Unsupervised white matter fiber clustering and tract probability map generation: Applications of a Gaussian process framework for white matter fibers

• Published in: NeuroImage (Orlando, Fla.) Vol. 51; no. 1; pp. 228 - 241
• Main Authors: Wassermann, D., Bloy, L., Kanterakis, E., Verma, R., Deriche, R.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 15.05.2010; Elsevier Limited; Elsevier
• Subjects: Adult; Algorithms; Anatomy, Artistic; Atlases as Topic; Automation; Bioengineering; Brain; Brain - anatomy & histology; Cluster Analysis; Clustering; Computer Simulation; Databases, Factual; Diffusion Magnetic Resonance Imaging - methods; Diffusion MRI; Female; Gaussian process; Humans; Image Processing, Computer-Assisted - methods; Imaging; Life Sciences; Male; Methods; Nerve Fibers, Myelinated; Neural Pathways - anatomy & histology; Normal Distribution; Organ Size; Probability; White matter fiber tracts; Diffusion MRI; Gaussian process; White matter fiber tracts; Clustering; white matter fibres; tractography; clustering; gaussian processes; white matter fibers
• DOI: 10.1016/j.neuroimage.2010.01.004
• ISSN: 1053-8119; 1095-9572

With the increasing importance of fiber tracking in diffusion tensor images for clinical needs, there has been a growing demand for an objective mathematical framework to perform quantitative analysis of white matter fiber bundles incorporating their underlying physical significance. This article presents such a novel mathematical framework that facilitates mathematical operations between tracts using an inner product between fibres. Such inner product operation, based on Gaussian processes, spans a metric space. This metric facilitates combination of fiber tracts, rendering operations like tract membership to a bundle or bundle similarity simple. Based on this framework, we have designed an automated unsupervised atlas-based clustering method that does not require manual initialization nor an a priori knowledge of the number of clusters. Quantitative analysis can now be performed on the clustered tract volumes across subjects, thereby avoiding the need for point parameterization of these fibers, or the use of medial or envelope representations as in previous work. Experiments on synthetic data demonstrate the mathematical operations. Subsequently, the applicability of the unsupervised clustering framework has been demonstrated on a 21-subject dataset.