Transfer learning promotes acquisition of individual BCI skills
PNAS Nexus · 01.02.2024 · 3 (2)
DOI: 10.1093/pnasnexus/pgae076 ↗Kurz gefasst
Fachartikel in der Zeitschrift PNAS Nexus, online erschienen am 16.02.2024. Verfasst von Satyam Kumar und 4 weiteren Autor:innen. Beteiligte Einrichtung: Chandra Family Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin, TX 78712, USA. Die Arbeit wurde bisher 52-mal zitiert (Crossref) und nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Coleman Fung Foundation, Sinclair Foundation.
Abstract (Originalsprache)
Subject training is crucial for acquiring brain–computer interface (BCI) control. Typically, this requires collecting user-specific calibration data due to high inter-subject neural variability that limits the usability of generic decoders. However, calibration is cumbersome and may produce inadequate data for building decoders, especially with naïve subjects. Here, we show that a decoder trained on the data of a single expert is readily transferrable to inexperienced users via domain adaptation techniques allowing calibration-free BCI training. We introduce two real-time frameworks, (i) Generic Recentering (GR) through unsupervised adaptation and (ii) Personally Assisted Recentering (PAR) that extends GR by employing supervised recalibration of the decoder parameters. We evaluated our frameworks on 18 healthy naïve subjects over five online sessions, who operated a customary synchronous bar task with continuous feedback and a more challenging car racing game with asynchronous control and discrete feedback. We show that along with improved task-oriented BCI performance in both tasks, our frameworks promoted subjects’ ability to acquire individual BCI skills, as the initial neurophysiological control features of an expert subject evolved and became subject specific. Furthermore, those features were task-specific and were learned in parallel as participants practiced the two tasks in every session. Contrary to previous findings implying that supervised methods lead to improved online BCI control, we observed that longitudinal training coupled with unsupervised domain matching (GR) achieved similar performance to supervised recalibration (PAR). Therefore, our presented frameworks facilitate calibration-free BCIs and have immediate implications for broader populations—such as patients with neurological pathologies—who might struggle to provide suitable initial calibration data.
Autor:innen und Institutionen
- Chandra Family Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin, TX 78712, USA.
- Department of Neurology, The University of Texas at Austin, Austin, TX 78712, USA.
- Mulva Clinic for the Neurosciences, The University of Texas at Austin, Austin, TX 78712, USA.
- Departement of Biomedical Engineering, The University of Texas at Austin, Austin, TX 78712, USA.
ORCID-Profile: Satyam Kumar, Hussein Alawieh, Frigyes Samuel Racz, Rawan Fakhreddine, José Del R Millán
Förderung
- Coleman Fung Foundation
- Sinclair Foundation
Zitieren
APA
Kumar, S., Alawieh, H., Racz, F. S., Fakhreddine, R., & Millán, J. D. R. (2024). Transfer learning promotes acquisition of individual BCI skills. PNAS Nexus, 3(2). https://doi.org/10.1093/pnasnexus/pgae076
Vancouver
Kumar S, Alawieh H, Racz FS, Fakhreddine R, Millán JDR. Transfer learning promotes acquisition of individual BCI skills. PNAS Nexus. 2024;3(2). doi:10.1093/pnasnexus/pgae076
BibTeX
@article{Kumar2024,
author = {Kumar, Satyam and Alawieh, Hussein and Racz, Frigyes Samuel and Fakhreddine, Rawan and Millán, José Del R},
title = {Transfer learning promotes acquisition of individual BCI skills},
journal = {PNAS Nexus},
year = {2024},
volume = {3},
number = {2},
doi = {10.1093/pnasnexus/pgae076}
}
EUVIMED