignment of multiple modalities. We adapt the Fast Primal-Dual (Fast-PD) approach for discrete Markov Random Field (MRF) optimisation to spherical registration by reframing the deformation labels as a discrete set of rotations and propose a novel regularisation term, derived from the geodesic distance between rotation matrices. This formulation allows significant flexibility in the choice of similarity metric. To this end we propose a new multivariate cost function based on the discretisation of a graph-based mutual information measure. Results are presented for alignment driven by scalar metrics of curvature and myelination, and multivariate features derived from functional task performance. These experiments demonstrate the potential of this approach for improving the integration of complementary brain data sets in the future.

MSM: A new flexible framework for Multimodal Surface Matching

“…The results here extend work presented in Robinson et al (2013), in which the utility of the Multimodal Surface Matching framework for surface registration was demonstrated through use of simulation and preliminary analyses using neurobiological data. Here, we significantly expand these analyses and present results using new multivariate and multimodal MRI data.…”

The WU-Minn Human Connectome Project: An overview

“…In addition to these scan sessions, participants complete extensive behavioral assessment outside the scanner, during two sessions lasting a total of several hours (see Tables 2 and 3 in Barch et al, 2013). One set of measures, from the NIH Toolbox (http://www.nihtoolbox.org/) is typically done on visit Day 1, takes about 2 hr and includes 19 subdomains within the broad domains of cognitive, emotional, motor, and sensory function (see Barch et al, 2013, Table 2).…”

“…One set of measures, from the NIH Toolbox (http://www.nihtoolbox.org/) is typically done on visit Day 1, takes about 2 hr and includes 19 subdomains within the broad domains of cognitive, emotional, motor, and sensory function (see Barch et al, 2013, Table 2). The other session (~1.5 hr duration) of 11 non-Toolbox measures is typically done on Day 2 and includes tests of vision (color vision, contrast sensitivity), attention, personality, episodic memory, emotion processing, spatial processing, fluid intelligence, and self-regulation (delay discounting).…”

“…Additional articles in this special issue go into greater detail in these specific areas and provide a wealth of information about our instrumentation and image acquisition methods (Ugurbil et al, 2013); preprocessing pipelines (Glasser et al, 2013b); diffusion imaging (Sotiropoulos et al, 2013c); resting-state fMRI (Smith et al, 2013); task-fMRI and behavior (Barch et al, 2013); MEG (Larson-Prior et al, 2013); and informatics and quality control processes (Marcus et al, 2013). Other special issue articles describe progress by the MGH-UCLA HCP consortium.…”

Micro-, Meso- and Macro-Connectomics of the Brain

“…This approach is inherently problematic for dealing with regions of high folding variability, especially since the location of the cortical areas and functionally specialized regions vary in relation to gyral and sulcal landmarks. Fortunately, new registration methods have recently emerged that capitalize on functionally relevant features (e.g., myelin maps, fMRI data) in conjunction with shape-based information (Sabuncu et al 2010;Robinson et al 2013). In the examples illustrated below, we capitalize on the Multimodal Surface Matching (MSM) method as applied to HCP and macaque datasets ).…”