Nonlinear Modeling of Cortical Responses to Mechanical Wrist Perturbations Using the NARMAX Method

EEG Signal Processing Techniques and Applications-2nd Edition

Electroencephalography (EEG), as a well-established, non-invasive tool, has been successfully applied to a wide range of conditions due to its many evident advantages, such as economy, portability, easy operation, easy accessibility, and widespread availability in hospitals [...].

Frequency-specific dual-attention based adversarial network for blood oxygen level-dependent time series prediction

Functional magnetic resonance imaging (fMRI) is currently one of the most popular technologies for measuring brain activity in both research and clinical contexts. However, clinical constraints often result in short fMRI scan durations, limiting the diagnostic performance for brain disorders. To address this limitation, we developed an end-to-end frequency-specific dual-attention-based adversarial network (FDAA-Net) to extend the time series of existing blood oxygen level-dependent (BOLD) data, enhancing their diagnostic utility. Our approach leverages the frequency-dependent nature of fMRI signals using variational mode decomposition (VMD), which adaptively tracks brain activity across different frequency bands. We integrated the generative adversarial network (GAN) with a spatial-temporal attention mechanism to fully capture relationships among spatially distributed brain regions and temporally continuous time windows. We also introduced a novel loss function to estimate the upward and downward trends of each frequency component. We validated FDAA-Net on the Human Connectome Project (HCP) database by comparing the original and predicted time series of brain regions in the default mode network (DMN), a key network activated during rest. FDAA-Net effectively overcame linear frequency-specific challenges and outperformed other popular prediction models. Test-retest reliability experiments demonstrated high consistency between the functional connectivity of predicted outcomes and targets. Furthermore, we examined the clinical applicability of FDAA-Net using short-term fMRI data from individuals with autism spectrum disorder (ASD) and major depressive disorder (MDD). The model achieved a maximum predicted sequence length of 40% of the original scan durations. The prolonged time series improved diagnostic performance by 8.0% for ASD and 11.3% for MDD compared with the original sequences. These findings highlight the potential of fMRI time series prediction to enhance diagnostic power of brain disorders in short fMRI scans.

Modeling the cortical response elicited by wrist manipulation via a nonlinear delay differential embedding

Regarding motor processes, modeling healthy people's brains is essential to understand the brain activity in people with motor impairments. However, little research has been undertaken when external forces disturb limbs, having limited information on physiological pathways. Therefore, in this paper, a nonlinear delay differential embedding model is used to estimate the brain response elicited by externally controlled wrist movement in healthy individuals. The aim is to improve the understanding of the relationship between a controlled wrist movement and the generated cortical activity of healthy people, helping to disclose the underlying mechanisms and physiological relationships involved in the motor event. To evaluate the model, a public database from the Delft University of Technology is used, which contains electroencephalographic recordings of ten healthy subjects while wrist movement was externally provoked by a robotic system. In this work, the cortical response related to movement is identified via Independent Component Analysis and estimated based on a nonlinear delay differential embedding model. After a cross-validation analysis, the model performance reaches 90.21% ± 4.46% Variance Accounted For, and Correlation 95.14% ± 2.31%. The proposed methodology allows to select the model degree, to estimate a general predominant operation mode of the cortical response elicited by wrist movement. The obtained results revealed two facts that had not previously been reported: the movement's acceleration affects the cortical response, and a common delayed activity is shared among subjects. Going forward, identifying biomarkers related to motor tasks could aid in the evaluation of rehabilitation treatments for patients with upper limbs motor impairments.

Decoding Electroencephalography Signal Response by Stacking Ensemble Learning and Adaptive Differential Evolution

