Dillen, A., Steckelmacher, D., Efthymiadis, K., Langlois, K., De Beir, A., Marusic, U., Vanderborght, B., Nowe, A., Meeusen, R., Ghaffari, F., Romain, O., & De Pauw, K. (2022). Deep learning for biosignal control: insights from basic to real-time methods with recommendations. Journal of Neural Engineering, 19(1), Article 011003. https://doi.org/10.1088/1741-2552/ac4f9a

@article{583e72997aad4a51af23b22d0ed4e0fe,
title = "Deep learning for biosignal control: insights from basic to real-time methods with recommendations",
abstract = "Objective.Biosignal control is an interaction modality that allows users to interact with electronicdevices by decoding the biological signals emanating from the movements or thoughts of the user. Thismanner of interaction with devices can enhance the sense of agency for users and enable persons sufferingfrom a paralyzing condition to interact with everyday devices that would otherwise be challenging forthem to use. It can also improve control of prosthetic devices and exoskeletons by making the interactionfeel more natural and intuitive. However, with the current state of the art, several issues still need to beaddressed to reliably decode user intent from biosignals and provide an improved user experience overother interaction modalities. One solution is to leverage advances in Deep Learning (DL) methods toprovide more reliable decoding at the expense of added computational complexity. This scoping reviewintroduces the basic concepts of DL and assists readers in deploying DL methods to a real-time controlsystem that should operate under real-world conditions.Approach.The scope of this review covers anyelectronic device, but with an emphasis on robotic devices, as this is the most active area of research inbiosignal control. We review the literature pertaining to the implementation and evaluation of controlsystems that incorporate DL to identify the main gaps and issues in the field, and formulate suggestionson how to mitigate them.Main results.The results highlight the main challenges in biosignal controlwith deep learning methods. Additionally, we were able to formulate guidelines on the best approachto designing, implementing and evaluating research prototypes that use DL in their biosignal controlsystems.Significance.This review should assist researchers that are new to the fields of biosignalcontrol and deep learning in successfully deploying a full biosignal control system. Experts in theirrespective fields can use this article to identify possible avenues of research that would further advancethe development of biosignal control with deep learning methods.",
keywords = "Deep Learning, Biosignal processing, control systems, Artificial intelligence, Human Computer Interaction",
author = "Arnau Dillen and Denis Steckelmacher and Kyriakos Efthymiadis and Kevin Langlois and {De Beir}, Albert and Uros Marusic and Bram Vanderborght and Ann Nowe and Romain Meeusen and Fakhreddine Ghaffari and Olivier Romain and {De Pauw}, Kevin",
year = "2022",
month = feb,
day = "28",
doi = "10.1088/1741-2552/ac4f9a",
language = "English",
volume = "19",
journal = "Journal of Neural Engineering",
issn = "1741-2552",
publisher = "IOP Publishing",
number = "1",
}