A major challenge in daily automotive use is the high incidence of car accidents, primarily attributed to human errors. This paper introduces Brain-Brake, an innovative brain-operated advanced driver-assistance system (ADAS) designed for collision avoidance. Utilizing electroencephalography (EEG) signals, Brain-Brake detects a driver's braking intention faster than traditional methods, addressing critical reaction time delays. It integrates a computer vision module that measures the proximity of objects to the vehicle, complementing the EEG- based intention detection for enhanced reliability. Implemented on embedded chips using the AUTOSAR framework, preliminary results indicate a significant reduction in reaction times during emergency braking scenarios, highlighting the system's potential to enhance road safety significantly.

The development of practical Brain-Computer Interface (BCI) systems has been hindered by significant issues related to data, specifically the lack of sufficient data needed for training. To address this challenge, generating synthetic data that mimics real recorded data has been proposed to augment the real data. One promising technique for data augmentation is through the use of Generative Adversarial Networks (GANs), which have been successfully applied in many other fields. This paper proposes a novel GAN-based approach for generating synthetic spectrum images of Motor Imagery (MI) Electroencephalogram (EEG). The proposed GAN is examined with two Convolutional Neural Network (CNN) architectures in the context of MI classification. Using the public dataset BCI competition IV, our findings reveal that the generated EEG spectrum images using GANs exhibit temporal, spectral, and spatial characteristics similar to the real ones. The average classification accuracy of right-hand versus left-hand MI using the proposed GAN/CNN models has improved to 76.71% with an enhancement of 2.5% in comparison to using the CNN applied to the real data only. These results suggest that using GANs could improve MI BCI systems with limited data.

The performance of P300-based Brain-Computer Interface (BCI) applications is highly dependent on both the quality and quantity of the recorded Electroencephalography (EEG) signals. As recording extended datasets from users for calibration is often a difficult and tedious task, data augmentation can be used to help supplement the training data for machine learning classifiers that are typically used in P300-based BCI applications. In this paper, we analyze and compare the performance of three different generative adversarial networks (GANs) as data augmentation techniques; namely, deep convolutional GAN (DCGAN), conditional GAN (cGAN), and the auxiliary classifier GAN (ACGAN). We first investigated the effect of increasing the training data size using each of these GANs on the performance of P300 classification. Our results revealed that the cGAN increased the classification accuracy by up to 18% relative to the baseline data under the best conditions. We also investigated the effect of decreasing the training data size and compensating for the reduced data size using data generated from the GANs. Our analysis indicated that the training data size could be reduced by ~30% while maintaining the accuracy on par with the baseline accuracy. These results demonstrate the utility of GANs in addressing the challenges associated with the limited data typically available for BCI applications.

Visual prostheses have been introduced to partially restore vision to the blind via visual pathway stimulation. Despite their success, some challenges have been reported by the implanted patients. One of those challenges is the difficulty of object recognition due to the low resolution of the images perceived through these devices. In this paper, a deep learning-based approach combined with image pre-processing is proposed to allow visual prostheses' users to recognize objects in a given scene. The approach simplifies the objects in the scene by displaying the objects in clip art form to enhance object recognition. These clip art images are generated by, first, identifying the objects in the scene using the You Only Look Once (YOLO) deep neural network. The clip art corresponding to each identified object is then retrieved via Google Images. Three experiments were conducted to measure the success of the proposed approach using simulated prosthetic vision. Our results reveal a remarkable decrease in the recognition time, increase in the recognition accuracy and confidence level when using the clip art representation as opposed to using the actual images of the objects. These results demonstrate the utility of object simplification in enhancing the perception of images in prosthetic vision.