Electroencephalography (EEG) is an exam widely adopted to monitor cerebral activities regarding external stimuli, and its signals compose a nonlinear dynamical system. There are many difficulties associated with EEG analysis. For example, noise can originate from different disorders, such as muscle or physiological activity. There are also artifacts that are related to undesirable signals during EEG recordings, and finally, nonlinearities can occur due to brain activity and its relationship with different brain regions. All these characteristics make data modeling a difficult task. Therefore, using a combined approach can be the best solution to obtain an efficient model for identifying neural data and developing reliable predictions. This paper proposes a new hybrid framework combining stacked generalization (STACK) ensemble learning and a differential-evolution-based algorithm called Adaptive Differential Evolution with an Optional External Archive (JADE) to perform nonlinear system identification. In the proposed framework, five base learners, namely, eXtreme Gradient Boosting, a Gaussian Process, Least Absolute Shrinkage and Selection Operator, a Multilayer Perceptron Neural Network, and Support Vector Regression with a radial basis function kernel, are trained. The predictions from all these base learners compose STACK's layer-0 and are adopted as inputs of the Cubist model, whose hyperparameters were obtained by JADE. The model was evaluated for decoding the electroencephalography signal response to wrist joint perturbations. The variance accounted for (VAF), root-mean-squared error (RMSE), and Friedman statistical test were used to validate the performance of the proposed model and compare its results with other methods in the literature, including the base learners. The JADE-STACK model outperforms the other models in terms of accuracy, being able to explain around, as an average of all participants, 94.50% and 67.50% (standard deviations of 1.53 and 7.44, respectively) of the data variability for one step ahead and three steps ahead, which makes it a suitable approach to dealing with nonlinear system identification. Also, the improvement over state-of-the-art methods ranges from 0.6% to 161% and 43.34% for one step ahead and three steps ahead, respectively. Therefore, the developed model can be viewed as an alternative and additional approach to well-established techniques for nonlinear system identification once it can achieve satisfactory results regarding the data variability explanation.

TV-NARX and Coiflets WPT based time-frequency Granger causality with application to corticomuscular coupling in hand-grasping

The study of the synchronous characteristics and functional connections between the functional cortex and muscles of hand-grasping movements is important in basic research, clinical disease diagnosis and rehabilitation evaluation. The electroencephalogram (EEG) and electromyographic signal (EMG) signals of 15 healthy participants were used to analyze the corticomuscular coupling under grasping movements by holding three different objects, namely, card, ball, and cup by using the time-frequency Granger causality method based on time-varying nonlinear autoregressive with exogenous input (TV-NARX) model and Coiflets wavelet packet transform. The results show that there is a bidirectional coupling between cortex and muscles under grasping movement, and it is mainly reflected in the beta and gamma frequency bands, in which there is a statistically significant difference (p < 0.05) among the different grasping actions during the movement execution period in the beta frequency band, and a statistically significant difference (p < 0.1) among the different grasping actions during the movement preparation period in the gamma frequency band. The results show that the proposed method can effectively characterize the EEG-EMG synchronization features and functional connections in different frequency bands during the movement preparation and execution phases in the time-frequency domain, and reveal the neural control mechanism of sensorimotor system to control the hand-grasping function achievement by regulating the intensity of neuronal synchronization oscillations.

A General Use QSAR-ARX Model to Predict the Corrosion Inhibition Efficiency of Drugs in Terms of Quantum Mechanical Descriptors and Experimental Comparison for Lidocaine

A study of 250 commercial drugs to act as corrosion inhibitors on steel has been developed by applying the quantitative structure-activity relationship (QSAR) paradigm. Hard-soft acid-base (HSAB) descriptors were used to establish a mathematical model to predict the corrosion inhibition efficiency (IE%) of several commercial drugs on steel surfaces. These descriptors were calculated through third-order density-functional tight binding (DFTB) methods. The mathematical modeling was carried out through autoregressive with exogenous inputs (ARX) framework and tested by fivefold cross-validation. Another set of drugs was used as an external validation, obtaining SD, RMSE, and MSE, obtaining 6.76%, 3.89%, 7.03%, and 49.47%, respectively. With a predicted value of IE% = 87.51%, lidocaine was selected to perform a final comparison with experimental results. By the first time, this drug obtained a maximum IE%, determined experimentally by electrochemical impedance spectroscopy measurements at 100 ppm concentration, of about 92.5%, which stands within limits of 1 SD from the predicted ARX model value. From the qualitative perspective, several potential trends have emerged from the estimated values. Among them, macrolides, alkaloids from Rauwolfia species, cephalosporin, and rifamycin antibiotics are expected to exhibit high IE% on steel surfaces. Additionally, IE% increases as the energy of HOMO decreases. The highest efficiency is obtained in case of the molecules with the highest ω and ΔN values. The most efficient drugs are found with pK_a ranging from 1.70 to 9.46. The drugs recurrently exhibit aromatic rings, carbonyl, and hydroxyl groups with the highest IE% values.

