Emotion Recognition Based on Microstates: A Comparison between Scalp and Source Analysis DOI

Jie Ruan,

Di Xiao

Published: Sept. 22, 2023

The microstate of the electroencephalogram (EEG) captures spatiotemporal information from all channels, encompassing extensive electrophysiological data. Its significance to emotion recognition is substantial. However, current research into based on microstates remains confined scalp level, and due effects volume conduction, accuracy might not be optimal. In this study, we employed sLORETA method map data onto cortex. Subsequently, applied analysis using techniques extracted various features, including duration, occurrence frequency, coverage, transition probability. We performed classification discrete emotional labels separately for source within SEED SEED-IV datasets. For dataset, use K-Nearest Neighbor (KNN) Support Vector Machine (SVM) classifiers resulted in an average increase 6.07% 5.93%, respectively, compared scalp. Similarly, corresponding increments 6.85% 7.5% were observed. These findings emphasize efficacy enhancing accuracy.

Language: Английский

Neuroimaging features for cognitive fatigue and its recovery with VR intervention: An EEG microstates analysis DOI Creative Commons

Jia-Cheng Han,

Chi Zhang, Yan Cai

et al.

Brain Research Bulletin, Journal Year: 2025, Volume and Issue: 221, P. 111223 - 111223

Published: Jan. 24, 2025

Language: Английский

Citations

0
NeuroIDBench: An open-source benchmark framework for the standardization of methodology in brainwave-based authentication research DOI Creative Commons

Avinash Kumar Chaurasia,

Matin Fallahi, Thorsten Strufe

et al.

Journal of Information Security and Applications, Journal Year: 2024, Volume and Issue: 85, P. 103832 - 103832

Published: July 18, 2024

Biometric systems based on brain activity have been proposed as an alternative to passwords or complement current authentication techniques. By leveraging the unique brainwave patterns of individuals, these offer possibility creating solutions that are resistant theft, hands-free, accessible, and potentially even revocable. However, despite growing stream research in this area, faster advance is hindered by reproducibility problems. Issues such lack standard reporting schemes for performance results system configuration, absence common evaluation benchmarks, make comparability proper assessment different biometric challenging. Further, barriers erected future work when, so often, source code not published open access. To bridge gap, we introduce NeuroIDBench, a flexible tool benchmark brainwave-based models. It incorporates nine diverse datasets, implements comprehensive set pre-processing parameters machine learning algorithms, enables testing under two adversary models (known vs unknown attacker), allows researchers generate full reports visualizations. We use NeuroIDBench investigate shallow classifiers deep learning-based approaches literature, test robustness across multiple sessions. observe 37.6% reduction Equal Error Rate (EER) attacker scenarios (typically tested literature), highlight importance session variability authentication. All all, our demonstrate viability relevance streamlining fair comparisons thereby furthering advancement through robust methodological practices.

Language: Английский

Citations

2
Effects of Anti-Seizure Medications on Resting-State Functional Networks in Juvenile Myoclonic Epilepsy: An EEG Microstate Analysis DOI
Ying Li, Yibo Zhao, Yanan Chen

et al.

Seizure, Journal Year: 2024, Volume and Issue: 124, P. 48 - 56

Published: Dec. 5, 2024

Language: Английский

Citations

1
Prediction of Attention Deficit Hyperactivity Disorder Using Machine Learning Models DOI
Sri Parameswaran,

S.R Gowsheeba,

E Praveen

et al.

Published: May 3, 2024

Language: Английский

Citations

0
EEG microstates as an important marker of depression: A systematic review and meta-analysis DOI
Si Zhang, Aiping Chi,

Li-quan Gao

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: July 12, 2024

Abstract This study conducts a literature search through databases such as PubMed, Web of Science, CNKI (China National Knowledge Infrastructure), and the Cochrane Library to collect case-control studies on microstates in patients with depression. Conducting bias risk assessment using Review Manager 5.4, meta-analysis is performed Stata 18.0 14.0 software. has been registered Prospero, CRD42024543793. Our research results suggest that increased duration frequency microstate A may serve potential biomarker for An increase parameter B also observed when individuals experience anxiety. The coverage C are closely related rumination levels. Abnormalities D among some depression indicate presence comorbid conditions overlapping mental disorders or attention executive function deficits. provides important insights into identifying symptoms etiology by examining differences between healthy individuals.

Language: Английский

Citations

0
Emotion Recognition Based on Microstates: A Comparison between Scalp and Source Analysis DOI

Jie Ruan,

Di Xiao

Published: Sept. 22, 2023

The microstate of the electroencephalogram (EEG) captures spatiotemporal information from all channels, encompassing extensive electrophysiological data. Its significance to emotion recognition is substantial. However, current research into based on microstates remains confined scalp level, and due effects volume conduction, accuracy might not be optimal. In this study, we employed sLORETA method map data onto cortex. Subsequently, applied analysis using techniques extracted various features, including duration, occurrence frequency, coverage, transition probability. We performed classification discrete emotional labels separately for source within SEED SEED-IV datasets. For dataset, use K-Nearest Neighbor (KNN) Support Vector Machine (SVM) classifiers resulted in an average increase 6.07% 5.93%, respectively, compared scalp. Similarly, corresponding increments 6.85% 7.5% were observed. These findings emphasize efficacy enhancing accuracy.

Language: Английский

Citations

0