A New Nonlinear Autoregressive Exogenous (NARX)-based Intra-spinal Stimulation Approach to Decode Brain Electrical Activity for Restoration of Leg Movement in Spinally-injured Rabbits.

Introduction: This study aimed at investigating the stimulation by intra-spinal signals decoded from electrocorticography (ECoG) assessments to restore the movements of the leg in an animal model of spinal cord injury (SCI). Methods: The present work is comprised of three steps. First, ECoG signals and the associated leg joint changes (hip, knee, and ankle) in sedated healthy rabbits were recorded in different trials. Second, an appropriate set of intra-spinal electric stimuli was discovered to restore natural leg movements, using the three leg joint movements under a fuzzy-controlled strategy in spinally-injured rabbits under anesthesia. Third, a nonlinear autoregressive exogenous (NARX) neural network model was developed to produce appropriate intra-spinal stimulation developed from decoded ECoG information. The model was able to correlate the ECoG signal data to the intra-spinal stimulation data and finally, induced desired leg movements. In this study, leg movements were also developed from offline ECoG signals (deciphered from rabbits that were not injured) as well as online ECoG data (extracted from the same rabbit after SCI induction). Results: Based on our data, the correlation coefficient was 0.74±0.15 and the normalized root means square error of the brain-spine interface was 0.22±0.10. Conclusion: Overall, we found that using NARX, appropriate information from ECoG recordings can be extracted and used for the generation of proper intra-spinal electric stimulations for restoration of natural leg movements lost due to SCI.

Extracting a Novel Emotional EEG Topographic Map Based on a Stacked Autoencoder Network.

Emotion recognition based on brain signals has increasingly become attractive to evaluate human's internal emotional states. Conventional emotion recognition studies focus on developing machine learning and classifiers. However, most of these methods do not provide information on the involvement of different areas of the brain in emotions. Brain mapping is considered as one of the most distinguishing methods of showing the involvement of different areas of the brain in performing an activity. Most mapping techniques rely on projection and visualization of only one of the electroencephalogram (EEG) subband features onto brain regions. The present study aims to develop a new EEG-based brain mapping, which combines several features to provide more complete and useful information on a single map instead of common maps. In this study, the optimal combination of EEG features for each channel was extracted using a stacked autoencoder (SAE) network and visualizing a topographic map. Based on the research hypothesis, autoencoders can extract optimal features for quantitative EEG (QEEG) brain mapping. The DEAP EEG database was employed to extract topographic maps. The accuracy of image classifiers using the convolutional neural network (CNN) was used as a criterion for evaluating the distinction of the obtained maps from a stacked autoencoder topographic map (SAETM) method for different emotions. The average classification accuracy was obtained 0.8173 and 0.8037 in the valence and arousal dimensions, respectively. The extracted maps were also ranked by a team of experts compared to common maps. The results of quantitative and qualitative evaluation showed that the obtained map by SAETM has more information than conventional maps.

Application of a neural interface for restoration of leg movements: Intra-spinal stimulation using the brain electrical activity in spinally injured rabbits.

This study aimed to design a neural interface that extracts movement commands from the brain to generate appropriate intra-spinal stimulation to restore leg movement. This study comprised four steps: (1) Recording electrocorticographic (ECoG) signals and corresponding leg movements in different trials. (2) Partial laminectomy to induce spinal cord injury (SCI) and detect motor modules in the spinal cord. (3) Delivering appropriate intra-spinal stimulation to the motor modules for restoration of the movements to those documented before SCI. (4) Development of a neural interface created by sparse linear regression (SLiR) model to detect movement commands transmitted from the brain to the modules. Correlation coefficient (CC) and normalized root mean square (NRMS) error was calculated to evaluate the neural interface effectiveness. It was found that by stimulating detected spinal cord modules, joint angle evaluated before SCI was not significantly different from that of post-SCI (P > 0.05). Based on results of SLiR model, overall CC and NRMS values were 0.63 ± 0.14 and 0.34 ± 0.16 (mean ± SD), respectively. These results indicated that ECoG data contained information about intra-spinal stimulations and the developed neural interface could produce intra-spinal stimulation based on ECoG data, for restoration of leg movements after SCI.

Combination of reinforcement learning and bee algorithm for controlling two-link arm with six muscle: simplified human arm model in the horizontal plane.

The aim of this study was to improve reinforcement learning algorithm by combining artificial bee colony algorithm. The traditional method of reinforcement learning algorithm has a very low convergence rate due to random choices. An ant algorithm will help to make random choices in reinforcement learning more appropriate. This hybrid algorithm called the bee colony reinforcement (BCR) algorithm. The tip of the arm must reach a predetermined purpose by BCR algorithm. The results show that the BCR algorithm in the model has been able to reduce the time to reach the goal than the reinforcement learning algorithm (In average 12 steps faster). Also, the path for reaching the goal in the BCR algorithm was far more direct and shorter than the reinforcement learning algorithm. This method also detects the optimal path towards the goal.

A fuzzy-genetic model for estimating forces from electromyographical activity of antagonistic muscles due to planar lower arm movements: the effect of nonlinear muscle properties.

The aim of this paper is to create a model for mapping the surface electromyogram (EMG) signals to the force that generated by human arm muscles. Because the parameters of each person's muscle are individual, the model of the muscle must have two characteristics: (1) The model must be adjustable for each subject. (2) The relationship between the input and output of model must be affected by the force-length and the force-velocity behaviors are proven through Hill's experiments. Hill's model is a kinematic mechanistic model with three elements, i.e. one contractile component and two nonlinear spring elements. In this research, fuzzy systems are applied to improve the muscle model. The advantages of using fuzzy system are as follows: they are robust to noise, they prove an adjustable nonlinear mapping, and are able to model the uncertainties of the muscle. Three fuzzy coefficients have been added to the relationships of force-length (active and passive) and force-velocity existing in Hill's model. Then, a genetic algorithm (GA) has been used as a biological search method that can adjust the parameters of the model in order to achieve the optimal possible fit. Finally, the accuracy of the fuzzy genetic implementation Hill-based muscle model (FGIHM) is invested as following: the FGIHM results have 12.4% RMS error (in worse case) in comparison to the experimental data recorded from three healthy male subjects. Moreover, the FGIHM active force-length relationship which is the key characteristics of muscles has been compared to virtual muscle (VM) and Zajac muscle model. The sensitivity of the FGIHM has been evaluated by adding a white noise with zero mean to the input and FGIHM has proved to have lower sensitivity to input noise than the traditional Hill's muscle model.