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

Search research articles

Related Experiment Video

Updated: Dec 24, 2025

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

Published on: March 10, 2011

Manifold Embedded Knowledge Transfer for Brain-Computer Interfaces.

Wen Zhang, Dongrui Wu

IEEE Transactions on Neural Systems and Rehabilitation Engineering : a Publication of the IEEE Engineering in Medicine and Biology Society

|April 15, 2020

Summary

This summary is machine-generated.

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

CKD: Contrastive Knowledge Distillation for Cross-Dataset EEG Classification.

IEEE transactions on bio-medical engineering·2026

Same author

fastSeizureNet: Accurate and efficient knowledge-data fusion for semi-supervised seizure detection.

Neural networks : the official journal of the International Neural Network Society·2026

Same author

SACM: SEEG-Audio Contrastive Matching for Chinese Speech Decoding.

IEEE transactions on bio-medical engineering·2026

Same author

Mirror Descent Safe Policy Optimization for Reinforcement Learning Agents.

IEEE transactions on pattern analysis and machine intelligence·2026

Same author

Electroencephalography Enables Continuous Decoding of Hand Motion Angles in Polar Coordinates.

Cyborg and bionic systems (Washington, D.C.)·2026

Same author

Discovery of the reward function for embodied reinforcement learning agents.

Nature communications·2025

Same journal

Pathology-Informed Augmentation Improves Cross-Cohort IMU-to-vGRF Estimation Between Healthy Adults and Adults With Osteoarthritis.

IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society·2026

Same journal

Effects of task-driven head orientations on gait and balance during walking in virtual reality.

IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society·2026

Same journal

Wearable sensor-based Mild Cognitive Impairment Identification: A Multi-Domain Gait Analysis Approach with Association Rule Mining.

IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society·2026

Same journal

Semi-implantable Micro-cooler for Dorsal Root Ganglion Enables Targeted, Sustained, and Cumulative Pain Relief.

IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society·2026

Same journal

Auditory Cue Integration for a Power-Assisted Gait Training System Based on Neurodevelopmental Treatment Principles.

IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society·2026

Same journal

Quantifying the dynamics that link leg tendon vibration to induced periodic postural oscillations in young subjects Differential effects of light touch on the induced sway.

IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society·2026

See all related articles

This study introduces Manifold Embedded Knowledge Transfer (MEKT) for brain-computer interfaces (BCIs). MEKT improves electroencephalogram (EEG) classification across subjects by aligning data on a Riemannian manifold, enhancing BCI performance.

Area of Science:

Neuroscience
Machine Learning
Signal Processing

Background:

Transfer learning is crucial for brain-computer interfaces (BCIs) to address subject and task variability.
Offline unsupervised cross-subject electroencephalogram (EEG) classification presents challenges due to domain shifts between subjects.

Purpose of the Study:

To propose a novel transfer learning approach for robust EEG classification in BCIs.
To enhance the efficiency and accuracy of cross-subject EEG analysis.

Main Methods:

Developed Manifold Embedded Knowledge Transfer (MEKT) to align EEG data covariance matrices on a Riemannian manifold.
Extracted features in the tangent space and performed domain adaptation by minimizing distribution shift while preserving geometric structures.
Introduced Domain Transferability Estimation (DTE) to select optimal source domains for efficient transfer learning.

More Related Videos

Assessment and Communication for People with Disorders of Consciousness

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

Published on: May 8, 2021

Related Experiment Videos

Last Updated: Dec 24, 2025

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

Published on: March 10, 2011

Assessment and Communication for People with Disorders of Consciousness

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

Published on: May 8, 2021

Main Results:

MEKT demonstrated superior performance compared to state-of-the-art transfer learning methods across four EEG datasets and two BCI paradigms.
DTE significantly reduced computational cost (over 50%) for large numbers of source subjects with minimal impact on classification accuracy.

Conclusions:

MEKT offers an effective and efficient solution for unsupervised cross-subject EEG classification in BCIs.
The proposed DTE method optimizes the selection of source domains, improving computational efficiency in transfer learning scenarios.