Search papers, labs, and topics across Lattice.
This paper introduces EEGForceFusion, a novel hybrid framework for continuous grasp force decoding that effectively combines tokenised and continuous representations to address the challenges of inter-subject variability and temporal dynamics in EEG data. By employing convolutional-recurrent networks alongside transformer-based models, the approach captures both fine-grained neural structures and long-range dependencies, achieving impressive R虏 scores of 0.817 in offline settings and 0.793 in simulated real-time scenarios. The results underscore the framework's strong generalisation capabilities across subjects, making it a promising solution for applications in assistive robotics and neuro-rehabilitation.
Achieving R虏 scores above 0.8 for grasp force decoding demonstrates a breakthrough in cross-subject generalisation using hybrid EEG representations.
Brain-machine interfaces provide a link between neural activity and external devices, enabling restoration of motor function and advancing human-machine interaction using non-invasive electroencephalography (EEG). However, continuous grasp force decoding remains challenging due to complex temporal dynamics, high inter-subject variability, and limited generalisation of existing approaches. To address this, we propose a hybrid EEG decoding framework that jointly models continuous and tokenised representations, enabling capture of both fine-grained neural structure and long-range temporal dependencies. The proposed approach integrates convolutional-recurrent representation learning, quantisation-based tokenisation, and transformer-based temporal modelling within a unified fusion-based regression architecture. Experimental evaluation on the WAY-EEG-GAL dataset under strict leave-one-subject-out conditions achieves $R^2$ = 0.817 in offline settings and $R^2$ = 0.793 in simulated real-time evaluation, with latency suitable for real-time deployment. These results demonstrate strong cross-subject generalisation and highlight the practicality of hybrid continuous-tokenised representations for real-time EEG-based force decoding in assistive robotics, neuro-rehabilitation, and human-machine interaction.
