English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Brain-Machine Interfacing (BMI) has greatly benefited from adopting machinelearning methods for feature learning that require extensive data for training,which are often unavailable from a single dataset. Yet, it is difficult tocombine data across labs or even data within the same lab collected over theyears due to the variation in recording equipment and electrode layoutsresulting in shifts in data distribution, changes in data dimensionality, andaltered identity of data dimensions. Our objective is to overcome thislimitation and learn from many different and diverse datasets across labs withdifferent experimental protocols. To tackle the domain adaptation problem, wedeveloped a novel machine learning framework combining graph neural networks(GNNs) and transfer learning methodologies for non-invasive Motor Imagery (MI)EEG decoding, as an example of BMI. Empirically, we focus on the challenges oflearning from EEG data with different electrode layouts and varying numbers ofelectrodes. We utilise three MI EEG databases collected using very differentnumbers of EEG sensors (from 22 channels to 64) and layouts (from customlayouts to 10-20). Our model achieved the highest accuracy with lower standarddeviations on the testing datasets. This indicates that the GNN-based transferlearning framework can effectively aggregate knowledge from multiple datasetswith different electrode layouts, leading to improved generalization insubject-independent MI EEG classification. The findings of this study haveimportant implications for Brain-Computer-Interface (BCI) research, as theyhighlight a promising method for overcoming the limitations posed bynon-unified experimental setups. By enabling the integration of diversedatasets with varying electrode layouts, our proposed approach can help advancethe development and application of BMI technologies.

EEG Decoding for Datasets with Heterogenous Electrode Configurations using Transfer Learning Graph Neural Networks