Classifyber, a robust streamline-based linear classifier for white matter bundle segmentation

Bertó, Giulia; Bullock, Daniel; Astolfi, Pietro; Hayashi, Soichi; Zigiotto, Luca; Annicchiarico, Luciano; Corsini, Francesco; Benedictis, Alessandro De; Sarubbo, Silvio; Pestilli, Franco; Avesani, Paolo; Olivetti, Emanuele

doi:10.1016/j.neuroimage.2020.117402

Cited by 32 publications

(29 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Third, seven major tracts were investigated in this study because of the restriction of example data in tract segmentation method. The other WM tracts could be considered in the future study with other tract segmentation method 54 – 56 . In addition, we focus on FA values since the meaning of abnormalities in other measures has not been fully understood in patients with RLS.…”

Section: Discussionmentioning

confidence: 99%

White matter tract-specific alterations in patients with primary restless legs syndrome

Park

Kim

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Prior diffusion tensor imaging (DTI) studies have investigated white matter (WM) changes in patients with primary restless legs syndrome (RLS), but the results were inconsistent. Here, we proposed using tract-specific statistical analysis (TSSA) to find alterations in specific WM tracts to clarify the pathophysiological mechanisms of RLS. We enrolled 30 patients with RLS and 31 age- and sex- matched controls who underwent brain magnetic resonance imaging, neuropsychological tests, and polysomnography. Fractional anisotropy (FA) maps obtained from whole-brain diffusion tensor imaging and TSSA were used to localize WM changes in patients with RLS. Subsequently, a comparison of FA values for each tract between patients and controls was performed. The associations between FA values and clinical, polysomnographic, and neuropsychological parameters in RLS patients were assessed. RLS patients demonstrated decreased FA values in the left corticospinal tract (CST) and cingulum, and in the right anterior thalamic radiation (ATR) and inferior fronto-occipital fasciculus (IFO). Patients’ attention/executive function and visual memory scores positively correlated with FA values in the right ATR, and anxiety levels negatively correlated with FA values in the right IFO. Additionally, the number of periodic leg movements and movement arousal index were negatively correlated with FA values in the left CST. The TSSA method identified previously unknown tract-specific alterations in patients with RLS and significant associations with distinct clinical manifestations of RLS.

show abstract

Section: Discussionmentioning

confidence: 99%

White matter tract-specific alterations in patients with primary restless legs syndrome

Park

Kim

et al. 2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…The bundle-wise distance matrix is constructed by computed the pairwise distances among all the bundle types RNN-based method naturally supports variable lengths of sequence input without the need to truncate or resample the original streamlines. Unlike manually designed features (Yendiki et al, 2011;Bertò et al, 2021), the latent space dimension in the proposed framework can be customized by adjusting the RNN hidden dimension. RNN networks with larger hidden dimension, i.e.…”

Section: Discussionmentioning

confidence: 99%

“…Olivetti et al (2012) designed three different policies to select reference streamlines, and used a symmetric minimum average distance (Zhang et al, 2008) to encode each streamline to a feature vector. Bertò et al (2021) selected reference landmarks from global and local perspectives, and considered the distances of the streamline endpoints to the global reference landmarks and the predefined ROIs. Lam and co-workers (Lam et al, 2018) also included curvature and torsion as features to represent a streamline.…”

Section: Introductionmentioning

confidence: 99%

Auto-encoded Latent Representations of White Matter Streamlines for Quantitative Distance Analysis

2022

View full text Add to dashboard Cite

Parcellation of whole brain tractograms is a critical step to study brain white matter structures and connectivity patterns. The existing methods based on supervised classification of streamlines into predefined streamline bundle types are not designed to explore sub-bundle structures, and methods with manually designed features are expensive to compute streamline-wise similarities. To resolve these issues, we propose a novel atlas-free method that learns a latent space using a deep recurrent auto-encoder trained in an unsupervised manner. The method efficiently embeds any length of streamlines to fixed-size feature vectors, named streamline embedding, for tractogram parcellation using non-parametric clustering in the latent space. The method was evaluated on the ISMRM 2015 tractography challenge dataset with discrimination of major bundles using clustering algorithms and streamline querying based on similarity, as well as real tractograms of 102 subjects Human Connectome Project. The learnt latent streamline and bundle representations open the possibility of quantitative studies of arbitrary granularity of sub-bundle structures using generic data mining techniques.

show abstract

“…Using these resources, it should be possible to re-execute our workflows and replicate most of our results (53). For example, if other researchers would be interested in comparing our TRR results to another tractometry pipeline (e.g., TRACULA (11), another popular tractometry pipeline) or another bundle recognition algorithm (e.g., TractSeg (54), which uses a neural network to recognize bundles, or Classifyber (55), which uses a linear classifier), they could do so with the HCP-TR dataset, inspired by our scripts and the visualization tools in the pyAFQ software.…”

Section: Exceptions and Limitationsmentioning

confidence: 99%

Evaluating the Reliability of Human Brain White Matter Tractometry

et al. 2021

View full text Add to dashboard Cite

The validity of research results depends on the reliability of analysis methods. In recent years, there have been concerns about the validity of research that uses diffusion-weighted MRI (dMRI) to understand human brain white matter connections in vivo, in part based on the reliability of analysis methods used in this field. We defined and assessed three dimensions of reliability in dMRI-based tractometry, an analysis technique that assesses the physical properties of white matter pathways: (1) reproducibility, (2) test-retest reliability, and (3) robustness. To facilitate reproducibility, we provide software that automates tractometry (https://yeatmanlab.github.io/pyAFQ). In measurements from the Human Connectome Project, as well as clinical-grade measurements, we find that tractometry has high test-retest reliability that is comparable to most standardized clinical assessment tools. We find that tractometry is also robust: showing high reliability with different choices of analysis algorithms. Taken together, our results suggest that tractometry is a reliable approach to analysis of white matter connections. The overall approach taken here both demonstrates the specific trustworthiness of tractometry analysis and outlines what researchers can do to establish the reliability of computational analysis pipelines in neuroimaging.

show abstract

Classifyber, a robust streamline-based linear classifier for white matter bundle segmentation

Cited by 32 publications

References 53 publications

White matter tract-specific alterations in patients with primary restless legs syndrome

White matter tract-specific alterations in patients with primary restless legs syndrome

Auto-encoded Latent Representations of White Matter Streamlines for Quantitative Distance Analysis

Evaluating the Reliability of Human Brain White Matter Tractometry

Contact Info

Product

Resources

About