Independent Components of Magnetoencephalography: Localization

Tang, Akaysha C.; Pearlmutter, Barak A.; Malaszenko, Natalie A.; Phung, Dan B.; Reeb, Bethany C.

doi:10.1162/089976602760128036

Cited by 95 publications

(79 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The idea of this simpler objective function is similar to the ideas already studied by many authors [10] [11] [12]. We shall not attempt to reproduce precisely the same methods as explained by these earlier authors, but instead we compare our new objective function to L old , which summarises these earlier methods compactly.…”

Section: Model and Posterior For Spatial Distribution 21 Definition mentioning

confidence: 95%

See 1 more Smart Citation

A Bayesian inverse solution using independent component analysis

Puuronen

Hyvärinen

2014

Neural Networks

View full text Add to dashboard Cite

We present new results about the simultaneous linear inverse problems using independent component analysis (ICA), which can be used to separate the data into statistically independent components. The idea of using ICA in solving such inverse problems, especially in EEG/MEG context, has been a known topic for at least more than a decade, but the known results have been justified heuristically, and their relationships are not understood properly. Here we show how to obtain a Bayesian posterior for a spatial source distribution, by using an ICA demixing matrix as an input. The posterior enables us to rederive and reinterpret the previously known methods, and also provides completely new methods.

show abstract

Section: Model and Posterior For Spatial Distribution 21 Definition mentioning

confidence: 95%

“…. + x mmax + ε, and then use some wellknown inverse operator to the components separately [10] [11] [12]. The components are usually obtained by a formula x m = (w + ) * m w m * x, or by some related method.…”

Section: Introductionmentioning

confidence: 99%

A Bayesian inverse solution using independent component analysis

Puuronen

Hyvärinen

2014

Neural Networks

View full text Add to dashboard Cite

show abstract

“…BSS of MEG data segregates noise from signal [Sander et al, 2002;Tang et al, 2000a;Vigário et al, 2000], raising the SNR sufficiently to allow single-trial analysis [Tang et al, 2000b]. Although the sensor attenuation vectors of BSS-separated components can be localized well to equivalent current dipoles [Tang et al, 2002;Vigário et al, 2000], the recovered field maps can be quite noisy. We applied the MLP and MLP-start-LM to localize single dipolar sources from various actual BSS-separated MEG signals.…”

Section: Localization On Real Meg Signals and Comparison With Commercmentioning

confidence: 99%

“…The input to the dipole-fitting algorithm of XFIT was the attenuation vector and the output was the location of equivalent current dipoles. From all separated components for four subjects and four tasks shown in the Appendix, we chose (using the procedures of [Tang et al, 2002]) the 14 BSS-separated MEG signal components for which XFIT had localized a single dipole source well and that matched other criteria 7 for correct localization (see Appendix for further experimental details).…”

Section: Localization On Real Meg Signals and Comparison With Commercmentioning

confidence: 99%

“…To improve the localization accuracy, we used a hybrid MLP-start-Levenberg-Marquardt (LM) method, in which the MLP output provides the starting point for an LM optimization [Levenbert, 1944;Marquardt, 1963]. We used the MLP and MLPstart-LM methods to localize single dipole sources from actual MEG signal components isolated by a BSS algorithm [Tang et al, 2002] and compared the results to the output of standard interactive commercial localization software.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fast robust subject‐independent magnetoencephalographic source localization using an artificial neural network

Jun

Pearlmutter

2004

Human Brain Mapping

Self Cite

View full text Add to dashboard Cite

Abstract:We describe a system that localizes a single dipole to reasonable accuracy from noisy magnetoencephalographic (MEG) measurements in real time. At its core is a multilayer perceptron (MLP) trained to map sensor signals and head position to dipole location. Including head position overcomes the previous need to retrain the MLP for each subject and session. The training dataset was generated by mapping randomly chosen dipoles and head positions through an analytic model and adding noise from real MEG recordings. After training, a localization took 0.7 ms with an average error of 0.90 cm. A few iterations of a Levenberg-Marquardt routine using the MLP output as its initial guess took 15 ms and improved accuracy to 0.53 cm, which approaches the natural limit on accuracy imposed by noise. We applied these methods to localize single dipole sources from MEG components isolated by blind source separation and compared the estimated locations to those generated by standard manually assisted commercial software. Hum Brain Mapp 24:21-34, 2005.

show abstract

Tracking gaze position from EEG: Exploring the possibility of an EEG‐based virtual eye‐tracker

Sun,

Cheng,

Chan

et al. 2023

Brain and Behavior

Self Cite

View full text Add to dashboard Cite

IntroductionOcular artifact has long been viewed as an impediment to the interpretation of electroencephalogram (EEG) signals in basic and applied research. Today, the use of blind source separation (BSS) methods, including independent component analysis (ICA) and second‐order blind identification (SOBI), is considered an essential step in improving the quality of neural signals. Recently, we introduced a method consisting of SOBI and a discriminant and similarity (DANS)‐based identification method, capable of identifying and extracting eye movement–related components. These recovered components can be localized within ocular structures with a high goodness of fit (>95%). This raised the possibility that such EEG‐derived SOBI components may be used to build predictive models for tracking gaze position.MethodsAs proof of this new concept, we designed an EEG‐based virtual eye‐tracker (EEG‐VET) for tracking eye movement from EEG alone. The EEG‐VET is composed of a SOBI algorithm for separating EEG signals into different components, a DANS algorithm for automatically identifying ocular components, and a linear model to transfer ocular components into gaze positions.ResultsThe prototype of EEG‐VET achieved an accuracy of 0.920° and precision of 1.510° of a visual angle in the best participant, whereas an average accuracy of 1.008° ± 0.357° and a precision of 2.348° ± 0.580° of a visual angle across all participants (N = 18).ConclusionThis work offers a novel approach that readily co‐registers eye movement and neural signals from a single‐EEG recording, thus increasing the ease of studying neural mechanisms underlying natural cognition in the context of free eye movement.

show abstract

Independent Components of Magnetoencephalography: Localization

Cited by 95 publications

References 43 publications

A Bayesian inverse solution using independent component analysis

A Bayesian inverse solution using independent component analysis

Fast robust subject‐independent magnetoencephalographic source localization using an artificial neural network

Tracking gaze position from EEG: Exploring the possibility of an EEG‐based virtual eye‐tracker

Contact Info

Product

Resources

About