Abstracts

Rationale: Resting-state fMRI (1) has been used to reveal connectivity of functional networks in the brain and recently has been shown to relate to cognitive task performance (2,3). This work introduces a region-of-interest (ROI) approach, for assessing functional connectivity within specific brain networks. The approach can be used to evaluate laterality of language or memory circuits, and can also reveal networks associated with inter-ictal activity. Methods: Resting state fMRI data were collected on a Siemens 3T MRI scanner using 10min. runs with 64x64 matrix, alpha=80, TE/TR=30/1550ms in 38 subjects with intractable epilepsy who were candidates for surgery. Simultaneous EEG data was collected to record inter-ictal spike activity for localization of generating tissue. Data were filtered to remove physiologic noise and reveal low frequency <1Hz signal fluctuations. ROIs defining the motor, language, and memory pathways were defined from other studies, and some regions were defined based on anatomy, and all data was co-registered to a reference brain space using in-house software (bioimagesuite.org). Connectivity is defined by the strength of temporal correlations between ROIs. Hemispheric asymmetry is obtained as a ratio of differences in left/right ROI connectivity within specific networks or globally.Results: The results demonstrate that this approach can be used to assess motor, language and memory asymmetries in epilepsy patients and that such an analysis can reveal specific aberrant nodes within a network. Comparison of patient to normative control data reveals nodes of increased and decreased connectivity that differ significantly from the control data. Figure 1 (top row) shows an example where inter-ictal spike fMRI reveals abnormal activity in right orbital frontal cortex. The matrix on the left represents the global interregional connectivity while the matrix on the right reveals connectivity asymmetry (L-R)/(L+R) in the orbital frontal cortex (4th row from the bottom - blue squares). Similar results are shown across a range of pathologies in these patients.Conclusions: This work reveals a new task-free fMRI based approach for assessing functional networks that should be spared in surgery, lateralization of language and memory cortex, and epileptogenic networks that may need to be considered in planning a surgical intervention. Clear evidence for such networks is provided and validated with data obtained from simultaneous spike triggered eeg-fMRI and clinical data. References 1. Biswal B, et al, Functional connectivity in the motor cortex of resting human brain using echo-planar MRI, Magn. Reson. In Med. 35(1): 107-113, 1996. 2. Hampson M, Driesen NR, Skudlarski P, Gore JC, Constable RT, Brain connectivity related to working memory performance, J. Neuroscience, 26(51): 13338-13343, 2006. 3. Hampson M, Tokoglu F, Sun Z, Schafer RJ, Skudlarski P, Gore JC, Constable RT, Connectivity-behavior analysis reveals that functional connectivity between left BA39 and Broca’s varies with reading ability, Neuroimage 31(2): 513-519, 2006. Acknowledgments: NIH NS047605