Fiberneat: Unsupervised Streamline Filtering In Latent Space

Whole-brain tractograms generated from diffusion are composed of millions of streamlines and can have false positive and anatomically implausible streamlines. To obtain anatomically relevant streamlines and tracts, supervised and unsupervised methods can be used for tractogram clustering and tract extraction. Here we propose FiberNeat, an unsupervised streamline filtering method. It takes an input set of streamlines that could either be unlabeled clusters or labeled tracts. Individual clusters/tracts are projected into a latent space using nonlinear dimensionality reduction techniques, such as t-SNE and UMAP, to find spurious and outlier streamlines. Outlier streamline clusters are detected using DBSCAN and removed from the data in streamline space. Quantitative comparisons with expertly delineated tracts show the promise of the approach. This approach can be deployed as a filtering step after tracts are extracted.