Brain topography beyond parcellations: Local gradients of functional maps

• Published in: NeuroImage (Orlando, Fla.) Vol. 229; p. 117706
• Main Authors: Dohmatob, Elvis, Richard, Hugo, Pinho, Ana Luísa, Thirion, Bertrand
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.04.2021; Elsevier Limited; Elsevier
• Subjects: Brain; Brain architecture; Brain mapping; Computer Science; Datasets; Functional gradients; Functional magnetic resonance imaging; Functional mapping; Medical Imaging; Model selection; Neuroimaging; Parcellation; Prediction; Topography; Functional mapping; Functional gradients; Parcellation; Model selection; Prediction
• ISSN: 1053-8119; 1095-9572; 1095-9572
• DOI: 10.1016/j.neuroimage.2020.117706

•Assessing the existence of local gradients in brain organization calls for multi-contrast analyses, such as task- vs. rest-fMRI.•Using about 200 parcels yields highest accuracy for local linear rest-to-task map prediction.•Combining results from multiple parcellations improves model accuracy.•Local linear models outperform whole-brain non-linear models.•Motor contrasts are less well predicted from resting-state activity than high-level contrasts. Functional neuroimaging provides the unique opportunity to characterize brain regions based on their response to tasks or ongoing activity. As such, it holds the premise to capture brain spatial organization. Yet, the conceptual framework to describe this organization has remained elusive: on the one hand, parcellations build implicitly on a piecewise constant organization, i.e. flat regions separated by sharp boundaries; on the other hand, the recently popularized concept of functional gradient hints instead at a smooth structure. Noting that both views converge to a topographic scheme that pieces together local variations of functional features, we perform a quantitative assessment of local gradient-based models. Using as a driving case the prediction of functional Magnetic Resonance Imaging (fMRI) data —concretely, the prediction of task-fMRI from rest-fMRI maps across subjects— we develop a parcel-wise linear regression model based on a dictionary of reference topographies. Our method uses multiple random parcellations —as opposed to a single fixed parcellation— and aggregates estimates across these parcellations to predict functional features in left-out subjects. Our experiments demonstrate the existence of an optimal cardinality of the parcellation to capture local gradients of functional maps.