PA-TCNet: Pathology-Aware Temporal Calibration with Physiology-Guided Target Refinement for Cross-Subject Motor Imagery EEG Decoding in Stroke Patients

arXiv cs.CV / 4/21/2026

📰 NewsSignals & Early TrendsTools & Practical UsageModels & Research

共有:

Key Points

The paper introduces PA-TCNet, a pathology-aware temporal calibration framework designed to improve cross-subject EEG decoding of motor imagery for stroke patients despite lesion-induced temporal abnormalities and strong inter-patient heterogeneity.
PA-TCNet uses two coordinated components: PRSM to separate slowly varying rhythmic context from fast transient perturbations and propagate pathological context more effectively, and PGTC to build physiology-consistent ROI templates that refine target-domain pseudo-labels.
The approach specifically targets common failure modes of existing adaptation methods, including being misled by pathological slow-wave activity and by unstable pseudo-labels in the target domain.
Leave-one-subject-out experiments on two datasets (XW-Stroke and 2019-Stroke) report mean accuracies of 66.56% and 72.75%, respectively, outperforming prior state-of-the-art baselines.
The authors provide an implementation via GitHub and argue that combining pathological temporal modeling with physiology-constrained pseudo-supervision yields more robust initialization for personalized post-stroke MI-BCI rehabilitation.

Abstract

Stroke patient cross-subject electroencephalography (EEG) decoding of motor imagery (MI) brain-computer interface (BCI) is essential for motor rehabilitation, yet lesion-related abnormal temporal dynamics and pronounced inter-patient heterogeneity often undermine generalization. Existing adaptation methods are easily misled by pathological slow-wave activity and unstable target-domain pseudo-labels. To address this challenge, we propose PA-TCNet, a pathology-aware temporal calibration framework with physiology-guided target refinement for stroke motor imagery decoding. PA-TCNet integrates two coordinated components. The Pathology-aware Rhythmic State Mamba (PRSM) module decomposes EEG spatiotemporal features into slowly varying rhythmic context and fast transient perturbations, injecting the fused pathological context into selective state propagation to more effectively capture abnormal temporal dynamics. The Physiology-Guided Target Calibration (PGTC) module constructs source-domain sensorimotor region-of-interest templates, imposing physiological consistency constraints and dynamically refining target-domain pseudo-labels, thereby improving adaptation reliability. Leave-one-subject-out experiments on two independent stroke EEG datasets, XW-Stroke and 2019-Stroke, yielded mean accuracies of 66.56\% and 72.75\%, respectively, outperforming state-of-the-art baselines. These results indicate that jointly modeling pathological temporal dynamics and physiology-constrained pseudo-supervision can provide more robust cross-subject initialization for personalized post-stroke MI-BCI rehabilitation. The implemented code is available at https://github.com/wxk1224/PA-TCNet.