Cardiovascular diseases and events are the principal source of mortality worldwide. To reduce their fatality, the prediction of such events is of high significance. The usage of signal processing and machine learning techniques applied to Electrocardiogram (ECG) signals in the prognosis has been investigated in recent years. This paper presents the Majority-vote Over Multiple ECG Segments for Risk Assessment (MOMESRA) approach that comprises the use of multiple ECG segments from a single patient and the utilization of majority-vote to predict cardiovascular risk. In this approach, time-domain, frequency-domain and nonlinear Heart Rate Variability (HRV) features are extracted from 1 hour ECG segments subdivided into 2-minute windows. Features are subsequently reduced via Principal Component Analysis (PCA). The reduced features are then used as input to a Support Vector Machine (SVM) classifier. The proposed approach attained an accuracy of 78 %, a sensitivity rate of 90 %, a specificity rate of 67 % and a precision of 79 %. These results indicate the utility of the proposed approach in predicting cardiovascular events.

P300 speller systems represent one of the most basic applications of Brain-Computer Interfaces (BCIs). A traditional P300 speller consists of a 6 by 6 grid of characters in which each column or row in this grid intensifies at random. During such intensification process, the electroencephalography (EEG) data of the subject is recorded and analyzed to determine the character to be spelled. In this paper, we demonstrate how to improve on the traditional P300 speller by investigating the effects of incorporating different color luminance in the columns and rows of the spellers grid-of-characters (i.e. red, green and blue) as opposed to the conventional one-color (i.e. gray-scale) luminance. In our analysis, we used the Emotiv Neuroheadset to record scalp EEG obtained from the frontal, parietal and occipital brain regions. We examine four different feature extraction techniques in addition to two classifiers, namely, Linear Discriminant Analysis (LDA) and Linear Support vector machines (LSVM). Offline and online tests conducted on four subjects demonstrate a significant performance increase (up to 16%) for the intermixed color luminance case compared to the gray luminance one. These results indicate the efficacy of incorporating colors into P300 spellers interface.

Predicting the occurrence of epileptic seizures can provide an enormous aid to epileptic patients. This paper introduces a novel patient-specific method for seizure prediction applied to scalp Electroencephalography (EEG) signals. The proposed method relies on the count of zero-crossings of wavelet detail coefficients of EEG signals as the major feature. This is followed by a binary classifier that discriminates between preictal and interictal states. The proposed method is practical for real-time applications given its computational efficiency as it uses an adaptive algorithm for channel selection to identify the optimum number of needed channels. Moreover, this method is robust against the variability across seizures for the same patient. Applied to data from 8 patients, the proposed method achieved high accuracy and sensitivity with an average accuracy of 94% and an average sensitivity of 96%. These results were obtained using only 10 minutes of training data as opposed to using hours of recordings typically used in traditional approaches.

Restoring vision is no longer impossible as a result of recent advances in neural interfaces. Successful demonstrations of retinal implants motivate the development of more effective visual prostheses. The thalamic Lateral Geniculate Nucleus (LGN) is one potential deep-brain interfacing site for visual prostheses. A main challenge in developing thalamic as well as other visual prostheses is optimizing the parameters of electrical stimulation. This paper introduces a Kalman-based optimal encoder whose function is to determine the optimal electrical stimulation parameters required to induce a certain visual sensation. The performance of the proposed approach is demonstrated using a probabilistic model of LGN neurons. Results demonstrate a significant similarity between neuronal responses obtained using electrical stimulation and the responses obtained using the corresponding visual stimuli with a mean correlation of 0.62 (P <; 0.01, n = 54). These results indicate the efficacy of the proposed optimal encoder in driving LGN neurons to induce visual sensations.

Coordination among cortical neurons is believed to be a key element in mediating many high-level cortical processes such as perception, attention, learning, and memory formation. Inferring the structure of the neural circuitry underlying this coordination is important to characterize the highly nonlinear, time-varying interactions between cortical neurons in the presence of complex stimuli. In this work, we investigate the applicability of dynamic Bayesian networks (DBNs) in inferring the effective connectivity between spiking cortical neurons from their observed spike trains. We demonstrate that DBNs can infer the underlying nonlinear and time-varying causal interactions between these neurons and can discriminate between mono- and polysynaptic links between them under certain constraints governing their putative connectivity. We analyzed conditionally Poisson spike train data mimicking spiking activity of cortical networks of small and moderately large size. The performance was assessed and compared to other methods under systematic variations of the network structure to mimic a wide range of responses typically observed in the cortex. Results demonstrate the utility of DBN in inferring the effective connectivity in cortical networks.

