DeepBundle: Fiber Bundle Parcellation with Graph Convolution Neural Networks
This addresses the issue of inaccurate registration in brain white matter analysis for researchers in neuroimaging, though it appears incremental as it applies a known deep learning technique to a specific domain problem.
The paper tackled the problem of parcellating whole-brain tractography streamlines without relying on registration to an atlas, which is challenging due to individual differences, by proposing DeepBundle, a method using graph convolution neural networks that achieved effective fiber parcellation as demonstrated on Human Connectome Project data.
Parcellation of whole-brain tractography streamlines is an important step for tract-based analysis of brain white matter microstructure. Existing fiber parcellation approaches rely on accurate registration between an atlas and the tractograms of an individual, however, due to large individual differences, accurate registration is hard to guarantee in practice. To resolve this issue, we propose a novel deep learning method, called DeepBundle, for registration-free fiber parcellation. Our method utilizes graph convolution neural networks (GCNNs) to predict the parcellation label of each fiber tract. GCNNs are capable of extracting the geometric features of each fiber tract and harnessing the resulting features for accurate fiber parcellation and ultimately avoiding the use of atlases and any registration method. We evaluate DeepBundle using data from the Human Connectome Project. Experimental results demonstrate the advantages of DeepBundle and suggest that the geometric features extracted from each fiber tract can be used to effectively parcellate the fiber tracts.