NIMG-49. THE ACCURACY EVALUATION OF ARCUATE FASCICULUS OF DIFFUSION TENSOR TRACTOGRAPHY BY FUSION OF REAL SPACE AND VIRTUAL SPACE

PURPOSE Diffusion tensor-based tractography (DTT) is a method to estimate the direction of white matter fibers, but it is difficult to verify the relationship with brain function spatially with high accuracy. We developed a registration method to fuse the real space (brain surface photograph) and preoperative fused 3D image (virtual space) using the landmark method and thin plate spline method. In our previous study, this method was able to achieve highly accurate alignment registration error 0.7±0.1mm (mean±SE) even after brain shift due to craniotomy. In this study, we proposed a method to examine spatial errors of DTT and direct cortical stimulation (DCS) and verify its accuracy. METHODS We included 7 gliomas performed awake surgery. We created the fused three – dimensional image before surgery and acquired the brain surface photograph immediately after craniotomy, then we aligned them using the proposed method. Sites that showed speech arrest by DCS were plotted on the fused image. A circle with a radius of 15 mm centered on the same site was taken as the range over which the current spreads. The surface area of each of the circles was calculated to make it true if there was arcuate fasciculus drawn with DTT in the circle, and false if it did not exist. By using this method, the accuracy of the DTT was verified. RESULT: In 7 cases, speech arrest was shown at 21 DCS plots. The probability of the presence of DTT within the current spread of DCS was 64.4%. CONCLUSION The proposed method indicates that DTT does not necessarily match the DCS results by verification using real space and virtual space. We present some illustrative cases.