Deep learning in fNIRS: a review

Neurophotonics, Journal Year: 2022, Volume and Issue: 9(04)

Published: July 20, 2022

Significance: Optical neuroimaging has become a well-established clinical and research tool to monitor cortical activations in the human brain. It is notable that outcomes of functional near-infrared spectroscopy (fNIRS) studies depend heavily on data processing pipeline classification model employed. Recently, deep learning (DL) methodologies have demonstrated fast accurate performances tasks across many biomedical fields. Aim: We aim review emerging DL applications fNIRS studies. Approach: first introduce some commonly used techniques. Then, summarizes current work most active areas this field, including brain-computer interface, neuro-impairment diagnosis, neuroscience discovery. Results: Of 63 papers considered review, 32 report comparative study techniques traditional machine where 26 been shown outperforming latter terms accuracy. In addition, eight also utilize reduce amount preprocessing typically done with or increase via augmentation. Conclusions: The application mitigate hurdles present such as lengthy small sample sizes while achieving comparable improved

Language: Английский

---

State-of-the-Art on Brain-Computer Interface Technology

Sensors, Journal Year: 2023, Volume and Issue: 23(13), P. 6001 - 6001

Published: June 28, 2023

This paper provides a comprehensive overview of the state-of-the-art in brain–computer interfaces (BCI). It begins by providing an introduction to BCIs, describing their main operation principles and most widely used platforms. The then examines various components BCI system, such as hardware, software, signal processing algorithms. Finally, it looks at current trends research related use for medical, educational, other purposes, well potential future applications this technology. concludes highlighting some key challenges that still need be addressed before widespread adoption can occur. By presenting up-to-date assessment technology, will provide valuable insight into where field is heading terms progress innovation.

Language: Английский

---

fNIRS-EEG BCIs for Motor Rehabilitation: A Review

Bioengineering, Journal Year: 2023, Volume and Issue: 10(12), P. 1393 - 1393

Published: Dec. 6, 2023

Motor impairment has a profound impact on significant number of individuals, leading to substantial demand for rehabilitation services. Through brain–computer interfaces (BCIs), people with severe motor disabilities could have improved communication others and control appropriately designed robotic prosthetics, so as (at least partially) restore their abilities. BCI plays pivotal role in promoting smoother interactions between individuals impairments others. Moreover, they enable the direct assistive devices through brain signals. In particular, most potential lies realm rehabilitation, where BCIs can offer real-time feedback assist users training continuously monitor brain’s state throughout entire process. Hybridization different brain-sensing modalities, especially functional near-infrared spectroscopy (fNIRS) electroencephalography (EEG), shown great creation rehabilitating motor-impaired populations. EEG, well-established methodology, be combined fNIRS compensate inherent disadvantages achieve higher temporal spatial resolution. This paper reviews recent works hybrid fNIRS-EEG emphasizing methodologies that utilized imagery. An overview system its key components was introduced, followed by an introduction various devices, strengths weaknesses signal processing techniques, applications neuroscience clinical contexts. The review concludes discussing possible challenges opportunities future development.

Language: Английский

---

Comparing Multi-Dimensional fNIRS Features Using Bayesian Optimization-Based Neural Networks for Mild Cognitive Impairment (MCI) Detection

IEEE Transactions on Neural Systems and Rehabilitation Engineering, Journal Year: 2023, Volume and Issue: 31, P. 1019 - 1029

Published: Jan. 1, 2023

The diagnosis of mild cognitive impairment (MCI), a prodromal stage Alzheimer's disease (AD), is essential for initiating timely treatment to delay the onset AD. Previous studies have shown potential functional near-infrared spectroscopy (fNIRS) diagnosing MCI. However, preprocessing fNIRS measurements requires extensive experience identify poor-quality segments. Moreover, few explored how proper multi-dimensional features influence classification results disease. Thus, this study outlined streamlined method process and compared with neural networks in order explore temporal spatial factors affect MCI normality. More specifically, proposed using Bayesian optimization-based auto hyperparameter tuning evaluate 1D channel-wise, 2D spatial, 3D spatiotemporal detecting patients. highest test accuracies 70.83%, 76.92%, 80.77% were achieved 1D, 2D, features, respectively. Through comparisons, time-point oxyhemoglobin feature was proven be more promising by an dataset 127 participants. Furthermore, presented approach data processing, designed models required no manual tuning, which promoted general utilization modality network-based detect

Language: Английский

---

A Comprehensive Review on Brain–Computer Interface (BCI)-Based Machine and Deep Learning Algorithms for Stroke Rehabilitation

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(14), P. 6347 - 6347

Published: July 21, 2024

This literature review explores the pivotal role of brain–computer interface (BCI) technology, coupled with electroencephalogram (EEG) in advancing rehabilitation for individuals damaged muscles and motor systems. study provides a comprehensive overview recent developments BCI control rehabilitation, emphasizing integration user-friendly technological support robotic prosthetics powered by brain activity. critically examines latest strides technology its application skill recovery. Special attention is given to prevalent EEG devices adaptable BCI-driven rehabilitation. The surveys significant contributions realm machine learning-based deep evaluation. demonstrates promising outcomes enhancing skills identifies key suitable applications, discusses advancements learning approaches assessment, highlights emergence novel Furthermore, it showcases successful case studies illustrating practical implementation techniques their positive impact on diverse patient populations. serves as cornerstone informed decision-making field results highlight BCI’s advantages, integration. findings potential reshaping practices offer insights recommendations future research directions. contributes significantly ongoing transformation particularly through utilization equipment, providing roadmap researchers this dynamic domain.

