Riemannian Approaches in Brain-Computer Interfaces: A Review

Yger, Florian; Bérar, Maxime; Lotte, Fabien

doi:10.1109/tnsre.2016.2627016

Cited by 297 publications

(243 citation statements)

References 63 publications

Supporting

Mentioning

240

Contrasting

Unclassified

Order By: Relevance

“…A more recent single-step approach consists in learning directly from spatially correlated power-spectra with linear models and Riemannian geometry (Barachant et al, 2011Yger et al, 2017). This mathematical framework provides principles to correct for the geometric distortions arising from linear mixing of non-linear sources.…”

Section: State-of-the Art Approaches To Predict From M/eeg Observationsmentioning

confidence: 99%

Predictive regression modeling with MEG/EEG: from source power to signals and cognitive states

Sabbagh

Ablin

Varoquaux

et al. 2019

Preprint

View full text Add to dashboard Cite

Predicting biomedical outcomes from Magnetoencephalography and Electroencephalography (M/EEG) is central to applications like decoding, brain-computer-interfaces (BCI) or biomarker development and is facilitated by supervised machine learning. Yet most of the literature is concerned with within-subject classification. Here, we focus on predicting continuous outcomes from M/EEG signal power across subjects. Considering different generative mechanisms for M/EEG signals and the biomedical outcome, we propose statistically-consistent predictive models that avoid sourcereconstruction based on the covariance as representation. Our mathematical analysis and ground truth simulations demonstrated that consistent parameter estimation can be obtained with Source Power Comodulation (SPoC) supervised spatial filtering or by embedding with Riemannian geometry. Additional simulations revealed that Riemannian methods were more robust to model violations, in particular geometric distortions induced by individual anatomy. To estimate the relative contribution of brain dynamics and anatomy to prediction performance, we propose a novel model inspection procedure based on biophysical forward modeling. Applied to cross-subject prediction, the analysis revealed that the Riemannian model better exploited anatomical information while sensitivity to brain dynamics was similar across methods. We then probed the robustness of the models across different data cleaning options. Environmental denoising was globally important but Riemannian models were strikingly robust and continued performing even without preprocessing. Our results suggest each method has its niche: SPoC is practical for within-subject prediction while the Riemannian model may enable simple end-to-end learning.

show abstract

Section: State-of-the Art Approaches To Predict From M/eeg Observationsmentioning

confidence: 99%

Predictive regression modeling with MEG/EEG: from source power to signals and cognitive states

Sabbagh

Ablin

Varoquaux

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…In order to compute a distance of a newly arriving ERP trial sample from the above mentioned class-characterizing mean covariance matrix k an appropriate metric, allowing a simple discrimination, is employed. A point on a Riemannian manifold represents the symmetric positive definite matrix (Barachant et al, 2012; Barachant and Congedo, 2014; Yger et al, 2016). …”

Section: Methodsmentioning

confidence: 99%

Robotic and Virtual Reality BCIs Using Spatial Tactile and Auditory Oddball Paradigms

Rutkowski

2016

Front. Neurorobot.

View full text Add to dashboard Cite

The paper reviews nine robotic and virtual reality (VR) brain–computer interface (BCI) projects developed by the author, in collaboration with his graduate students, within the BCI–lab research group during its association with University of Tsukuba, Japan. The nine novel approaches are discussed in applications to direct brain-robot and brain-virtual-reality-agent control interfaces using tactile and auditory BCI technologies. The BCI user intentions are decoded from the brainwaves in realtime using a non-invasive electroencephalography (EEG) and they are translated to a symbiotic robot or virtual reality agent thought-based only control. A communication protocol between the BCI output and the robot or the virtual environment is realized in a symbiotic communication scenario using an user datagram protocol (UDP), which constitutes an internet of things (IoT) control scenario. Results obtained from healthy users reproducing simple brain-robot and brain-virtual-agent control tasks in online experiments support the research goal of a possibility to interact with robotic devices and virtual reality agents using symbiotic thought-based BCI technologies. An offline BCI classification accuracy boosting method, using a previously proposed information geometry derived approach, is also discussed in order to further support the reviewed robotic and virtual reality thought-based control paradigms.

show abstract

“…Number of iterations N ; Weights α, β, ρ; Dimensionality of the shared subspace, p. Output:Ŷ T ∈ R nT ×l , the labels for {X T,i } nT i=1 . Calculate the covariance matrices {P S,i } nS i=1 and their mean matrix M in the source domain, using (6), (7), or (8); (14); Map each P ′ S,i to a tangent space feature vector x S,i ∈ R d×1 using (19) Then, the transferability of Source Domain S i is computed as:…”

Section: Algorithm 1: Manifold Embedded Knowledge Transfer (Mekt)mentioning

confidence: 99%

Manifold Embedded Knowledge Transfer for Brain-Computer Interfaces

Zhang

2020

IEEE Trans. Neural Syst. Rehabil. Eng.

105

101

View full text Add to dashboard Cite

Transfer learning makes use of data or knowledge in one problem to help solve a different, yet related, problem. It is particularly useful in brain-computer interfaces (BCIs), for coping with variations among different subjects and/or tasks. This paper considers offline unsupervised cross-subject electroencephalogram (EEG) classification, i.e., we have labeled EEG trials from one or more source subjects, but only unlabeled EEG trials from the target subject. We propose a novel manifold embedded knowledge transfer (MEKT) approach, which first aligns the covariance matrices of the EEG trials in the Riemannian manifold, extracts features in the tangent space, and then performs domain adaptation by minimizing the joint probability distribution shift between the source and the target domains, while preserving their geometric structures. MEKT can cope with one or multiple source domains, and can be computed efficiently. We also propose a domain transferability estimation (DTE) approach to identify the most beneficial source domains, in case there are a large number of source domains. Experiments on four EEG datasets from two different BCI paradigms demonstrated that MEKT outperformed several stateof-the-art transfer learning approaches, and DTE can reduce more than half of the computational cost when the number of source subjects is large, with little sacrifice of classification accuracy.

show abstract

Riemannian Approaches in Brain-Computer Interfaces: A Review

Cited by 297 publications

References 63 publications

Predictive regression modeling with MEG/EEG: from source power to signals and cognitive states

Predictive regression modeling with MEG/EEG: from source power to signals and cognitive states

Robotic and Virtual Reality BCIs Using Spatial Tactile and Auditory Oddball Paradigms

Manifold Embedded Knowledge Transfer for Brain-Computer Interfaces

Contact Info

Product

Resources

About