Nan-kuei Chen1, Ying-hui Chou2, Lawrence P. Panych3,4, David J. Madden1, Allen W. Song1
1Brain Imaging and Analysis Center, Duke University, Durham, NC, USA; 2Occupational Therapy, Fu Jen Catholic University, Taipei, Taiwan; 3Radiology, Brigham and Women's Hospital, Boston, MA, USA; 4Harvard Medical School, Boston, MA, USA
Here we report an automatic post-processing pipeline to reliably identify functionally correlated regions in resting-state fMRI data. In comparison to existing analysis methods that are optimized for detecting functionally connecting brain regions (e.g. ICA), our new approach provides a comprehensive view of the functional connectivity across the whole brain and is better suited for identifying brain regions with low connectivity, without a prior assumption. We have successfully identified several brain regions that are functioning more independently from other cortical regions. We expect the developed method can reliably identify brain regions that are functionally deficient due to neurological diseases.