Neuroactivation imaging using a monogenic framework

Intentions of specific motor movements are known to generate event-related (de)synchronization (ERD/ERS) patterns which may be interpreted as changes in the degree of synchronization of underlying neuronal networks. When activated, the neural populations in a certain region of the brain de-synchronize, leading to a decrease in the EEG power signal. Notably, this phenomena happens not only in time but also in space. Here, we propose a novel method to detect brain activity based on this spatial desynchronization effect. The electrical activity at the scalp is described by a finite dimension continuous 2D field estimated from the EEG data acquired at the discrete locations of the electrodes and analyzed using the monogenic framework. The method was tested with synthetic data and the results are compared with other methods, namely, the traditional power-based ERD/ERS estimation method. It is shown that the proposed approach yields more space-specific and less blurred results than traditional approaches.