Yinyan Wang, Presenter: Nothing to Disclose

Tianyi Qian PhD, Abstract Co-Author: Nothing to Disclose

Lei Wang, Abstract Co-Author: Nothing to Disclose

Jun Ma MD, Abstract Co-Author: Nothing to Disclose

Tao Jiang, Abstract Co-Author: Nothing to Disclose

Aiming at evaluating the reliability of the rs-fMRI in depicting the motor functional plasticity following brain tumors, we include a cohort of glioma cases with a tumor involved in the motor area. To calculate/generate the functional correlation map, four seed regions were selected from each case according to either anatomical mark or task-based functional activation in each hemisphere.

A total of 14 consecutive patients (9 males) with gliomas treated were participated in this study. fMRI data were acquired on a SIEMENS Magnetom Trio 3T (Siemens, Germany), with using echo-planar imaging (EPI). Intra-operative ECS mapping was performed using a bipolar electrode with a 5-mm interelectrode distance. Peak areas of the task-based activations were selected as the seed regions in the lesion hemisphere and healthy hemisphere. The spatial distributions of motor network were obtained by computing the correlation coefficient between each seed regions and the whole brain.

Receiver operating characteristic (ROC) curves of various functional neuroimaging analyses were acquired by calculating the sensitivity and specificity under different thresholds. Motor functional brain regions located by ECS were considered as the reference. In group level, among five analyzing methods, task-based functional MR achieved the highest accuracy in predicting the hand motor area according to the ROC curve. Its sensitivity and specificity are 81.6% and 80.0% for the patient group at the fixed-threshold analysis. For seed-dependent functional connectivity mapping, motor networks calculated from task-determined seeds have higher conformity with ECS result, when compared to ones acquired from the anatomy-determined seeds. The motor functional network calculated from unilateral task-based seed showed a similar high reliability to task fMRI in predicting the motor plasticity determined by ECS. 

Resting-state functional MR imaging can be used to perform individualized locating of motor functional area and in consequence assist surgical planning in patients with brain tumors. Methods of seed selection influence the accuracy in predicting functional areas for clinical cases.

When tumor involved the motor area, the most reliable motor network was calculated from the seed which was determined by unilateral task-based activation.

Wang, Y, Qian, T, Wang, L, Ma, J, Jiang, T, Validation of Motor Network on Resting-state fMRI with Direct Cortical Stimulation for Glioma Cases.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14013590.html