Analyzing EEG signals to detect unexpected obstacles during walking.

BACKGROUND: When an unexpected perturbation in the environment occurs, the subsequent alertness state may cause a brain activation responding to that perturbation which can be detected and employed by a Brain-Computer Interface (BCI). In this work, the possibility of detecting a sudden obstacle appearance analyzing electroencephalographic (EEG) signals is assessed. For this purpose, different features of EEG signals are evaluated during the appearance of sudden obstacles while a subject is walking on a treadmill. The future goal is to use this procedure to detect any obstacle appearance during walking when the user is wearing a lower limb exoskeleton in order to generate an emergency stop command for the exoskeleton. This would enhance the user-exoskeleton interaction, improving the safety mechanisms of current exoskeletons. METHODS: In order to detect the change in the brain activity when an obstacle suddenly appears, different features of EEG signals are evaluated using the recordings of five healthy subjects. Since the change in the brain activity occurs in the time domain, the features evaluated are: common spatial patterns, average power, slope, and the coefficients of a polynomial fit. A Linear Discriminant Analysis-based classifier is used to differentiate between two conditions: the appearance or not of an obstacle. The evaluation of the performance to detect the obstacles is made in terms of accuracy, true positive (TP) and false positive (FP) rates. RESULTS: From the offline analysis, the best performance is achieved when the slope or the polynomial coefficients are used as features, with average detection accuracy rates of 74.0 and 79.5 %, respectively. These results are consistent with the pseudo-online results, where a complete EEG recording is segmented into windows of 500 ms and overlapped 400 ms, and a decision about the obstacle appearance is made for each window. The results of the best subject were 11 out of 14 obstacles detected with a rate of 9.09 FPs/min, and 10 out of 14 obstacles detected with a rate of 6.34 FPs/min using slope and polynomial coefficients features, respectively. CONCLUSIONS: An EEG-based BCI can be developed to detect the appearance of unexpected obstacles. The average accuracy achieved is 79.5 % of success rate with a low number of false detections. Thus, the online performance of the BCI would be suitable for commanding in a safely way a lower limb exoskeleton during walking.

Design of a brain-machine interface for reducing false activations of a lower-limb exoskeleton based on error related potential.

BACKGROUND AND OBJECTIVE: Brain-Machine Interfaces (BMIs) based on a motor imagination paradigm provide an intuitive approach for the exoskeleton control during gait. However, their clinical applicability remains difficulted by accuracy limitations and sensitivity to false activations. A proposed improvement involves integrating the BMI with methods based on detecting Error Related Potentials (ErrP) to self-tune erroneous commands and enhance not only the system accuracy, but also its usability. The aim of the current research is to characterize the ErrP at the beginning of the gait with a lower limb exoskeleton to reduce the false starts in the BMI system. Furthermore, this study is valuable for determining which type of feedback, Tactile, Visual, or Visuo-Tactile, achieves the best performance in evoking and detecting the ErrP. METHODS: The initial phase of the research concentrates on detecting ErrP at the beginning of gait to improve the efficiency of an asynchronous BMI based on motor imagery (BMI-MI) to control a lower limb exoskeleton. Initially, an experimental protocol is designed to evoke ErrP at the start of gait, employing three different stimuli: Tactile, Visual, and Visuo-Tactile. An iterative selection process is then utilized to characterize ErrP in both time and frequency domains and fine-tune various parameters, including electrode distribution, feature combinations, and classifiers. A generic classifier with 6 subjects is employed to configure an ensemble classification system for detecting ErrP and reducing the false starts. RESULTS: The ensembled system configured with the selected parameters yields an average correction of false starts of 72.60 % ± 10.23, highlighting its corrective efficacy. Tactile feedback emerges as the most effective stimulus, outperforming Visual and Visuo-Tactile feedback in both training types. CONCLUSIONS: The results suggest promising prospects for reducing the false starts when integrating ErrP with BMI-MI, employing Tactile feedback. Consequently, the security of the system is enhanced. Subsequent, further research efforts will focus on detecting error potential during movement for gait stop, in order to limit undesired stops.

Analysis of Error Potentials generated by a lower limb exoskeleton feedback in a BMI for gait control.

Brain-machine interfaces (BMIs) based on motor imagery (MI) for controlling lower-limb exoskeletons during the gait have been gaining importance in the rehabilitation field. However, these MI-BMI are not as precise as they should. The detection of error related potentials (ErrP) as a self-tune parameter to prevent wrong commands could be an interesting approach to improve their performance. For this reason, in this investigation ErrP elicited by the movement of a lower-limb exoskeleton against subject's will is analyzed in the time, frequency and time-frequency domain and compared with the cases where the exoskeleton is correctly commanded by motor imagery (MI). The results of the ErrP study indicate that there is statistical significative evidence of a difference between the signals in the erroneous events and the success events. Thus, ErrP could be used to increase the accuracy of BMIs which commands exoskeletons.Clinical Relevance- This investigation has the purpose of improving brain-machine interfaces (BMIs) based on motor imagery (MI) by means of the detection of error potentials. This could promote the adoption of robotic exoskeletons commanded by BMIs in rehabilitation therapies.

