Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Fatal Neonatal Cardiac Rhabdomyoma Presenting With Pulmonary Hypoplasia: An Autopsy Case Report From Jamaica.

Cureus·2026

Same author

Impact of adding the immune checkpoint inhibitor atezolizumab to first-line chemotherapy on progression-free survival in poor-prognosis ovarian cancer: A retrospective analysis from the IMagyn050 trial.

Gynecologic oncology·2025

Same author

Case report of stage IIIC low-grade serous ovarian cancer in a 13-year-old female treated with novel therapy.

Gynecologic oncology reports·2025

Same author

Impact of chronic ethanol consumption and SARS-COV-2 on the liver and intestine: A pilot dose-response study in mice.

Alcohol, clinical & experimental research·2025

Same author

Development of an affordable light emitting diode spectrophotometer paired with a Python program for calibration and linearity testing and the measurement of uranium(VI).

PloS one·2024

Same author

Testing the structural disconnection hypothesis: Myelin content correlates with memory in healthy aging.

Neurobiology of aging·2024

Search research articles

Related Experiment Video

Updated: Jun 18, 2026

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Published on: November 1, 2019

Nonlinear dimensionality reduction of electroencephalogram (EEG) for Brain Computer interfaces.

Mohammad Nayeem Teli¹, Charles Anderson

¹Department of Computer Science, Colorado State University, Fort Collins, CO 80526, USA. mayeem@cs.colostate.edu

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference

|December 8, 2009

Summary

This study explores using neural networks to reduce the dimensionality of electroencephalogram (EEG) data for brain-computer interfaces (BCI). While bottleneck networks effectively transform EEG signals, they do not enhance classification accuracy over traditional methods.

More Related Videos

Recording Human Electrocorticographic (ECoG) Signals for Neuroscientific Research and Real-time Functional Cortical Mapping

Recording Human Electrocorticographic (ECoG) Signals for Neuroscientific Research and Real-time Functional Cortical Mapping

Published on: June 26, 2012

Assessment and Communication for People with Disorders of Consciousness

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

Related Experiment Videos

Last Updated: Jun 18, 2026

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Published on: November 1, 2019

Recording Human Electrocorticographic (ECoG) Signals for Neuroscientific Research and Real-time Functional Cortical Mapping

Recording Human Electrocorticographic (ECoG) Signals for Neuroscientific Research and Real-time Functional Cortical Mapping

Published on: June 26, 2012

Assessment and Communication for People with Disorders of Consciousness

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

Area of Science:

Neuroscience
Computer Science
Signal Processing

Background:

Electroencephalogram (EEG) signals are crucial for Brain-Computer Interfaces (BCI).
High-dimensional EEG data requires effective transformation into lower-dimensional spaces for accurate classification of mental tasks.
Existing methods for optimizing neural networks can be slow to converge for complex, multi-layered networks.

Purpose of the Study:

To investigate the efficacy of auto-encoder neural networks with a bottleneck configuration for transforming high-dimensional EEG data.
To explore the use of approximate second-order methods for optimizing neural network weights in this context.
To assess whether these non-linear transformations improve the classification of mental tasks from EEG signals.

Main Methods:

Utilized a neural network (NN) with an auto-encoder bottleneck configuration to perform non-linear dimensionality reduction on EEG data.
Implemented two approximate second-order optimization methods to train the neural networks, addressing slow convergence issues of first-order methods.
Analyzed the resulting low-dimensional projections of time-embedded EEG signals for task-related separations.

Main Results:

The bottleneck neural networks successfully discovered non-linear transformations to low-dimensional spaces.
These transformations captured significant information from the original EEG signals, showing task-related separations.
Classification rates using the transformed data did not surpass those achieved with Quadratic Discriminant Analysis (QDA) on the original time-lagged EEG data.

Conclusions:

Auto-encoder neural networks with bottleneck configurations can effectively perform non-linear dimensionality reduction on EEG data.
While these transformations reveal informative patterns, they do not inherently improve classification performance compared to established methods like QDA.
Further research may be needed to optimize the use of these dimensionality reduction techniques for BCI applications.