Comparing functional connectivity based predictive models across datasets

Abstract

To cite this version:Kamalaker Dadi, Alexandre Abraham, Mehdi Rahim, Bertrand Thirion, Gaël Varoquaux. Comparing functional connectivity based predictive models across datasets. Abstract-Resting-state functional Magnetic Resonance Imaging (rs-fMRI) holds the promise of easy-to-acquire and widespectrum biomarkers. However, there are few predictivemodeling studies on resting state, and processing pipelines all vary. Here, we systematically study resting state functionalconnectivity (FC)-based prediction across three different cohorts. Analysis pipelines consist of four steps: Delineation of brain regions of interest (ROIs), ROI-level rs-fMRI time series signal extraction, FC estimation and linear model classification analysis of FC features. For each step, we explore various methodological choices: ROI set selection, FC metrics, and linear classifiers to compare and evaluate the dominant strategies for the sake of prediction accuracy. We achieve good prediction results on the three different targets. With regard to pipeline selection, we obtain consistent results in two pipeline steps -FC metrics and linear classifiers-that are vital in the diagnosis of rs-fMRI based disease biomarkers. Regarding brain ROIs selection, we observe that the effects of different diseases are best characterized by different strategies: Schizophrenia discrimination is best performed in dataset-specific ROIs, which is not clearly the case for other pathologies. Overall, we outline some dominant strategies, in spite of the specificity of each brain disease in term of FC pattern disruption.

MIST: A multi-resolution parcellation of functional brain networks

“…We discuss the results of this analysis in light of similar analyses in the existing parcellation literature. We have also included a discussion of a recent benchmark analysis by Dadi and colleagues [3] that show good performance of the MIST_64 parcellation on machine learning applications. Overall, although the dataset used here is smaller than others used in the field, e.g.…”

“…An early and important example is the parcellation of the brain into distinct areas of homogeneous cytoarchitecture by Korbinian Brodmann at the beginning of last century 1 . This line of work has since been extended to include brain parcellations based on a range of brain modalities, such as sulcal landmarks 2 , functional connectivity 3 , task activation 4 , gene expression patterns 5 , and combinations of different imaging modalities 6 . Good parcellations generate homogeneous parcels that tell us something about the organization of the brain along the corresponding modality, provide a common frame of reference for the localization of new findings, and serve the additional purpose of meaningful dimensionality reduction 7 .…”

“…As a result, a wide range of functional brain parcellations have been proposed in the literature, ranging in resolution from large functional systems (e.g. 3) to individual functional nodes (e.g. 4).…”

Tangent functional connectomes uncover more unique phenotypic traits

“…Until recently, this important fact that FCs lie on or inside a non-linear manifold has been scarcely considered by the neuroscientific community. 5 , 8 , 16 , 17 , 18 , 19 , 20 , 21 , 22 As a result, most analyses and frameworks did not take a full advantage of functional connectivity data to uncover their fingerprinting and/or biomarker capacity. This may have also limited the capacity of FC to predict cognitive outcomes or serve as reliable and robust clinical biomarkers of brain disorders.…”

“…The use of Riemannian geometry may mitigate such limitations and enable comparing FCs with basic algebraic operations on the manifold when the underlying non-linear geometry of the correlation-based FCs is incorporated. In comparison to “regular” FCs, 16 , 23 tangent-FCs have been proven to provide more accurate predictions of disease 19 , 21 , 24 , 25 and aging. 26 Also, Riemannian geometry-based approaches applied to functional connectivity have been recently used for harmonization of multi-site data, 27 as well as for brain connectivity interface.…”