Decoding kinetic features of hand motor preparation from single-trial EEG using convolutional neural networks.

Building accurate movement decoding models from brain signals is crucial for many biomedical applications. Predicting specific movement features, such as speed and force, before movement execution may provide additional useful information at the expense of increasing the complexity of the decoding problem. Recent attempts to predict movement speed and force from the electroencephalogram (EEG) achieved classification accuracies at or slightly above chance levels, highlighting the need for more accurate prediction strategies. Thus, the aims of this study were to accurately predict hand movement speed and force from single-trial EEG signals and to decode neurophysiological information of motor preparation from the prediction strategies. To these ends, a decoding model based on convolutional neural networks (ConvNets) was implemented and compared against other state-of-the-art prediction strategies, such as support vector machines and decision trees. ConvNets outperformed the other prediction strategies, achieving an overall accuracy of 84% in the classification of two different levels of speed and force (four-class classification) from pre-movement single-trial EEG (100 ms and up to 1,600 ms prior to movement execution). Furthermore, an analysis of the ConvNet architectures suggests that the network performs a complex spatiotemporal integration of EEG data to optimize classification accuracy. These results show that movement speed and force can be accurately predicted from single-trial EEG, and that the prediction strategies may provide useful neurophysiological information about motor preparation.

Electrical stimulation of the brain and the development of cortical visual prostheses: An historical perspective.

Rapid advances are occurring in neural engineering, bionics and the brain-computer interface. These milestones have been underpinned by staggering advances in micro-electronics, computing, and wireless technology in the last three decades. Several cortically-based visual prosthetic devices are currently being developed, but pioneering advances with early implants were achieved by Brindley followed by Dobelle in the 1960s and 1970s. We have reviewed these discoveries within the historical context of the medical uses of electricity including attempts to cure blindness, the discovery of the visual cortex, and opportunities for cortex stimulation experiments during neurosurgery. Further advances were made possible with improvements in electrode design, greater understanding of cortical electrophysiology and miniaturisation of electronic components. Human trials of a new generation of prototype cortical visual prostheses for the blind are imminent. This article is part of a Special Issue entitled Hold Item.

Web-based neuromuscular simulator applied to the teaching of principles of neuroscience

INTRODUCTION: The learning of core concepts in neuroscience can be reinforced by a hands-on approach, either experimental or computer-based. In this work, we present a web-based multi-scale neuromuscular simulator that is being used as a teaching aid in a campus-wide course on the Principles of Neuroscience. METHODS: The simulator has several built-in individual models based on cat and human biophysics, which are interconnected to represent part of the neuromuscular system that controls leg muscles. Examples of such elements are i) single neurons, representing either motor neurons or interneurons mediating reciprocal, recurrent and Ib inhibition; ii) afferent fibers that can be stimulated to generate spinal reflexes; iii) muscle unit models, generating force and electromyogram; and iv) stochastic inputs, representing the descending volitional motor drive. RESULTS: Several application examples are provided in the present report, ranging from studies of individual neuron responses to the collective action of many motor units controlling muscle force generation. A subset of them was included in an optional homework assignment for Neuroscience and Biomedical Engineering graduate students enrolled in the course cited above at our University. Almost all students rated the simulator as a good or an excellent learning tool, and approximately 90% declared that they would use the simulator in future projects. CONCLUSION: The results allow us to conclude that multi-scale neuromuscular simulator is an effective teaching tool. Special features of this free teaching resource are its direct usability from any browser (http://remoto.leb.usp.br/), its user-friendly graphical user interface (GUI) and the preset demonstrations.