Brain functional and effective connectivity based on electroencephalography recordings: A review

Functional connectivity and effective connectivity of the human brain, representing statistical dependence and directed information flow between cortical regions, significantly contribute to the study of the intrinsic brain network and its functional mechanism. Many recent studies on electroencephalography (EEG) have been focusing on modeling and estimating brain connectivity due to increasing evidence that it can help better understand various brain neurological conditions. However, there is a lack of a comprehensive updated review on studies of EEG-based brain connectivity, particularly on visualization options and associated machine learning applications, aiming to translate those techniques into useful clinical tools. This article reviews EEG-based functional and effective connectivity studies undertaken over the last few years, in terms of estimation, visualization, and applications associated with machine learning classifiers. Methods are explored and discussed from various dimensions, such as either linear or nonlinear, parametric or nonparametric, time-based, and frequency-based or time-frequency-based. Then it is followed by a novel review of brain connectivity visualization methods, grouped by Heat Map, data statistics, and Head Map, aiming to explore the variation of connectivity across different brain regions. Finally, the current challenges of related research and a roadmap for future related research are presented.

Analyzing efficacy and safety of anti-fungal blue light therapy via kernel-based modeling the reactive oxygen species induced by light

OBJECTIVE

The goal of this study is to investigate the efficacy, safety, and mechanism of ABL for inactivating Candida albicans (C. albicans), and to determine the best wavelength for treating candida infected disease, by experimental measurements and dynamic modeling.

METHODS

The changes in reactive oxygen species (ROS) in C. albicans and human host cells under the irradiation of 385, 405, and 415nm wavelengths light with irradiance of 50mW/cm2 were measured. Moreover, a kernel-based nonlinear dynamic model, i.e., nonlinear autoregressive with exogenous inputs (NARX), was developed and applied to predict the concentration of light-induced ROS, whose kernels were selected by a newly developed algorithm based on particle swarm optimization (PSO).

RESULTS

The ROS concentration was increased respectively about 10-12 times in C. albicans and about 3-6 times in human epithelial cells by the ABL treatment with the same fluence of 90J/cm2. The NARX models were respectively fitted to the data from the experiments on both types of cells. Besides, four different kernel functions, including Gaussian, Laplace, linear and polynomial kernels, were compared in their fitting accuracies. The errors with the Laplace kernel turned out to be only 0.2704 and 0.0593, as respectively fitted to the experimental data of the C. albicans and human host cells.

CONCLUSION

The results demonstrated the effectiveness of the NARX modeling approach, and revealed that the 415nm light was more effective as an anti-fungal treatment with less damage to the host cells than the 405 or 385nm light.

SIGNIFICANCE

The kernel-based NARX model identification algorithm offers opportunities for determining the effective and safe light dosages in treating various fungal infection diseases.

Nonlinear System Identification of Neural Systems from Neurophysiological Signals

The human nervous system is one of the most complicated systems in nature. Complex nonlinear behaviours have been shown from the single neuron level to the system level. For decades, linear connectivity analysis methods, such as correlation, coherence and Granger causality, have been extensively used to assess the neural connectivities and input-output interconnections in neural systems. Recent studies indicate that these linear methods can only capture a certain amount of neural activities and functional relationships, and therefore cannot describe neural behaviours in a precise or complete way. In this review, we highlight recent advances in nonlinear system identification of neural systems, corresponding time and frequency domain analysis, and novel neural connectivity measures based on nonlinear system identification techniques. We argue that nonlinear modelling and analysis are necessary to study neuronal processing and signal transfer in neural systems quantitatively. These approaches can hopefully provide new insights to advance our understanding of neurophysiological mechanisms underlying neural functions. These nonlinear approaches also have the potential to produce sensitive biomarkers to facilitate the development of precision diagnostic tools for evaluating neurological disorders and the effects of targeted intervention.