Francisco J. Pelayo

Time series analysis using normalized PG-RBF network with regression weights

This paper proposes a framework for constructing and training a radial basis function (RBF) neural network. For this purpose, a sequential learning algorithm is presented to adapt the structure of the network, in which it is possible to create a new hidden unit and also to detect and remove inactive units. The structure of the Gaussian functions is modified using a pseudo-Gaussian function (PG) in which two scaling parameters σ are introduced, which eliminates the symmetry restriction and provides the neurons in the hidden layer with greater flexibility with respect to function approximation. Other important characteristics of the proposed neural system are that the activation of the hidden neurons is normalized which, as described in the bibliography, provides a better performance than nonnormalization and instead of using a single parameter for the output weights, these are functions of the input variables which leads to a significant reduction in the number of hidden units compared to the classical RBF network. Finally, we examine the result of applying the proposed algorithm to time series prediction.

Toward the development of a cortically based visual neuroprosthesis

Motivated by the success of cochlear implants for deaf patients, we are now facing the goal of creating a visual neuroprosthesis designed to interface with the occipital cortex as a means through which a limited but useful sense of vision could be restored in profoundly blind patients. We review the most important challenges regarding this neuroprosthetic approach and emphasize the need for basic human psychophysical research on the best way of presenting complex stimulating patterns through multiple microelectrodes. Continued research will hopefully lead to the development of and design specifications for the first generation of a cortically based visual prosthesis system.

Development of a cortical visual neuroprosthesis for the blind: the relevance of neuroplasticity

Clinical applications such as artificial vision require extraordinary, diverse, lengthy and intimate collaborations among basic scientists, engineers and clinicians. In this review, we present the state of research on a visual neuroprosthesis designed to interface with the occipital visual cortex as a means through which a limited, but useful, visual sense could be restored in profoundly blind individuals. We review the most important physiological principles regarding this neuroprosthetic approach and emphasize the role of neural plasticity in order to achieve desired behavioral outcomes. While full restoration of fine detailed vision with current technology is unlikely in the immediate near future, the discrimination of shapes and the localization of objects should be possible allowing blind subjects to navigate in a unfamiliar environment and perhaps even to read enlarged text. Continued research and development in neuroprosthesis technology will likely result in a substantial improvement in the quality of life of blind and visually impaired individuals.

Real-time tone mapping on GPU and FPGA

Low-level computer vision algorithms have high computational requirements. In this study, we present two real-time architectures using resource constrained FPGA and GPU devices for the computation of a new algorithm which performs tone mapping, contrast enhancement, and glare mitigation. Our goal is to implement this operator in a portable and battery-operated device, in order to obtain a low vision aid specially aimed at visually impaired people who struggle to manage themselves in environments where illumination is not uniform or changes rapidly. This aid device processes in real-time, with minimum latency, the input of a camera and shows the enhanced image on a head mounted display (HMD). Therefore, the proposed operator has been implemented on battery-operated platforms, one based on the GPU NVIDIA ION2 and another on the FPGA Spartan III, which perform at rates of 30 and 60 frames per second, respectively, when working with VGA resolution images (640 × 480).

AN AUDITORY BRAIN-COMPUTER INTERFACE WITH ACCURACY PREDICTION

Fully auditory Brain-computer interfaces based on the dichotic listening task (DL-BCIs) are suited for users unable to do any muscular movement, which includes gazing, exploration or coordination of their eyes looking for inputs in form of feedback, stimulation or visual support. However, one of their disadvantages, in contrast with the visual BCIs, is their lower performance that makes them not adequate in applications that require a high accuracy. To overcome this disadvantage, we employed a Bayesian approach in which the DL-BCI was modeled as a Binary phase shift keying receiver for which the accuracy can be estimated a priori as a function of the signal-to-noise ratio. The results showed the measured accuracy to match the predefined target accuracy, thus validating this model that made possible to estimate in advance the classification accuracy on a trial-by-trial basis. This constitutes a novel methodology in the design of fully auditory DL-BCIs that let us first, define the target accuracy for a specific application and second, classify when the signal-to-noise ratio guarantees that target accuracy.

Analysis of the Functional Block Involved in the Design of Radial Basis Function Networks

The main architectures, learning abilities and applications of radial basis function (RBF) neural networks are well documented. However, to the best of our knowledge, no in-depth analyses have been carried out into the influence on the behaviour of the neural network arising from the use of different alternatives for the design of an RBF (different non-linear functions, distances, number of neurons, structures, etc.). Thus, as a complement to the existing intuitive knowledge, it is necessary to have a more precise understanding of the significance of the different alternatives. In the present contribution, the relevance and relative importance of the parameters involved in such a design are investigated by using a statistical tool, the ANalysis Of the VAriance (ANOVA). In order to obtain results that are widely applicable, various problems of classification, functional approximation and time series estimation are analyzed. Conclusions are drawn regarding the whole set.

Use of Phase in Brain---Computer Interfaces based on Steady-State Visual Evoked Potentials

Brain–computer interfaces based on steady-state visual evoked potentials (SSVEP-BCIs) extract the amplitude of these potentials for classification. The use of the phase has not yet been widely used in on-line classification, since it requires a very accurate real time system that keeps synchronized the stimulation, recording and processing. In this paper, it has been presented an experiment, based on the AM modulation of flickering stimuli, that demonstrates that first, the phase shifts of different stimuli can be recovered from that of the corresponding SSVEPs without the need of a real time system; second, this information can be used efficiently to develop a BCI based on the classification of the phase shifts of the SSVEPs. Since the phase is statistically independent of the amplitude, the joint use in classification of both would improve the performance of this type of BCI.

An auditory brain–computer interface evoked by natural speech

Brain–computer interfaces (BCIs) are mainly intended for people unable to perform any muscular movement, such as patients in a complete locked-in state. The majority of BCIs interact visually with the user, either in the form of stimulation or biofeedback. However, visual BCIs challenge their ultimate use because they require the subjects to gaze, explore and shift eye-gaze using their muscles, thus excluding patients in a complete locked-in state or under the condition of the unresponsive wakefulness syndrome. In this study, we present a novel fully auditory EEG-BCI based on a dichotic listening paradigm using human voice for stimulation. This interface has been evaluated with healthy volunteers, achieving an average information transmission rate of 1.5 bits min⁻¹ in full-length trials and 2.7 bits min⁻¹ using the optimal length of trials, recorded with only one channel and without formal training. This novel technique opens the door to a more natural communication with users unable to use visual BCIs, with promising results in terms of performance, usability, training and cognitive effort.

Translating image sequences into spike patterns for cortical neuro-stimulation

Abstract This paper describes a bioinspired preprocessing and coding system devised for producing optimal multi-electrode stimulation at the cortical level, starting from image sequences and working at video rates. A hybrid platform with software and reconfigurable hardware delivers a continuously varying stream of pulses or spike patterns. The main objective of this work is to build a portable system for a visual neuro-prosthesis to stimulate efficiently an array of intra-cortical implanted microelectrodes. A set of parameters can be adjusted in the processing and spike-coding modules to trade-off their technology constraints with the biological plausibility of their functional features.

Analog CMOS implementation of a discrete time CNN with programmable cloning templates

An analog CMOS implementation of cells for building discrete-time cellular neural networks (DT-CNNs), based on current-mode multipliers and capacitive storage for the analog initial states and cloning templates, is presented. Since the cloning templates are programmable, the circuit could be used for several applications, or it could be reconfigured to perform different tasks on the initial input data sequentially. A chip prototype containing to DT-CNN cells has been designed and manufactured, and some experimental results showing the functionality of the circuit are provided. >