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

Multiple binding modes of the tyrosine kinase inhibitor dovitinib to fibroblast growth factor receptor 1 characterized by surface plasmon resonance, 19F-NMR, and molecular dynamics simulations.

Journal of biochemistry·2026
Same author

Template-Based Label Propagation for Mouse Brain MRI Skull Stripping.

Neuroinformatics·2026
Same author

Diagnostic Pitfalls of Rapidly Growing Mycobacteria: Cutaneous Mycobacterium chelonae Infection Detected by Extended Routine Culture Despite Negative Mycobacterial Tests.

International journal of dermatology·2026
Same author

Consciousness Disturbances Reported With Clindamycin Versus Cefazolin in Surgical Patients: A Global Pharmacovigilance Analysis Using VigiBase.

Clinical and translational science·2026
Same author

CRTC1 knockdown in the marmoset visual cortex induces neuronal IEG overexpression, HFOs, and neurodegeneration.

Neuroscience research·2026
Same author

Improving purity and apparent thermal stability of single-chain Fv antibodies by heat treatment in a soluble E. coli cytoplasmic expression system.

Journal of biochemistry·2026

Related Experiment Video

Updated: Dec 30, 2025

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

1.9K

EEG-based personal identification method using unsupervised feature extraction and its robustness against

Takashi Nishimoto1,2, Hiroshi Higashi1, Hiroshi Morioka3,4

  • 1Graduate School of Informatics, Kyoto University, Kyoto 606-8501, Japan.

Journal of Neural Engineering
|January 21, 2020
PubMed
Summary

Electroencephalography (EEG) signals show variability, impacting personal identification accuracy. However, novel unsupervised learning methods robustly extract unique EEG features for reliable authentication, even across different sessions.

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

43.9K
Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy
11:15

Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy

Published on: June 27, 2013

34.3K

Related Experiment Videos

Last Updated: Dec 30, 2025

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

1.9K
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

43.9K
Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy
11:15

Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy

Published on: June 27, 2013

34.3K

Area of Science:

  • Neuroscience
  • Biometrics
  • Machine Learning

Background:

  • Brain activity signals, specifically electroencephalography (EEG), are explored as potential biomarkers for personal authentication.
  • Signal variability due to measurement and subject-dependent factors poses a significant challenge to the consistency and performance of EEG-based identification systems.
  • This variability can lead to decreased accuracy in identifying individuals across multiple, distinct EEG recordings.

Purpose of the Study:

  • To evaluate the influence of EEG signal variability on personal identification performance.
  • To develop and assess unsupervised learning methods for extracting personal EEG features invariant across different measurement sessions.
  • To test the robustness of these methods in both personal identification and unknown subject detection scenarios.

Main Methods:

  • Collected EEG signals from twenty subjects across four sessions over two days, re-installing the EEG cap each time.
  • Applied unsupervised learning techniques, including common dictionary learning and t-distributed stochastic neighbor embedding, to extract session-invariant personal EEG features.
  • Compared identification performance between training and testing data from the same round (Setting SR) and different rounds (Setting DR).

Main Results:

  • Personal identification performance was higher in Setting SR (same round) than in Setting DR (different rounds), indicating session-dependent feature dominance.
  • Despite variability, the proposed method achieved a 40% accuracy rate in Setting DR, significantly above chance, demonstrating robust feature extraction.
  • The methods also showed applicability in detecting unknown subjects, with good performance even in Setting DR, though round variability influenced results.

Conclusions:

  • Variability in EEG data significantly affects personal features used for authentication.
  • The proposed unsupervised learning methods are applicable for personal authentication, including both identification and unknown subject detection, even when accounting for EEG data variability.