Separating Diagnosis from Disease Representation: Dual-View EEG Learning with Neural-Dynamics-Guided Deformation

2026-09-26

AI Summary

This study addresses the challenge of simultaneously preserving anatomical structure and predictive signals in EEG-based diagnosis by proposing the DMD-EEG framework. Architecturally, it decouples scalp-level diagnostics from source-space representations, fusing them exclusively at the decision layer. By incorporating neural dynamics priors through low-rank sparse iterative optimization and dual-view multi-scale deformation, the method explicitly models 46 brain regions, five frequency bands, and four time lags to generate interpretable ROI-frequency-lag attributions. Experimental results demonstrate superior performance on major depressive disorder (MDD), first-episode psychosis (FEP), and Parkinson’s disease (PD) classification tasks. The derived saliency coordinates align closely with established pathological circuits, and the framework exhibits robust cross-montage transferability.

Abstract

Electroencephalography (EEG)-based closed-loop neuromodulation calls for a subject-specific structured state, as opposed to a single disease probability, specifying which brain regions are deviant, at which frequencies, and at which lags. Sensor-space models keep the strongest diagnostic evidence without anatomy, source-space models give anatomy at a loss of predictive signal, and post-hoc attributions stay outside the prediction. We separate the two instead of forcing them into one representation, and propose DMD-EEG (Dual-view Multiscale Deformation for EEG), which keeps a fixed scalp spectral expert for diagnosis and models the source-space disease-related representation as a low-rank, sparse, iterative deformation of a healthy neural-dynamics prior in a $46$-region-of-interest (ROI) $\times$ $5$-frequency $\times$ $4$-lag (autocorrelation-timescale) space. The two experts meet only at a fixed decision level, so the source state is architecturally separate from the scalp expert. Across major depressive disorder (MDD), first-episode psychosis (FEP), and Parkinson's disease (PD), decision-level fusion matches the strongest single expert on MDD and FEP and exceeds the source branch on PD. On FEP the source expert is the strongest branch, the task where the deformation contributes most. The source state is an explicit ROI-frequency-lag attribution defined in a shared source coordinate system across montages, which we treat as an anatomically-coordinated predictive representation whose coordinates are directly readable and hypothesis-generating. The highest-saliency coordinates align with established disease circuitry (fronto-limbic-temporal regions in MDD, motor-cortex beta in PD), and the MDD state transfers by rank to an unseen cohort recorded with a different montage.

Problem

• EEG
• closed-loop neuromodulation
• disease representation
• source-space modeling
• interpretability

Innovation

• Dual-view EEG learning
• Neural-dynamics-guided deformation
• Source-space representation
• Decision-level fusion
• Low-rank sparse modeling