Meeting Banner
Abstract #2268

Identifying Group Differences in Functional Subnetworks: A Novel Whole-Brain Method Applied to Dyslexia

Emily S. Finn1, Xilin Shen2, John M. Holahan3, Xenophon Papademetris2, 4, Dustin Scheinost4, Cheryl Lacadie2, Sally E. Shaywitz3, Bennett A. Shaywitz3, Robert Todd Constable2, 4

1Interdepartmental Neuroscience Program, Yale University, New Haven, CT, United States; 2Diagnostic Radiology, Yale University, New Haven, CT, United States; 3Yale Center for Dyslexia and Creativity, Yale University, New Haven, CT, United States; 4Biomedical Engineering, Yale University, New Haven, CT, United States

We applied a novel data-driven functional connectivity (FC) analysis to fMRI data from dyslexic vs. non-impaired readers to reveal functional subnetworks involved in successful reading. Our method improves upon previous FC analyses by not requiring a priori seed regions or arbitrary thresholds to determine connectivity. We found a network of occipitoparietal (visual association) and frontal (attention) areas that were better connected in non-impaired readers, suggesting that these subjects are better able to process word shapes and modulate their attention to visual stimuli. We believe this method can be extended to examine differentially connected functional subnetworks in other neural disorders.

Abstract PDF

Keywords

able activation analyses analyzed applied association atlases attention automatic axial balanced believe better beyond biomedical blue bold brain cell characterizing chosen component connected connection connections connectivity consider constable containing continue control controls corrected correlations datasets deepen define defined depiction detected determined diagnostic differentially diffuse disorders disrupted distinct done driven dyslexia dyslexic edge edges engineering examine examining executive extended frequently frontal functional haven head highlighting identifying impaired importance improve indicates inferior instead interdepartmental involved isolated john known laborious largest limit magnitude many matrix mixing modulate motion moving network networks neural node nodes normalized novel older pair permutation persistent phonology powerful previous priori process profiles program radiology readers reading recognizing regressed rely representing responsible rhyming sally score seed shapes sight slice sounding statistic stimuli strategies stronger strongly studies subject subjects subnetwork subnetworks subunits successful suggesting synchrony system task temporal threshold tool transformed trends understanding upon various versus visual whole word words young younger