An essential step towards understanding how the brain orchestrates information processing at the cellular and population levels is to simultaneously observe the spiking activity of cortical neurons that mediate perception, learning, and motor processing. In this paper, we formulate an information theoretic approach to determine whether cooperation among neurons may constitute a governing mechanism of information processing when encoding external covariates. Specifically, we show that conditional independence between neuronal outputs may not provide an optimal encoding strategy when the firing probability of a neuron depends on the history of firing of other neurons connected to it. Rather, cooperation among neurons can provide a "message-passing" mechanism that preserves most of the information in the covariates under specific constraints governing their connectivity structure. Using a biologically plausible statistical learning model, we demonstrate the performance of the proposed approach in synergistically encoding a motor task using a subset of neurons drawn randomly from a large population. We demonstrate its superiority in approximating the joint density of the population from limited data compared to a statistically independent model and a maximum entropy (MaxEnt) model.

A fundamental goal in systems neuroscience is to infer the functional connectivity among neuronal elements coordinating information processing in the brain. In this work, we investigate the applicability of dynamic Bayesian networks (DBN) in inferring the structure of cortical networks from the observed spike trains. DBNs have unique features that make them capable of detecting causal relationships between spike trains such as modeling time-dependent relationships, detecting non-linear interactions and inferring connectivity between neurons from the observed ensemble activity. A probabilistic point process model was used to assess the performance under systematic variations of the model parameters. Results demonstrate the utility of DBN in inferring functional connectivity in cortical network models.

Understanding the mechanism underlying distributed neural coding is a fundamental goal in computational neuroscience. With the ability to simultaneously observe the activity of large networks of neurons in response to external stimuli, a natural question arises: how the outside world is represented in the collective activity of these neurons? In this work, we provide an information theoretic approach for determining the role of cooperation among neurons in encoding external stimuli. Specifically, we show that statistical independence between neuronal outputs may not provide the best coding strategy when these outputs depend on the history of other neuronal constituents in the network. Rather, cooperation among neurons can provide a near optimal and lossless coding strategy under specific constraints governing their network structure. Using a statistical learning model, we demonstrate the performance of the proposed approach in decoding a motor task with both discrete targets and continuous trajectory using spike trains from a small subset of a large network. We demonstrate its superiority in minimizing the decoding error compared to a statistically independent model and to other classical decoders reported in the literature.

Identifying functional connectivity between neuronal elements is an essential first step toward understanding how the brain orchestrates information processing at the single-cell and population levels to carry out biological computations. This letter suggests a new approach to identify functional connectivity between neuronal elements from their simultaneously recorded spike trains. In particular, we identify clusters of neurons that exhibit functional interdependency over variable spatial and temporal patterns of interaction. We represent neurons as objects in a graph and connect them using arbitrarily defined similarity measures calculated across multiple timescales. We then use a probabilistic spectral clustering algorithm to cluster the neurons in the graph by solving a minimum graph cut optimization problem. Using point process theory to model population activity, we demonstrate the robustness of the approach in tracking a broad spectrum of neuronal interaction, from synchrony to rate co-modulation, by systematically varying the length of the firing history interval and the strength of the connecting synapses that govern the discharge pattern of each neuron. We also demonstrate how activity-dependent plasticity can be tracked and quantified in multiple network topologies built to mimic distinct behavioral contexts. We compare the performance to classical approaches to illustrate the substantial gain in performance.

Brain machine interfaces (BMIs) have recently received significant attention from the neuroscience and engineering communities as a result of striking advances in monitoring, processing, and modeling brain function at multiple temporal and spatial resolutions. These advances, however, have also raised significant challenges to both communities that are becoming the focus of numerous ongoing research efforts. Broadly categorized based on their level of invasiveness, BMIs relying on implantable microelectrode arrays (MEAs) have received the most attention. This paper briefly reviews some fundamental concepts underlying the operation of MEA-based BMIs and highlights in particular the signal processing challenges faced by these systems in light of their resource-constrained operation. Finally, we summarize some of our recent progress in this area and suggest some open questions for future research.