Investigation of different approaches for noise reduction in functional near-infrared spectroscopy signals for brain–computer interface applications

Hammerstein-Wiener Motion Artifact Correction for Functional Near-Infrared Spectroscopy: A Novel Inertial Measurement Unit-Based Technique

Participant movement is a major source of artifacts in functional near-infrared spectroscopy (fNIRS) experiments. Mitigating the impact of motion artifacts (MAs) is crucial to estimate brain activity robustly. Here, we suggest and evaluate a novel application of the nonlinear Hammerstein-Wiener model to estimate and mitigate MAs in fNIRS signals from direct-movement recordings through IMU sensors mounted on the participant's head (head-IMU) and the fNIRS probe (probe-IMU). To this end, we analyzed the hemodynamic responses of single-channel oxyhemoglobin (HbO) and deoxyhemoglobin (HbR) signals from 17 participants who performed a hand tapping task with different levels of concurrent head movement. Additionally, the tapping task was performed without head movements to estimate the ground-truth brain activation. We compared the performance of our novel approach with the probe-IMU and head-IMU to eight established methods (PCA, tPCA, spline, spline Savitzky-Golay, wavelet, CBSI, RLOESS, and WCBSI) on four quality metrics: SNR, △AUC, RMSE, and R. Our proposed nonlinear Hammerstein-Wiener method achieved the best SNR increase (p < 0.001) among all methods. Visual inspection revealed that our approach mitigated MA contaminations that other techniques could not remove effectively. MA correction quality was comparable with head- and probe-IMUs.

End-to-end methane gas detection algorithm based on transformer and multi-layer perceptron

In this paper, an end-to-end methane gas detection algorithm based on transformer and multi-layer perceptron (MLP) for tunable diode laser absorption spectroscopy (TDLAS) is presented. It consists of a Transformer-based U-shaped Neural Network (TUNN) filtering algorithm and a concentration prediction network (CPN) based on MLP. This algorithm employs an end-to-end architectural design to extract information from noisy transmission spectra of methane and derive the CH4 concentrations from denoised spectra, without intermediate steps. The results demonstrate the superiority of the proposed TUNN filtering algorithm over other typically employed digital filters. For concentration prediction, the determination coefficient (R2) reached 99.7%. Even at low concentrations, R2 remained notably high, reaching up to 89%. The proposed algorithm results in a more efficient, convenient, and accurate spectral data processing for TDLAS-based gas sensors.

Learning based motion artifacts processing in fNIRS: a mini review

This paper provides a concise review of learning-based motion artifacts (MA) processing methods in functional near-infrared spectroscopy (fNIRS), highlighting the challenges of maintaining optimal contact during subject movement, which can lead to MA and compromise data integrity. Traditional strategies often result in reduced reliability of the hemodynamic response and statistical power. Recognizing the limited number of studies focusing on learning-based MA removal, we examine 315 studies, identifying seven pertinent to our focus area. We discuss the current landscape of learning-based MA correction methods and highlight research gaps. Noting the absence of standard evaluation metrics for quality assessment of MA correction, we suggest a novel framework, integrating signal and model quality considerations and employing metrics like ΔSignal-to-Noise Ratio (ΔSNR), confusion matrix, and Mean Squared Error. This work aims to facilitate the application of learning-based methodologies to fNIRS and improve the accuracy and reliability of neurovascular studies.

Motion artifacts removal and evaluation techniques for functional near-infrared spectroscopy signals: A review

With the emergence of an increasing number of functional near-infrared spectroscopy (fNIRS) devices, the significant deterioration in measurement caused by motion artifacts has become an essential research topic for fNIRS applications. However, a high requirement for mathematics and programming limits the number of related researches. Therefore, here we provide the first comprehensive review for motion artifact removal in fNIRS aiming to (i) summarize the latest achievements, (ii) present the significant solutions and evaluation metrics from the perspective of application and reproduction, and (iii) predict future topics in the field. The present review synthesizes information from fifty-one journal articles (screened according to three criteria). Three hardware-based solutions and nine algorithmic solutions are summarized, and their application requirements (compatible signal types, the availability for online applications, and limitations) and extensions are discussed. Five metrics for noise suppression and two metrics for signal distortion were synthesized to evaluate the motion artifact removal methods. Moreover, we highlight three deficiencies in the existing research: (i) The balance between the use of auxiliary hardware and that of an algorithmic solution is not clarified; (ii) few studies mention the filtering delay of the solutions, and (iii) the robustness and stability of the solution under extreme application conditions are not discussed.

