Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

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

Transformative Technologies in Brain-Computer Interfaces-Innovations, Applications, and Challenges.

Brain connectivity·2026
Same author

Increasing EEG electrode density improves decoding of visual categories and source localization: an exploratory ultra-high-density EEG study.

Communications engineering·2026
Same author

Toward Brain-Computer Interface motor rehabilitation for people with Multiple Sclerosis.

Frontiers in medicine·2026
Same author

Reliable and efficient transcranial magnetic stimulation-electroencephalography (TMS-EEG) using ultra-thin active electrodes.

Communications engineering·2025
Same author

Correction: Biomarkers of professional cybersportsmen: Event related potentials and cognitive tests study.

PloS one·2025
Same author

Motor Imagery Acquisition Paradigms: In the Search to Improve Classification Accuracy.

Sensors (Basel, Switzerland)·2025

Related Experiment Video

Updated: Jan 5, 2026

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

1.1K

Time-Variant Linear Discriminant Analysis Improves Hand Gesture and Finger Movement Decoding for Invasive

Johannes Gruenwald1,2, Andrei Znobishchev3, Christoph Kapeller1

  • 1g.tec Medical Engineering GmbH, Schiedlberg, Austria.

Frontiers in Neuroscience
|October 17, 2019
PubMed
Summary

We introduce time-variant linear discriminant analysis (TVLDA), a novel method for brain-computer interfaces. TVLDA significantly improves classification accuracy for motor tasks by accounting for transient neural information, outperforming existing techniques.

Keywords:
brain-computer interfaceelectrocorticographylinear discriminant analysismovement decodingspectral whitening

More Related Videos

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

44.0K
P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation
06:09

P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation

Published on: September 8, 2023

888

Related Experiment Videos

Last Updated: Jan 5, 2026

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

1.1K
Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

44.0K
P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation
06:09

P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation

Published on: September 8, 2023

888

Area of Science:

  • Neuroscience and Biomedical Engineering
  • Brain-Computer Interface (BCI) Technology

Background:

  • Invasive brain-computer interfaces (BCIs) demonstrate high performance but often rely on static analysis methods like Linear Discriminant Analysis (LDA).
  • Standard LDA struggles to capture transient information present in high-gamma bandpower features, limiting BCI accuracy.
  • Existing feature reduction techniques like Common Spatial Patterns (CSPs) add complexity and may not be optimal for dynamic neural signals.

Purpose of the Study:

  • To develop an advanced classification method, time-variant linear discriminant analysis (TVLDA), that effectively processes time-varying high-gamma bandpower features.
  • To introduce a time-domain whitening stage to normalize brain-wave spectral characteristics (1/f shape) for improved feature analysis.
  • To evaluate TVLDA's performance against established methods using real-world BCI data from epilepsy patients performing motor tasks.

Main Methods:

  • Developed TVLDA, an extension of LDA incorporating a feature reduction stage and a novel time-domain whitening pre-processing step.
  • Utilized log-transformed high-gamma bandpower features (50-300 Hz) from intracranial EEG recordings of 15 epilepsy patients.
  • Compared TVLDA against LDA with feature selection, LDA with CSPs, and regularized LDA using motor tasks (3 high-level gestures, individual finger movements).

Main Results:

  • The whitening stage improved classification performance by an average of 11%.
  • TVLDA significantly outperformed LDA variants, achieving improvements of 11.8% (vs. feature selection), 13.9% (vs. CSPs), and 16.4% (vs. regularized LDA).
  • Achieved high classification accuracies: 99% for three-level gestures and 96% for individual finger movements, with minimal features (1-2) and rapid training (<1 second).

Conclusions:

  • TVLDA offers a robust, efficient, and accurate method for decoding motor intentions in real-time BCIs, outperforming current state-of-the-art techniques.
  • The proposed method demonstrates stability even with limited trial data and is suitable for high-density electrode arrays.
  • TVLDA represents a significant advancement in BCI signal processing, achieving record-breaking accuracies for motor-control BCIs.