Transfer Learning with CNN Models for Brain-Machine Interfaces to command lower-limb exoskeletons: A Solution for Limited Data.

This study evaluates the performance of two convolutional neural networks (CNNs) in a brain-machine interface (BMI) based on motor imagery (MI) by using a small dataset collected from five participants wearing a lower-limb exoskeleton. To address the issue of limited data availability, transfer learning was employed by training models on EEG signals from other subjects and subsequently fine-tuning them to specific users. A combination of common spatial patterns (CSP) and linear discriminant analysis (LDA) was used as a benchmark for comparison. The study's primary aim is to examine the potential of CNNs and transfer learning in the development of an automatic neural classification system for a BMI based on MI to command a lower-limb exoskeleton that can be used by individuals without specialized training.Clinical Relevance- BMI can be used in rehabilitation for patients with motor impairment by using mental simulation of movement to activate robotic exoskeletons. This can promote neural plasticity and aid in recovery.

Effects of an Exoskeleton-Assisted Gait Motor Imagery Training in Functional Brain Connectivity.

Lower-limb exoskeletons have been used in gait rehabilitation to facilitate the restoration of motor skills. These robotics systems could be complemented by Brain-Computer Interfaces (BCIs) to assist or rehabilitate people with walking disabilities. In this preliminary study, electroencephalography-based brain functional connectivity is analyzed during exoskeleton-assisted gait motor imagery (MI) training. Partial Directed Coherence (PDC) analysis was employed to assess the exchange of information flow between EEG signals during gait MI in four healthy subjects, two using an exoskeleton and two without using it. Besides, in order to explore the functional connectivity, an outflow index based on the number of significant directed connectivities revealed by the PDC analysis is proposed. We found that the outflow index increases in the central zone (C2, C3, C4) while decreases in the central-parietal (CP1, CP2) and fronto-central (FC1) zones when the training was assisted by an exoskeleton. The results obtained can be useful to obtain informative features for BCI applications as well as in motor rehabilitation.

Novel tDCS montage favors lower limb motor imagery detection.

This work studies a novel transcranial direct current stimulation (tDCS) montage to improve a brain-machine interface (BMI) lower limb motor imagery detection. The tDCS montage is composed by two anodes and one cathode. One anode is located over the motor cortex and the other one over the cerebellum. Ten healthy subjects participated in this experiment. They were randomly separated into two groups: sham, which received a fake stimulation, and active tDCS, which received a real stimulation. Each subject was experimented on five consecutive days. Results pointed out that there was a significant difference $(p < 0 .05)$ in the classification accuracy between the sham and the active tDCS group. On each of the five days of the experiment the active tDCS group achieved better accuracy results than the sham group: 4%, 10%, 10%, 9% and 7% higher respectively.

Effect on the classification of motor imagery in EEG after applying anodal tDCS with a 4×1 ring montage over the motor cortex.

Transcranial direct stimulation (tDCS) is a technique for modulating brain excitability that has potential to be used in motor neurorehabilitation by enhancing motor activity, such as motor imagery (MI). tDCS effects depend on different factors, like current density and the position of the stimulating electrodes. This study presents preliminary results of the evaluation of the effect of current density on MI performance by measuring right-hand/feet MI accuracy of classification from electroencephalographic (EEG) measurements after anodal tDCS is applied with a 4×1 ring montage over the right-hand or feet motor cortex. Results suggest that there might be an enhancement of feet MI when tDCS is applied over the right-hand motor cortex, but further evaluation is required. If results are confirmed with a larger sample, the montage could be used to optimize feet MI performance and improve the outcome of MI-based brain-computer interfaces, which are used during motor neurorehabilitation.

Similar short elements in the 5' regions of the STA2 and SGA genes from Saccharomyces cerevisiae.

The 5' region of the SGA and STA2 genes, encoding the intra- and extracellular glucoamylases, respectively, from Saccharomyces cerevisiae have been sequenced. In addition, the transcription initiation sites have been determined. Four distinct short elements (named I to IV) were found in both genes. Element III has the consensus sequence PuCATTTAPiG with a bilateral symmetry around the central T, and is present in both genes as a direct repeat. This motive seems responsible for the coregulation of STA2 and SGA by the repressor STA10 gene of S. cerevisiae.

Transposition of Tn 1 to the Rhizobium meliloti genome.

A derivative of the IncP1 plasmid RP4, carrying the thermoinducible prophage Mucts62, was obtained in Escherichia coli K12 J53 (RP4). It was impossible to maintain the hybrid plasmid RP4::Mucts62 in Rhizobium meliloti GR4. Thus, it was used as a vehicle for introducing the ampicillin-resistant transposon Tn1 into the R. meliloti genome. Transposition of Tn1 did not generate auxotrophic strains, suggesting that the insertion of Tn1 into the R. meliloti genome was relatively specific. Two chromosomal hot spots for Tn1 insertion were identified by cotransductional analysis, after general transduction by phage DF2. Plasmid-curing experiments, carried out by heat treatment, revealed that symbiotic plasmid(s) also contain at least one site for Tn1 insertion.