Auto-Denoising for EEG Signals Using Generative Adversarial Network

The brain-computer interface (BCI) has many applications in various fields. In EEG-based research, an essential step is signal denoising. In this paper, a generative adversarial network (GAN)-based denoising method is proposed to denoise the multichannel EEG signal automatically. A new loss function is defined to ensure that the filtered signal can retain as much effective original information and energy as possible. This model can imitate and integrate artificial denoising methods, which reduces processing time; hence it can be used for a large amount of data processing. Compared to other neural network denoising models, the proposed model has one more discriminator, which always judges whether the noise is filtered out. The generator is constantly changing the denoising way. To ensure the GAN model generates EEG signals stably, a new normalization method called sample entropy threshold and energy threshold-based (SETET) normalization is proposed to check the abnormal signals and limit the range of EEG signals. After the denoising system is established, although the denoising model uses the different subjects' data for training, it can still apply to the new subjects' data denoising. The experiments discussed in this paper employ the HaLT public dataset. Correlation and root mean square error (RMSE) are used as evaluation criteria. Results reveal that the proposed automatic GAN denoising network achieves the same performance as the manual hybrid artificial denoising method. Moreover, the GAN network makes the denoising process automatic, representing a significant reduction in time.

Machine Learning-Based Heavy Metal Ion Detection Using Surface-Enhanced Raman Spectroscopy

Surface-Enhanced Raman Spectroscopy (SERS) is often used for heavy metal ion detection. However, large variations in signal strength, spectral profile, and nonlinearity of measurements often cause problems that produce varying results. It raises concerns about the reproducibility of the results. Consequently, the manual classification of the SERS spectrum requires carefully controlled experimentation that further hinders the large-scale adaptation. Recent advances in machine learning offer decent opportunities to address these issues. However, well-documented procedures for model development and evaluation, as well as benchmark datasets, are missing. Towards this end, we provide the SERS spectral benchmark dataset of lead(II) nitride (Pb(NO3)2) for a heavy metal ion detection task and evaluate the classification performance of several machine learning models. We also perform a comparative study to find the best combination between the preprocessing methods and the machine learning models. The proposed model can successfully identify the Pb(NO3)2 molecule from SERS measurements of independent test experiments. In particular, the proposed model shows an 84.6% balanced accuracy for the cross-batch testing task.

Proposing a convolutional neural network for stress assessment by means of derived heart rate from functional near infrared spectroscopy

BACKGROUND

Stress is known as one of the major factors threatening human health. A large number of studies have been performed in order to either assess or relieve stress by analyzing the brain and heart-related signals.

METHOD

In this study, a method based on the Convolutional Neural Network (CNN) approach is proposed to assess stress induced by the Montreal Imaging Stress Task. The proposed model is trained on the heart rate signal derived from functional Near-Infrared Spectroscopy (fNIRS), which is referred to as HRF. In this regard, fNIRS signals of 20 healthy volunteers were recorded using a configuration of 23 channels located on the prefrontal cortex. The proposed deep learning system consists of two main parts where in the first part, the one-dimensional convolutional neural network is employed to build informative activation maps, and then in the second part, a stack of deep fully connected layers is used to predict the stress existence probability. Thereafter, the employed CNN method is compared with the Dense Neural Network, Support Vector Machine, and Random Forest regarding various classification metrics.

RESULTS

Results clearly showed the superiority of CNN over all other methods. Additionally, the trained HRF model significantly outperforms the model trained on the filtered fNIRS signals, where the HRF model could achieve 98.69 ± 0.45% accuracy, which is 10.09% greater than the accuracy obtained by the fNIRS model.

CONCLUSIONS

Employment of the proposed deep learning system trained on the HRF measurements leads to higher stress classification accuracy than the accuracy reported in the existing studies where the same experimental procedure has been done. Besides, the proposed method suggests better stability with lower variation in prediction. Furthermore, its low computational cost opens up the possibility to be applied in real-time monitoring of stress assessment.

Dual-optimized adaptive Kalman filtering algorithm based on BP neural network and variance compensation for laser absorption spectroscopy

A dual-optimized adaptive Kalman filtering (DO-AKF) algorithm based on back propagation (BP) neural network and variance compensation was developed for high-sensitivity trace gas detection in laser spectroscopy. The BP neural network was used to optimize the Kalman filter (KF) parameters. Variance compensation was introduced to track the state of the system and to eliminate the variations in the parameters of dynamic systems. The proposed DO-AKF algorithm showed the best performance compared with the traditional multi-signal average, extended KF, unscented KF, KF optimized by BP neural network (BP-KF) and KF optimized by variance compensation (VC-KF). The optimized DO-AKF algorithm was applied to a QCL-based gas sensor system for an exhaled CO analysis. The experimental results revealed a sensitivity enhancement factor of 23. The proposed algorithm can be widely used in the fields of environmental pollutant monitoring, industrial process control, and breath gas diagnosis.

Adaptive filtering of physiological noises in fNIRS data

The study presents a recursive least-squares estimation method with an exponential forgetting factor for noise removal in functional near-infrared spectroscopy data and extraction of hemodynamic responses (HRs) from the measured data. The HR is modeled as a linear regression form in which the expected HR, the first and second derivatives of the expected HR, a short-separation measurement data, three physiological noises, and the baseline drift are included as components in the regression vector. The proposed method is applied to left-motor-cortex experiments on the right thumb and little finger movements in five healthy male participants. The algorithm is evaluated with respect to its performance improvement in terms of contrast-to-noise ratio in comparison with Kalman filter, low-pass filtering, and independent component method. The experimental results show that the proposed model achieves reductions of 77% and 99% in terms of the number of channels exhibiting higher contrast-to-noise ratios in oxy-hemoglobin and deoxy-hemoglobin, respectively. The approach is robust in obtaining consistent HR data. The proposed method is applied for both offline and online noise removal.

