Subject selection framework to improve personalised models for motor-imagery BCIs via wavelets and graph diffusion
File(s)
Published version
OA Location
Author(s)
Barmpas, Konstantinos
Panagakis, Ioannis
Adamos, Dimitrios
Laskaris, Nikolaos
Zafeiriou, Stefanos
Type
Conference Paper
Abstract
Personalized electroencephalogram (EEG) decoders hold a distinct preference in healthcare applications, especially in the context of Motor-Imagery (MI) Brain-Computer Interfaces (BCIs), owing to their inherent capability to effectively tackle inter-subject variability. This study introduces a novel subject selection framework that blends ideas from discriminative learning (based on continuous wavelet transform) and graph-signal processing (over the sensor array). Through experimentation with a publicly available MI dataset, we showcase enhanced personalized performance for MI-BCIs. Notably, it proves particularly advantageous for subjects who initially demonstrated suboptimal personalized performance.
Date Issued
2024-03-08
Date Acceptance
2024-03-05
Citation
ICLR2024 Workshop on Learning from Time Series for Health, 2024, pp.1-7
Start Page
1
End Page
7
Journal / Book Title
ICLR2024 Workshop on Learning from Time Series for Health
Copyright Statement
© 2024 The Author(s). This poaper is available under a CC-BY licence (https://creativecommons.org/licenses/by/4.0/)
License URL
Source
ICLR2024 Workshop on Learning from Time Series for Health
Publication Status
Published
Start Date
2024-05-11
Coverage Spatial
Vienna, Austria
Date Publish Online
2024-03-08