Inference of structural brain networks and modeling of cortical multi-sensory integration

Anatomical brain connectome;Multisensory integration

Shadi, Kamal ; Computer Science Dovrolis, Constantine Zegura, Ellen Keilholz, Shella Dyer, Eva Kira, Zsolt

University:Georgia Institute of Technology

Department:Computer Science

关键词: Anatomical brain connectome;    Multisensory integration;   

Others  :  https://smartech.gatech.edu/bitstream/1853/62258/1/SHADI-DISSERTATION-2019.pdf

美国|英语

来源: SMARTech Repository

【 摘 要 】

Recent advances in neuroimaging have enabled major progress in the field of brain connectomics, i.e., constructing maps of connections between brain regions at different scales. Diffusion MRI (dMRI) and probabilistic tractography algorithms are state of the art methods to map the structural connectome of the brain non-invasively and in vivo. Although probabilistic tractography can detect many major connections in the brain, it also reports some spurious ones. We propose and evaluate a method, referred to as MANIA (Minimum Asymmetry Network Inference Algorithm) that can infer the structural brain network that interconnects a given set of Regions of Interest (ROIs) from probabilistic tractography data in a threshold-free manner. Given that diffusion MRI is unable to detect the direction of each connection, we formulate the network inference process as an optimization problem that minimizes the (appropriately normalized) asymmetry of the observed network. The most fundamental property of the human connectome, its density, is still elusive and debated. MANIA is well-positioned to address this open question because it does not depend on an arbitrary weight threshold. We use MANIA to infer the human cortico-cortical connectome from the data published by Human Connectome Project (HCP). MANIA reports connectomes that are highly consistent across individuals at a density of approximately 3.2\%. We validate the accuracy of these connectomes by comparing the connections inferred using MANIA at 3T MRI acquisitions with 7T high-resolution MRI acquisitions of the same subjects. Having a structural network is instrumental in analyzing communication dynamics and information processing in the brain. The last research problem, we focus on relates to multi-sensory integration in the cortex. We model this process on the mouse cortical connectome (provided by the Allen Institute) by employing an Asynchronous Linear Threshold (ALT) diffusion model on that connectome. The ALT model captures how evoked activity that originates at a primary sensory region of the cortex “ripples through” other cortical regions. We validate the ALT model using Voltage Sensitive Dye (VSD) imaging data. Our results show that a small number of cortical regions (including the Claustrum) integrate almost all sensory information streams, suggesting that the cortex uses an hourglass architecture to integrate and compress multi-sensory information.