ECoG based cortical function mapping using general linear model

Electrocorticography (ECoG) is an emerging tool to map brain functions in the context of neurosurgical intervention. Previous mapping methods based on the event related power spectrum are prone to noise. To improve the robustness of cortical function mapping, general linear model (GLM), which has been widely used in the analysis of functional magnetic resonance imaging (fMRI) data, is applied to bandpass filtered ECoG signals from each electrode. For a specific task, electrodes with best fitting parameters of the signal are identified, and the statistical significance of the fitting is mapped on the standard 3D brain model to provide a personalized map of sensorimotor functions. With the analysis of four patients´ data, the proposed approach yields consistent results with those obtained by electrical cortical stimulation (ECS), while showing promising performance against noise.