Comparison of Models for EEG-based Cross-Subject Movement Prediction
Infinite Science GmbH · 21.01.2025
DOI: 10.18416/scp.2025.1947 ↗Kurz gefasst
Fachartikel in der Zeitschrift Infinite Science GmbH, erschienen am 21.01.2025. Verfasst von Moosdorf und 4 weiteren Autor:innen.
Abstract (Originalsprache)
Brain-computer interfaces for robotic-assisted rehabilitation based on EEG recordings can play an important role in the rehabilitation of stroke patients. To effectively classify movement intentions from EEG signals, machine learning models are needed that can work with noisy and heterogeneous data. This work investigates four neural networks, namely the MLP, TKAN, EEGNet, and SincEEGNet, in their ability to predict movement intentions in intra- and cross-subject classification. Additionally, the amount of calibration data for adapting the pre-trained models of the cross-subject task to data of the target subject was explored. The model performance for intra-subject classification is highest with the EEGNet, however, SincEEGNet performs best in cross-subject classification. Calibrating the pre-trained models results in a performance gain with the overall highest accuracy of 88.1 % with SincEEGNet. Those results motivate to apply such models for EEG classification across subjects to reduce calibration times.
Schlagwörter
Zitieren
APA
Moosdorf, Laeticia, Kueper, N., A, H. H., & Kirchner, E. A. (2025). Comparison of Models for EEG-based Cross-Subject Movement Prediction. Infinite Science GmbH. https://doi.org/10.18416/scp.2025.1947
Vancouver
Moosdorf, Laeticia, Kueper N, A HH, Kirchner EA. Comparison of Models for EEG-based Cross-Subject Movement Prediction. Infinite Science GmbH. 2025. doi:10.18416/scp.2025.1947
BibTeX
@article{Moosdorf2025,
author = {Moosdorf and Laeticia and Kueper, Niklas and A, Heinz Handels and Kirchner, Elsa Andrea},
title = {Comparison of Models for EEG-based Cross-Subject Movement Prediction},
journal = {Infinite Science GmbH},
year = {2025},
doi = {10.18416/scp.2025.1947}
}
EUVIMED