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 Experiment Video

Updated: Dec 7, 2025

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces
10:51

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

Published on: March 10, 2011

14.1K

Affective brain-computer interfaces: Choosing a meaningful performance measuring metric.

Md Rakibul Mowla1, Rachael I Cano2, Katie J Dhuyvetter1

  • 1Mike Wiegers Department of Electrical & Computer Engineering, Kansas State University, Manhattan, KS, 66506, USA.

Computers in Biology and Medicine
|October 2, 2020
PubMed
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

Human forebrain neural synchronization and entrainment to breathing during wakefulness, sleep, and external mechanical ventilation.

Nature communications·2026
Same author

Efficient coherence inference on complex time-frequency coefficients using a general linear model.

Journal of neuroscience methods·2026
Same author

Cycle-by-cycle respiration waveforms are coupled with the shape of neural oscillations.

bioRxiv : the preprint server for biology·2026
Same author

A motor thalamic site in humans that suppresses involuntary breathing without awareness.

Journal of neurophysiology·2026
Same author

Investigation of blood-brain barrier penetration and pharmacokinetics of a new formulation of cyanide antidote dimethyl trisulfide.

Toxicology and environmental health sciences·2025
Same author

Human forebrain neural synchronization and entrainment to breathing during wakefulness, sleep, and external mechanical ventilation.

Research square·2025

Affective brain-computer interfaces (BCIs) using electroencephalography (EEG) must account for class imbalance. Balanced accuracy is recommended for reliable emotion recognition, with specific EEG features showing promise for classifying valence, arousal, and dominance.

Area of Science:

  • Affective computing and brain-computer interfaces (BCIs).
  • Neuroscience and signal processing for emotion recognition.

Background:

  • Affective BCIs aim to improve human-computer interaction and care for individuals with severe disabilities.
  • Electroencephalography (EEG) is a key technology for assessing affective states.
  • Publicly available datasets like DEAP are crucial for BCI research.

Purpose of the Study:

  • To evaluate the effectiveness of EEG recordings for recognizing affective states.
  • To identify the impact of class imbalance in affective BCI research.
  • To propose appropriate methods for handling class imbalance and comparing results across studies.

Main Methods:

  • Utilized EEG data from an in-house lab collection and the public DEAP database.
  • Reviewed existing literature on DEAP database usage, focusing on class imbalance.
Keywords:
Affective brain-computer interfacesBalanced accuracyElectroencephalogramEmotion classificationPerformance measurementSupport vector machines

More Related Videos

Assessing the Multiple Dimensions of Engagement to Characterize Learning: A Neurophysiological Perspective
13:57

Assessing the Multiple Dimensions of Engagement to Characterize Learning: A Neurophysiological Perspective

Published on: July 1, 2015

13.0K
Assessment and Communication for People with Disorders of Consciousness
07:37

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

9.5K

Related Experiment Videos

Last Updated: Dec 7, 2025

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces
10:51

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

Published on: March 10, 2011

14.1K
Assessing the Multiple Dimensions of Engagement to Characterize Learning: A Neurophysiological Perspective
13:57

Assessing the Multiple Dimensions of Engagement to Characterize Learning: A Neurophysiological Perspective

Published on: July 1, 2015

13.0K
Assessment and Communication for People with Disorders of Consciousness
07:37

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

9.5K
  • Employed features from prior research and introduced the theta/beta-1 ratio for classification.
  • Recommended balanced accuracy and its posterior distribution for performance evaluation.

    • Main Results:

    • A significant number of studies using the DEAP database neglect class imbalance, leading to misleading results.
    • Class imbalance complicates cross-study comparisons and affects the determination of chance levels.
    • Beta band power, theta band power, and the theta/beta-1 ratio emerged as effective features for classifying valence, arousal, and dominance, respectively.

    Conclusions:

    • Addressing class imbalance is critical for accurate and comparable results in affective BCI research.
    • Balanced accuracy provides a more reliable performance metric than standard accuracy in imbalanced datasets.
    • Specific EEG frequency bands and ratios offer promising feature sets for robust emotion recognition.