A review of hybrid EEG-based multimodal human–computer interfaces using deep learning: applications, advances, and challenges DOI
Hyung-Tak Lee, Miseon Shim,

Xianghong Liu

et al.

Biomedical Engineering Letters, Journal Year: 2025, Volume and Issue: unknown

Published: March 22, 2025

Language: Английский

Joint Filter-Band-Combination and Multi-View CNN for Electroencephalogram Decoding DOI Creative Commons
Zhuyao Fan, Xugang Xi, Yunyuan Gao

et al.

IEEE Transactions on Neural Systems and Rehabilitation Engineering, Journal Year: 2023, Volume and Issue: 31, P. 2101 - 2110

Published: Jan. 1, 2023

Motor imagery (MI) electroencephalogram (EEG) signals have an important role in brain-computer interface (BCI) research. However, effectively decoding these remains a problem to be solved. Traditional EEG signal algorithms rely on parameter design extract features, whereas deep learning represented by convolution neural network (CNN) can automatically which is more suitable for BCI applications. when data taken as input raw time series, traditional 1D-CNNs are unable acquire both frequency domain and channel association information. To solve this problem, study proposes novel algorithm inserting two modules into CNN. One the Filter Band Combination (FBC) Module, preserves many features possible while maintaining characteristics of EEG. Another module Multi-View structure that from output FBC module. prevent over fitting, we used cosine annealing with restart strategy update rate. The proposed was validated competition dataset experiment dataset, using accuracy, standard deviation, kappa coefficient. Compared algorithms, our achieved improvement maximum average correct rate 6.6% motion 4-classes recognition mission 11.3% 2-classes classification task.

Language: Английский

Citations

13

Applications of near-infrared spectroscopy in neurocritical care DOI Creative Commons
Rachel Thomas, Samuel S. Shin, Ramani Balu

et al.

Neurophotonics, Journal Year: 2023, Volume and Issue: 10(02)

Published: June 30, 2023

SignificanceAcute brain injuries are commonly encountered in the intensive care unit. Alterations cerebrovascular physiology triggered by initial insult can lead to neurological worsening, further injury, and poor outcomes. Robust methods for assessing continuously at bedside limited.AimIn this review, we aim assess potential of near-infrared spectroscopy (NIRS) as a tool monitor critically ill patients with acute injury well those who high risk developing injury.ApproachWe first review basic principles cerebral blood flow regulation how these altered after injury. We then discuss role NIRS different injuries. pay specific attention (1) identify new clinical (2) non-invasively measure intracranial pressure (ICP) autoregulation, (3) optimal (BP) targets that may improve patient outcomes.ResultsA growing body work supports use injured patients. is routinely used during cardiac surgeries neurologic events, there some evidence treatment algorithms using oximetry result improved In be autoregulation an "optimum" BP where status best preserved. Finally, has been utilized thresholds correlate outcome focal hemorrhages.ConclusionsNIRS emerging function Future will aimed technical refinements diagnostic accuracy, larger scale trials establish definitive impact on

Language: Английский

Citations

11

The control patterns of affective processing and cognitive reappraisal: insights from brain controllability analysis DOI
Feng Fang, Antônio Lúcio Teixeira, Rihui Li

et al.

Cerebral Cortex, Journal Year: 2024, Volume and Issue: 34(2)

Published: Jan. 9, 2024

Abstract Perceiving and modulating emotions is vital for cognitive function often impaired in neuropsychiatric conditions. Current tools evaluating emotional dysregulation suffer from subjectivity lack of precision, especially when it comes to understanding emotion a regulatory or control-based perspective. To address these limitations, this study leverages an advanced methodology known as functional brain controllability analysis. We simultaneously recorded electroencephalography (EEG) magnetic resonance imaging (fMRI) data 17 healthy subjects engaged processing regulation tasks. then employed novel EEG/fMRI integration technique reconstruct cortical activity high spatiotemporal resolution manner. Subsequently, we conducted analysis explore the neural network control patterns underlying different Our findings demonstrated that dorsolateral ventrolateral prefrontal cortex exhibited increased during negative compared neutral emotion. Besides, anterior cingulate was notably more active managing than either controlling regulating Finally, posterior parietal emerged central controller This offers valuable insights into mechanisms support perception regulation.

Language: Английский

Citations

4

Enhancing cross-subject EEG emotion recognition through multi-source manifold metric transfer learning DOI

XinSheng Shi,

Qingshan She, Feng Fang

et al.

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 174, P. 108445 - 108445

Published: April 9, 2024

Language: Английский

Citations

4

EEG emotion recognition based on data-driven signal auto-segmentation and feature fusion DOI
Yunyuan Gao, Zehao Zhu, Feng Fang

et al.

