White matter fiber analysis using kernel dictionary learning and sparsity priors. (November 2019)

• Record Type: Journal Article
• Authors: Kumar, Kuldeep; Siddiqi, Kaleem; Desrosiers, Christian
• Is Part Of: Pattern recognition. Volume 95(2019:Nov.)
• Page Start: 83
• Page End: 95
• Publication Date: 2019-11
• Subjects: Diffusion MRI -- White matter fibers -- Clustering -- Sparsity priors -- Kernel dictionary learning -- Human Connectome project; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2019.06.002
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: Novel fiber clustering approaches based on kernel dictionary learning and sparsity priors. Priors include: L-0 norm, global sparsity, group sparsity, & manifold regularization. Clustering analysis and validation on expert labeled set as well as subjects from Human Connectome Project. Results show proposed approaches group streamlines into plausible bundles, and illustrate the benefits of employing sparsity priors. Abstract: Diffusion magnetic resonance imaging, a non-invasive tool to infer white matter fiber connections, produces a large number of streamlines containing a wealth of information on structural connectivity. The size of these tractography outputs makes further analyses complex, creating a need for methods to group streamlines into meaningful bundles. In this work, we address this problem by proposing a set of flexible and efficient streamline clustering approaches based on kernel dictionary learning and sparsity priors. Proposed approaches, which include L 0 norm, group sparsity, and manifold regularization prior, allow streamlines to be assigned to more than one bundle, making the clustering more robust to overlapping bundles and inter-subject variations. We evaluate the performance of our method on an expert labeled dataset as well as data from the Human Connectome Project. Results highlight the ability of our method to group streamlines into plausible bundles and illustrate the impact of sparsity priors on the performance of the proposed methods. Methods presented in this work are relevant for the neuroscience studies on diffusion tractography analysis, as well as pattern recognition applications requiring the unsupervised clustering of 3D curves.