Biomedical Signal Processing and Control· 2026Q1
Classification-aware conditional diffusion-augmented criss-cross graph transformer for EEG-based emotion recognition
- 0citations
- Q1SCImago
- 2026year
Short summary
A novel framework couples EEG generation and classification via asymmetric alternating optimization, using conditional diffusion to create synthetic EEG data that improves subject-independent emotion recognition accuracy by expanding the classifier's training distribution.
AI-generated from the title and abstract; the full text is not read.
Key points
- A novel framework uses asymmetric alternating optimization to couple EEG generation and classification for emotion recognition.
- Conditional diffusion augmentation with gated emotion-subject-time conditioning generates synthetic EEG samples to improve classifier training.
- A criss-cross graph ODE model captures temporal and inter-channel dependencies for refined emotion prediction.
- Experiments on DEAP, SEED, and DREAMER datasets show consistent improvements over conventional and generative augmentation baselines.
AI-generated from the title and abstract; the full text is not read.
Abstract
Reliable electroencephalogram (EEG)-based emotion recognition remains challenging because labeled recordings are limited, noisy, non-stationary, and strongly subject dependent. This paper proposes a classification-aware conditional diffusion-augmented criss-cross graph transformer for raw EEG emotion recognition. Unlike generation-centered augmentation methods, the proposed framework couples EEG generation and classification through asymmetric alternating optimization. Detached multi-step synthetic samples expand the classifier training distribution, while differentiable one-step clean estimates receive feedback from a temporarily fixed classifier to improve the task relevance of the generator. The CondDiffEEG generator employs gated emotion-subject-time conditioning for subject-calibrated synthesis, while orthogonality regularization discourages linear overlap between the projected emotion and subject conditions. Temporal and frequency-domain constraints further preserve essential signal characteristics. For classification, CGODE-CCT first factorizes temporal and inter-channel dependencies through criss-cross attention, then models continuously evolving electrode interactions using a dynamically updated graph ODE, and finally refines the graph-enhanced representations for emotion prediction. Experiments on DEAP, SEED, and DREAMER under subject-independent and subject-dependent protocols demonstrate consistent improvements over conventional classifiers, EEG Transformer models, and generative augmentation baselines. Quantitative evaluation further indicates improved spectral consistency, real-synthetic distributional similarity, downstream transferability, and sample diversity. These findings demonstrate that task-oriented coordination between conditional EEG generation and continuous graph-based classification provides an effective approach to robust EEG emotion recognition.
The authors' abstract, as published at the source. Biomedical Signal Processing and Control, 2026 · DOI ↗
The rest is in the Pofolia app
Takeaways and questions to the paper; new summaries every day for your field. Free.
Sign in on the web to openField: Experimental and Cognitive Psychology
Experimental and Cognitive PsychologyPsychology