Language: Английский

---

A Survey of EEG and Machine Learning-Based Methods for Neural Rehabilitation

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 114155 - 114171

Published: Jan. 1, 2023

One approach to therapy and training for the restoration of damaged muscles motor systems is rehabilitation. EEG-assisted Brain-Computer Interface (BCI) may aid in restoring or enhancing brain's lost abilities. Assisted by brain activity, BCI offers simple-to-use technology aids robotic prosthetics. This systematic literature review (SLR) aims explore latest developments control Additionally, typical EEG apparatuses available BCI-driven rehabilitative purposes have been explored. Furthermore, a comparison significant studies rehabilitation assessment using machine learning techniques has summarized. The results this study influence policymakers' decisions regarding use equipment, particularly wireless devices, implement technology. Moreover, SLR offer suggestions further study. To identify additional characteristics each equipment determine which one most suited industry, we plan on measuring user experience based various devices future research.

Language: Английский

---

Application of Transfer Learning for Biomedical Signals: A Comprehensive Review of the Last Decade (2014-2024)

Information Fusion, Journal Year: 2025, Volume and Issue: 118, P. 102982 - 102982

Published: Jan. 30, 2025

Language: Английский

---

Improved performance of fNIRS-BCI by stacking of deep learning-derived frequency domain features

PLoS ONE, Journal Year: 2025, Volume and Issue: 20(4), P. e0314447 - e0314447

Published: April 17, 2025

The functional near-infrared spectroscopy-based brain-computer interface (fNIRS-BCI) systems recognize patterns in brain signals and generate control commands, thereby enabling individuals with motor disabilities to regain autonomy. In this study hand gripping data is acquired using fNIRS neuroimaging system, preprocessing performed nirsLAB features extraction deep learning (DL) Algorithms. For feature classification stack fft methods are proposed. Convolutional neural networks (CNN), long short-term memory (LSTM), bidirectional long-short-term (Bi-LSTM) employed extract features. method classifies these a model the enhances by applying fast Fourier transformation which followed model. proposed applied from twenty participants engaged two-class hand-gripping activity. performance of compared conventional CNN, LSTM, Bi-LSTM algorithms one another. yield 90.11% 87.00% accuracies respectively, significantly higher than those achieved CNN (85.16%), LSTM (79.46%), (81.88%) algorithms. results show that can be effectively used for two three-class problems fNIRS-BCI applications.

Language: Английский

---

Multimodal Model for Automated Pain Assessment: Leveraging Video and fNIRS

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(9), P. 5151 - 5151

Published: May 6, 2025

Pain assessment is a challenging task for clinicians due to its subjective nature, particularly in individuals with communication difficulties, cognitive impairments, or severe disabilities. Traditional methods such as the Visual Analogue Scale (VAS), Numerical Rating (NRS), and Verbal (VRS) rely heavily on patient feedback, which can be inconsistent subjective. To address these limitations, developing objective reliable pain tools that incorporate advanced technologies, multimodal data integration from video fNIRS, important improving clinical outcomes. However, challenges noise susceptibility fNIRS signals must carefully addressed realize their full potential. Recent studies have explored automatic using machine learning deep techniques, require high-quality accurately represent categories. In response introduction of new dataset AI4Pain Challenge, we proposed neural network model utilizing attention-based fusion improve overall accuracy (MMAPA). Our leverages modalities well manually extracted statistical features. We also implemented signal preprocessing artifact filtering, significantly improved performance both feature branches. On hidden test set, our achieved an 51.33%, outperforming official baseline 43.33%. evaluate generalizability, further tested method BioVid Heat Database, where highest 10-fold cross-validation setting, PainAttNet unimodal variants. These results highlight effectiveness approach classification across datasets.

Language: Английский

---

Brain-Computer interface control of stepping from invasive electrocorticography upper-limb motor imagery in a patient with quadriplegia

Frontiers in Human Neuroscience, Journal Year: 2023, Volume and Issue: 16

Published: Jan. 9, 2023

Introduction: Most spinal cord injuries (SCI) result in lower extremities paralysis, thus diminishing ambulation. Using brain-computer interfaces (BCI), patients may regain leg control using neural signals that actuate assistive devices. Here, we present a case of subject with cervical SCI an implanted electrocorticography (ECoG) device and determined whether the system is capable motor-imagery-initiated walking ambulator. Methods: A 24-year-old male (C5 ASIA A) was before study ECoG sensing over sensorimotor hand region brain. The used motor-imagery (MI) to train decoders classify rhythms. Fifteen sessions closed-loop trials followed which ambulated for one hour on robotic-assisted weight-supported treadmill three times per week. We evaluated stability best-performing decoder time initiate by decoding upper-limb (UL) MI. Results: An online bagged trees classifier performed best accuracy 84.15% averaged across 9 weeks. Decoder remained stable following throughout data collection. Discussion: These results demonstrate UL MI feasible signal use lower-limb motor control. Invasive BCI systems designed upper-extremity can be extended controlling beyond upper extremity alone. Importantly, were able invasive several weeks accurately from signal. More work needed determine long-term consequence between resulting

Language: Английский

---