Early Detection of Hemodynamic Responses Using EEG: A Hybrid EEG-fNIRS Study

Enhanced classification accuracy and a sufficient number of commands are highly demanding in brain computer interfaces (BCIs). For a successful BCI, early detection of brain commands in time is essential. In this paper, we propose a novel classifier using a modified vector phase diagram and the power of electroencephalography (EEG) signal for early prediction of hemodynamic responses. EEG and functional near-infrared spectroscopy (fNIRS) signals for a motor task (thumb tapping) were obtained concurrently. Upon the resting state threshold circle in the vector phase diagram that uses the maximum values of oxy- and deoxy-hemoglobin (ΔHbO and ΔHbR) during the resting state, we introduce a secondary (inner) threshold circle using the ΔHbO and ΔHbR magnitudes during the time window of 1 s where an EEG activity is noticeable. If the trajectory of ΔHbO and ΔHbR touches the resting state threshold circle after passing through the inner circle, this indicates that ΔHbO was increasing and ΔHbR was decreasing (i.e., the start of a hemodynamic response). It takes about 0.5 s for an fNIRS signal to cross the resting state threshold circle after crossing the EEG-based circle. Thus, an fNIRS-based BCI command can be generated in 1.5 s. We achieved an improved accuracy of 86.0% using the proposed method in comparison with the 63.8% accuracy obtained using linear discriminant analysis in a window of 0~1.5 s. Moreover, the active brain locations (identified using the proposed scheme) were spatially specific when a t-map was made after 10 s of stimulation. These results demonstrate the possibility of enhancing the classification accuracy for a brain-computer interface with a time window of 1.5 s using the proposed method.

Applications of Functional Near-Infrared Spectroscopy (fNIRS) Neuroimaging in Exercise⁻Cognition Science: A Systematic, Methodology-Focused Review

For cognitive processes to function well, it is essential that the brain is optimally supplied with oxygen and blood. In recent years, evidence has emerged suggesting that cerebral oxygenation and hemodynamics can be modified with physical activity. To better understand the relationship between cerebral oxygenation/hemodynamics, physical activity, and cognition, the application of state-of-the art neuroimaging tools is essential. Functional near-infrared spectroscopy (fNIRS) is such a neuroimaging tool especially suitable to investigate the effects of physical activity/exercises on cerebral oxygenation and hemodynamics due to its capability to quantify changes in the concentration of oxygenated hemoglobin (oxyHb) and deoxygenated hemoglobin (deoxyHb) non-invasively in the human brain. However, currently there is no clear standardized procedure regarding the application, data processing, and data analysis of fNIRS, and there is a large heterogeneity regarding how fNIRS is applied in the field of exercise⁻cognition science. Therefore, this review aims to summarize the current methodological knowledge about fNIRS application in studies measuring the cortical hemodynamic responses during cognitive testing (i) prior and after different physical activities interventions, and (ii) in cross-sectional studies accounting for the physical fitness level of their participants. Based on the review of the methodology of 35 as relevant considered publications, we outline recommendations for future fNIRS studies in the field of exercise⁻cognition science.

Critical bounds on noise and SNR for robust estimation of real-time brain activity from functional near infra-red spectroscopy

The robust characterization of real-time brain activity carries potential for many applications. However, the contamination of measured signals by various instrumental, environmental, and physiological sources of noise introduces a substantial amount of signal variance and, consequently, challenges real-time estimation of contributions from underlying neuronal sources. Functional near infra-red spectroscopy (fNIRS) is an emerging imaging modality whose real-time potential is yet to be fully explored. The objectives of the current study are to (i) validate a time-dependent linear model of hemodynamic responses in fNIRS, and (ii) test the robustness of this approach against measurement noise (instrumental and physiological) and mis-specification of the hemodynamic response basis functions (amplitude, latency, and duration). We propose a linear hemodynamic model with time-varying parameters, which are estimated (adapted and tracked) using a dynamic recursive least square algorithm. Owing to the linear nature of the activation model, the problem of achieving robust convergence to an accurate estimation of the model parameters is recast as a problem of parameter error stability around the origin. We show that robust convergence of the proposed method is guaranteed in the presence of an acceptable degree of model misspecification and we derive an upper bound on noise under which reliable parameters can still be inferred. We also derived a lower bound on signal-to-noise-ratio over which the reliable parameters can still be inferred from a channel/voxel. Whilst here applied to fNIRS, the proposed methodology is applicable to other hemodynamic-based imaging technologies such as functional magnetic resonance imaging.