Journal of Affective Disorders, Journal Year: 2024, Volume and Issue: 361, P. 356 - 366

Published: June 15, 2024

Language: Английский

Citations

4

Monitoring nap deprivation-induced fatigue using fNIRS and deep learning DOI
Pei Ma,

Chenyang Pan,

Huijuan Shen

et al.

Cognitive Neurodynamics, Journal Year: 2025, Volume and Issue: 19(1)

Published: Jan. 23, 2025

Language: Английский

Citations

0

From light to insight: Functional near-infrared spectroscopy for unravelling cognitive impairment during task performance DOI Open Access
Na Liu,

Yang Ling-ling,

Xiu‐Qing Yao

et al.

BioScience Trends, Journal Year: 2025, Volume and Issue: 19(1), P. 53 - 71

Published: Jan. 24, 2025

Cognitive impairment refers to the of higher brain functions such as perception, thinking or memory that affects individual's ability perform daily social activities. Studies have found changes in neuronal activity during tasks patients with cognitive are closely related cerebral cortical hemodynamics. Functional near-infrared spectroscopy is an indirect method measure neural based on blood oxygen concentration cortex. Due its strong anti-motion interference, high compatibility, and almost no restriction participants environment, it has shown great potential research field impairment. Recognizing these benefits, this comprehensive review systematically elucidates rationale, historical development, advantages disadvantages functional spectroscopy, also discusses applications combining other detection techniques. Additionally, summarized how can be applied caused by different diseases, ultimately aiding study mechanisms activities, which crucial for diagnosis, differentiation treatment

Language: Английский

Citations

0

Outbalanced: The cross-cortical effects of prefrontal neuromodulation in posterior parietal cortex DOI Creative Commons
Maryam Farshad, Beatrix Barth, Jennifer Svaldi

et al.

Cortex, Journal Year: 2025, Volume and Issue: 185, P. 96 - 112

Published: Feb. 5, 2025

Cognitive phenomena such as the Spatial-Numerical Association of Response Codes (SNARC) effect can arise in fronto-parietal cortical network. Prior neuromodulation studies with cathodal transcranial direct current stimulation (tDCS) over left prefrontal cortex (PFC) reduced SNARC effect. neuroimaging functional near-infrared spectroscopy (fNIRS), however, showed signatures posterior parietal (PPC). In this study, we investigated distant neural on hemodynamic activity by combining tDCS fNIRS. The and numerical distance (NDE) were assessed an event-related cross-over design (N = 45), when 1 mA at PFC was applied simultaneously during measurement fNIRS covering bilateral PPC. At behavioral level, did not significantly reduce effect, indicating that replication failed here. Crucially, neuronal activation associated but NDE. This a remote site shown preregistered primary region-of-interest analyses secondary all-channel analyses. results how combination shed light network responsible for cognition, assess effects tDCS.

Language: Английский

Citations

0

Neuroimaging and cognitive correlates of postural control in Parkinson’s disease: a systematic review DOI Creative Commons

P Tait,

Lisa Graham, Rodrigo Vitório

et al.

Journal of NeuroEngineering and Rehabilitation, Journal Year: 2025, Volume and Issue: 22(1)

Published: Feb. 8, 2025

Language: Английский

Citations

0

Clinical value of predicting relapse within 3 months in alcohol-dependent patients using fNIRS in verbal fluency task DOI Creative Commons
Anqi Huang,

Ran Wang,

Aiping Wen

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 12, 2025

To investigate the hemodynamic differences in various brain regions between alcohol dependence (AlcD) patients and healthy controls during a verbal fluency task (VFT) using functional near-infrared spectroscopy (fNIRS), to further explore clinical predictive value of fNIRS before therapy for outcome relapse AlcD after 3 months. A retrospective survey was conducted on 123 149 same period. Baseline assessment performed analyze two groups different regions. During hospitalization, underwent 3-week benzodiazepine substitution therapy, gradually tapering off medication achieve withdrawal treatment goals. Three months discharge, we follow-up phone calls assess status patients. Compared control group, group had significantly lower integral values frontal bilateral temporal lobes, as well β-values all channels lobe except Ch13, lobes (p < 0.005), with no significant difference parietal channel(p > 0.05). ROC (Receiver Operating Characteristic Curve) analysis predicting within showed that area under curve highest (0.951, sensitivity 0.924, specificity 0.886). Patients exhibit impairments lobes. based VFT have good pharmacotherapy can be applied practice.

Language: Английский

Citations

0