Abstracts

Rationale: Functional brain mapping prior to a brain surgery is an important step for optimizing the surgical outcome. Like electrical cortical stimulation (ECS) mapping, event-related activation in the high gamma frequency band (HGA) highlights functional regions of the cortex. This HGA also occurs during face recognition in the fusiform face area (FFA). Since ECS on the temporal base can be problematic for the patient, an alternative method using electrocorticography can overcome ECS-related issues like pain or seizures. Methods: In this work we used a real-time mapping system (cortiQ, g.tec, Austria) to map the FFA. The system identifies activated cortical areas within minutes and brings the great advantage that the patient can perform tasks voluntarily, without the danger of stimulation-elicited seizures. The study included ECS and cortiQ mapping with two patients who underwent neuro-monitoring prior to brain surgery. Subject S1 had 144 electrodes with 2.3 mm exposure diameters and a 10 mm inter-electrode distance that covered the right ventral stream (VS), and S2 had 188 sites with high-density grids having 1.5 mm exposure diameters and a 5 mm inter-electrode distance, which were implanted across the VS of both hemispheres. During the cortiQ mapping, S1 looked at a black screen for 25.6 s and then at 32 alternating presentations of faces and a blank screen for 0.4 s each. S2 looked at scrambled faces for 25.6 s and then 32 times at a face and a scrambled face for 0.4 s each. Both subjects repeated the mapping three times, resulting in a recording time of 153.6 s. For the ECS mapping, the subjects were asked to report any stimulation-related change in perception while looking at real faces and objects. Results: Figure 1 shows face-related HGA at specific electrode positions, where significant activation is indicated by a yellow border around the electrode. While S1 showed activation on the dorsal stream (left, visual processing area) and VS (middle, face/symbol recognition area), S2 showed activation on the VS and tongue motor region. The R represent the ratio of change in signal variance that is due to the face stimulation. Validation with ECS gave a sensitivity/specificity of 75/97 % and 100/76 % for S1 and S2, respectively. Conclusions: This study shows that FFA can be mapped within a short recording time, adding to a large battery of mapping procedures that can already be accomplished using cortiQ. These include localization of sensorimotor, receptive and expressive language, and auditory areas. In the future, localization of areas engaged during memory function will also be investigated. Funding: The research was supported by the European Union Eurostars Project RAPIDMAPS 2020 (No. 9 273) .