A Framework for Inter-Subject Prediction of Functional Connectivity From Structural Networks

• Published in: IEEE transactions on medical imaging Vol. 32; no. 12; pp. 2200 - 2214
• Main Authors: Deligianni, Fani, Varoquaux, Gael, Thirion, Bertrand, Sharp, David J., Ledig, Christian, Leech, Robert, Rueckert, Daniel
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.12.2013; Institute of Electrical and Electronics Engineers
• Series: epub ahead of print
• Subjects: Bioengineering; Brain models; Cognitive science; Computer Science; Correlation; Covariance matrices; Functional brain connectivity; Life Sciences; Matrix decomposition; Medical Imaging; Neuroscience; predictive modeling; Predictive models; statistical associations; structural brain connectivity; Symmetric matrices; functional brain connectivity; predictive modeling; statistical associations; structural brain connectivity
• ISSN: 0278-0062; 1558-254X; 1558-254X
• DOI: 10.1109/TMI.2013.2276916

Functional connections between brain regions are supported by structural connectivity. Both functional and structural connectivity are estimated from in vivo magnetic resonance imaging and offer complementary information on brain organization and function. However, imaging only provides noisy measures, and we lack a good neuroscientific understanding of the links between structure and function. Therefore, inter-subject joint modeling of structural and functional connectivity, the key to multimodal biomarkers, is an open challenge. We present a probabilistic framework to learn across subjects a mapping from structural to functional brain connectivity. Expanding on our previous work [1], our approach is based on a predictive framework with multiple sparse linear regression. We rely on the randomized LASSO to identify relevant anatomo-functional links with some confidence interval. In addition, we describe resting-state functional magnetic resonance imaging in the setting of Gaussian graphical models, on the one hand imposing conditional independences from structural connectivity and on the other hand parameterizing the problem in terms of multivariate autoregressive models. We introduce an intrinsic measure of prediction error for functional connectivity that is independent of the parameterization chosen and provides the means for robust model selection. We demonstrate our methodology with regions within the default mode and the salience network as well as, atlas-based cortical parcellation.