Synthesizing EEG Signals from Event-Related Potential Paradigms with Conditional Diffusion Models

arXiv:2403.18486v1 Announce Type: new
Abstract: Data scarcity in the brain-computer interface field can be alleviated through the use of generative models, specifically diffusion models. While diffusion models have previously been successfully applied to electroencephalogram (EEG) data, existing models lack flexibility w.r.t.~sampling or require alternative representations of the EEG data. To overcome these limitations, we introduce a novel approach to conditional diffusion models that utilizes classifier-free guidance to directly generate subject-, session-, and class-specific EEG data. In addition to commonly used …

abstract arxiv brain brain-computer interface computer cs.ai cs.lg data diffusion diffusion models eeg eess.sp event flexibility generative generative models sampling through type