CAN PRE-OPERATIVE RESTING STATE FMRI PROVIDE INSIGHTS INTO DISRUPTED LANGUAGE NETWORKS? A PRELIMINARY ANALYSIS

Abstract AIMS Awake craniotomy with intra-operative direct cortical stimulation is now considered the gold standard for pa- tients with brain tumours affecting the eloquent areas of the brain. Functional magnetic resonance imaging (fMRI) provides a non-invasive method for localisation of language dominant areas using task-specific brain activation maps. In this study, the authors evaluate the role of resting state fMRI in providing insights into dis- rupted language networks in patients with intrinsic brain tumours. METHODS 14participants were recruited prospectively for this study. After completion of anatomical and resting state fMRI, the participants were asked to perform a set of tasks – naming, semantics association and sentence com- pletion. The results of task-based fMRI were used to generate the masks of activation for each task. These masks were then used to calculate the seed-based connectivity at rest for each task. 10 patients with intrinsic brain tumours that presented with deficits in the above stated language domains were recruited in this study. The resting state seed-based connectivity was calculated for each patient in the 3 domains. A two-sample t test was conducted to understand if there were any differences between the two groups. RESULTS A task specific resting state network was determined using the healthy volunteer data set. The task-based ac- tivation was found to be a subset of a much more extensive resting state network. In presence of existing neurological deficits in a particular language domain, a relatively increased co-fluctuations were noted in the resting state network. For example, in a patient 10 with deficits in picture naming and semantic processing, corresponding increased functional connectivity was noted. CONCLUSION Resting state fMRI may provide insights into disrupted language networks resulting in the observed deficits in patients with intrinsic brain tumours.