Tractography-Based Parcellation of Cerebellar Dentate Nuclei via a Deep Nonnegative Matrix Factorization Clustering Method

TL;DR

This paper introduces a deep nonnegative matrix factorization clustering method for parcellating the human dentate nucleus based on structural connectivity from diffusion MRI, revealing its topography with improved consistency.

Contribution

The study presents a novel deep NMF clustering approach for cerebellar dentate nucleus parcellation using tractography data, outperforming existing methods.

Findings

DNMFC produces higher quality parcellations.

Parcellations are more consistent across subjects.

Method validated on Human Connectome Project data.

Abstract

As the largest human cerebellar nucleus, the dentate nucleus (DN) functions significantly in the communication between the cerebellum and the rest of the brain. Structural connectivity-based parcellation has the potential to reveal the topography of the DN and enable the study of its subregions. In this paper, we investigate a deep nonnegative matrix factorization clustering method (DNMFC) for parcellation of the human DN based on its structural connectivity using diffusion MRI tractography. We propose to describe the connectivity of the DN using a set of curated tractography fiber clusters within the cerebellum. Experiments are conducted on the diffusion MRI data of 50 healthy adults from the Human Connectome Project. In comparison with state-of-the-art clustering methods, DN parcellations resulting from DNMFC show better quality and consistency of parcels across subjects.