Effective Connectivity-Based Neural Decoding: A Causal Interaction-Driven Approach
This work addresses brain decoding for neuroscience applications, presenting an incremental improvement by applying a novel causality measure to MEG data.
The authors tackled the problem of decoding visual stimuli from brain activity by developing a model-free causality measure for detecting linear and nonlinear interactions in time series, which they applied to MEG data to construct effective connectivity maps. They found that these maps for structured images show more geometric patterns, as revealed through persistent homology analysis.
We propose a geometric model-free causality measurebased on multivariate delay embedding that can efficiently detect linear and nonlinear causal interactions between time series with no prior information. We then exploit the proposed causal interaction measure in real MEG data analysis. The results are used to construct effective connectivity maps of brain activity to decode different categories of visual stimuli. Moreover, we discovered that the MEG-based effective connectivity maps as a response to structured images exhibit more geometric patterns, as disclosed by analyzing the evolution of toplogical structures of the underlying networks using persistent homology. Extensive simulation and experimental result have been carried out to substantiate the capabilities of the proposed approach.