EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Comparison and extension of the Riemannian Procrustes analysis and tangent space alignment transfer learning methods

Martin Gemborn Nilsson iD, Frida Heskebeck iD

Journal of Neural Engineering · 16.11.2026

DOI: 10.1088/1741-2552/aec2aa ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Neural Engineering, online erschienen am 16.11.2026. Verfasst von Martin Gemborn Nilsson und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Automatic Control, Lund University Faculty of Engineering, Ole Römers väg 1, Lund, 22100, Sweden. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Excellence Center at Linköping – Lund in Information Technology, Knut och Alice Wallenbergs Stiftelse.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective. This study compares Riemannian Procrustes analysis (RPA) and tangent space alignment (TSA) for transfer learning of motor-imagery EEG data. The study also examines how performance changes when the EEG data is represented using analytic Hermitian positive definite covariance matrices instead of standard symmetric positive definite covariance matrices.

Approach. The methods' geometric constraints were compared and explained. Additionally, the methods were evaluated in two scenarios on real EEG data, one with substantial target-domain knowledge and one calibration-like with a limited amount of training data and thus limited target-domain knowledge. Six public EEG datasets with multiple subjects performing two or multiple motor imagery tasks were used. We report paired comparisons of classification performance for the different transfer learning methods and covariance matrix representations across the datasets in order to draw general conclusions about performance.

Main results. In the scenario with substantial target-domain knowledge, the TSA method outperformed the RPA method, and representing the EEG data as analytical covariance matrices outperformed standard covariance matrices. In the scenario with limited target-domain knowledge, performance depended on the amount of available target data, and source-data selection emerged as an important component of successful transfer learning. 

Significance. This study gives a deeper understanding of TSA and RPA transfer learning methods commonly used in brain-computer interfaces, which may inform future research and implementation. The analysis also demonstrates the importance of source data selection in brain-computer interface transfer learning.

Schlagwörter

Stichwörter der Autor:innen

Riemannian geometrybrain-computer interfacecovariance matrix representationmultivariate time-series datatransfer learning

Autor:innen und Institutionen

  • Department of Automatic Control, Lund University Faculty of Engineering, Ole Römers väg 1, Lund, 22100, Sweden.
  • Department of Automatic Control, Lunds Tekniska Hogskola, John Ericsons väg 3, Lund, 221 00, Sweden.

ORCID-Profile: Martin Gemborn Nilsson, Frida Heskebeck

Förderung

  • Excellence Center at Linköping – Lund in Information Technology
  • Knut och Alice Wallenbergs Stiftelse

Zitieren

APA

Gemborn Nilsson, M., & Heskebeck, F. (2026). Comparison and extension of the Riemannian Procrustes analysis and tangent space alignment transfer learning methods. Journal of Neural Engineering. https://doi.org/10.1088/1741-2552/aec2aa

Vancouver

Gemborn Nilsson M, Heskebeck F. Comparison and extension of the Riemannian Procrustes analysis and tangent space alignment transfer learning methods. Journal of Neural Engineering. 2026. doi:10.1088/1741-2552/aec2aa

BibTeX
@article{GembornNilsson2026,
  author = {Gemborn Nilsson, Martin and Heskebeck, Frida},
  title = {Comparison and extension of the Riemannian Procrustes analysis and tangent space alignment transfer learning methods},
  journal = {Journal of Neural Engineering},
  year = {2026},
  doi = {10.1088/1741-2552/aec2aa}
}
Kostenlos starten →