Arne Hans1, Adam Wittek2, Grand
Joldes2, Karol Miller2, Neil I. Weisenfeld1,
Mark Alexiuk3, John Saunders3, Einat Liebenthal4,
Garnette R. Sutherland5, Simon K. Warfield1
1Radiology, Children's
Hospital Boston & Harvard Medical School, Boston, MA, United States; 2University
of Western Australia, Perth, Australia; 3IMRIS, Winnipeg, Canada; 4National
Research Council Canada, Canada; 5University of Calgary
We propose to significantly increase the intraoperative utility of data acquired preoperatively by compensating, in real time, for brain shift during surgery. This is demonstrated on craniotomy case data that has been processed using our non-rigid registration pipeline. We have achieved intraoperative visualization of preoperative fMRI and DTI aligned to intraoperative MRI, and we have validated the alignment quality by comparing automatically generated edges. Our study shows that preoperative data can be warped to the current configuration of the patient's brain in real time and with subvoxel accuracy. Such visualization may dramatically improve surgical decision making.