Generative Augmentation for EEG Motor Imagery Classification: A Class-Conditional VAE with Cycle-Consistent Decoder Refinement

2026-07-22

AI Summary

This study addresses the challenge of limited labeled data in electroencephalography (EEG)-based motor imagery (MI) classification by proposing a novel class-conditional variational autoencoder (CVAE) framework. The approach explicitly preserves the Riemannian geometry of EEG signals during generation—not by mimicking raw waveforms, but through a covariance matrix constraint and a cycle-consistent decoder refinement strategy. This enables the synthesis of label-consistent and geometrically plausible EEG samples for data augmentation. Experimental results demonstrate that, under both cross-subject and within-subject settings, the proposed method yields modest yet consistent performance gains for covariance-structure-dependent classifiers such as Minimum Distance to Mean (MDM), thereby validating the efficacy and potential of structure-aware generative modeling in EEG data augmentation.

Abstract

We investigate whether a generative model can supply useful synthetic motor-imagery (MI) electroencephalography (EEG) trials that improve the accuracy of independent downstream classifiers. We train a class-conditional variational autoencoder (CVAE) with an integrated latent classifier on the Zhou motor-imagery dataset, using the learned per-class prior as a generator: sampling the prior for a given label and decoding it into a synthetic, label-consistent signal. A constraint on the covariance matrix of the generated data encourages preservation of covariance structure, and the model is trained with a schedule that alternates ordinary VAE training with a decoder-focused phase that sharpens the generative pathway used for augmentation. We measure the effect of adding synthetic trials to the training set under two evaluation protocols -- within-user (pooled 60/20/20 split across subjects) and cross-user (leave-one-subject-out, LOSO) -- across four representative EEG classification pipelines: Common Spatial Patterns with Linear Discriminant Analysis (CSP+LDA), tangent-space features with a Support Vector Machine (TGSP+SVM), Minimum Distance to Riemannian Mean (MDM), and a neural network based on EEGNetv4 (henceforth EEGNet). Results are aggregated across independent augmentation draws, random seeds (within-user), or leave-one-subject-out folds (cross-user), with uncertainty reported as 95\% confidence intervals (Student's $t$-distribution) computed over per-seed/per-fold averages. We find that synthetic EEG from the CVAE is most credible as a source of class-structured, covariance-like data rather than as a substitute for real raw EEG: it can raise the point estimate for MDM, but the broader augmentation claim remains conservative -- observed gains are small and classifier-dependent.

Problem

• EEG
• Motor Imagery
• Generative Augmentation
• Classification
• Synthetic Data

Innovation

• class-conditional VAE
• covariance-preserving generation
• decoder refinement
• EEG data augmentation
• motor imagery classification