Due to the increasing fatal traffic accidents, there are strong desire for more effective and convenient techniques for driving fatigue detection. Here, we propose a unified frameworkE-Keyto simultaneously perform personal identification (PI) and driving fatigue detection using a convolutional attention neural network (CNN-Attention). The performance was assessed using EEG data collected through a wearable dry-sensor system from 31 healthy subjects undergoing a 90-min simulated driving task. In comparison with three widely-used competitive models (including CNN, CNN-LSTM, and Attention), the proposed scheme achieved the best (p < 0.01) performance in both PI (98.5%) and fatigue detection (97.8%). Besides, the spatial-temporal structure of the proposed framework exhibits an optimal balance between classification performance and computational efficiency. Additional validation analyses were conducted to assess the reliability and practicability of the model via re-configuring the kernel size and manipulating the input data, showing that it can achieve a satisfactory performance using a subset of the input data. In sum, these findings would pave the way for further practical implementation of in-vehicle expert system, showing great potential in autonomous driving and car-sharing where currently monitoring of PI and driving fatigue are of particular interest.

A single dataset could hide a significant number of relationships among its feature set. Learning these relationships simultaneously avoids the time complexity associated with running the learning algorithm for every possible relationship, and affords the learner with an ability to recover missing data and substitute erroneous ones by using available data. In our previous research, we introduced the gate-layer autoencoders (GLAEs), which offer an architecture that enables a single model to approximate multiple relationships simultaneously. GLAE controls what an autoencoder learns in a time series by switching on and off certain input gates, thus, allowing and disallowing the data to flow through the network to increase network's robustness. However, GLAE is limited to binary gates. In this article, we generalize the architecture to weighted gate layer autoencoders (WGLAE) through the addition of a weight layer to update the error according to which variables are more critical and to encourage the network to learn these variables. This new weight layer can also be used as an output gate and uses additional control parameters to afford the network with abilities to represent different models that can learn through gating the inputs. We compare the architecture against similar architectures in the literature and demonstrate that the proposed architecture produces more robust autoencoders with the ability to reconstruct both incomplete synthetic and real data with high accuracy.

Schizophrenia seriously affects the quality of life. To date, both simple (e.g., linear discriminant analysis) and complex (e.g., deep neural network) machine-learning methods have been utilized to identify schizophrenia based on functional connectivity features. The existing simple methods need two separate steps (i.e., feature extraction and classification) to achieve the identification, which disables simultaneous tuning for the best feature extraction and classifier training. The complex methods integrate two steps and can be simultaneously tuned to achieve optimal performance, but these methods require a much larger amount of data for model training. To overcome the aforementioned drawbacks, we proposed a multikernel capsule network (MKCapsnet), which was developed by considering the brain anatomical structure. Kernels were set to match partition sizes of the brain anatomical structure in order to capture interregional connectivities at the varying scales. With the inspiration of the widely used dropout strategy in deep learning, we developed capsule dropout in the capsule layer to prevent overfitting of the model. The comparison results showed that the proposed method outperformed the state-of-the-art methods. Besides, we compared performances using different parameters and illustrated the routing process to reveal characteristics of the proposed method. MKCapsnet is promising for schizophrenia identification. Our study first utilized the capsule neural network for analyzing functional connectivity of magnetic resonance imaging (MRI) and proposed a novel multikernel capsule structure with the consideration of brain anatomical parcellation, which could be a new way to reveal brain mechanisms. In addition, we provided useful information in the parameter setting, which is informative for further studies using a capsule network for other neurophysiological signal classification.

Trust in autonomous vehicles (AV) is of critical importance and demands comprehensive interdisciplinary research. While most studies utilize subjective measures, we employ electroencephalography (EEG) to study in a more objective manner the cognitive states associated with trust during AV driving. Subjects drove a simulated AV in Conditional Automation Driving (SAE L3) and Full Automation Driving (SAE L5) modes. In the experimental design, malfunctions were induced at both automation levels. Self-reported trust in the AV was reduced immediately after Full Automation malfunctions, but not after Conditional Automation malfunctions when subjects were able to take over vehicle control to avoid danger. EEG analyses reveal that during Full Automation malfunctions, there was a significant enhancement in approach motivation (i.e. desire to re-engage) and a disruption of right frontal functional clustering that supports executive cognition (i.e. planning and decision-making). No neurocognitive disruptions were observed during Conditional Automation malfunctions. Our results demonstrate that it is not automation malfunctions per se (e.g. failure to decelerate) that deteriorate trust, but rather the inability for human drivers to adaptively mitigate the risk of negative outcomes (e.g. risk of crashing) resulting from those malfunctions. This is reflected in changes in brain activity associated with motivational state and action planning. Keeping the human driver on-the-loop protects against trust loss. Frontal alpha EEG is a neural correlate of trust-in-automation, with potential for trust monitoring using wearable technology to support driver-vehicle adaptivity.

In the nascent field of neuroergonomics, mental workload assessment is one of the most important issues and has an apparent significance in real-world applications. Although prior research has achieved efficient single-task classification, scatted studies on cross-task mental workload assessment usually result in unsatisfactory performance. Here, we introduce a data-driven analysis framework to overcome the challenges regarding task-independent workload assessment using a fusion of EEG spectral characteristics and unveil the common neural mechanisms underlying mental workload. Specifically, multi-frequency power spectrum and functional connectivity (FC) were estimated for two workload levels in two working-memory tasks performed by 40 healthy participants, subsequently being fed into a machine learning approach to obtain the importance of each feature vector and evaluate classification performance in a cross-task fashion. Our framework achieved a classification accuracy of 0.94 for task-independent mental workload discrimination. Further investigation of the designated features in terms of their spectral and localization properties revealed task-independent common patterns in the neural mechanisms governing workload. In particular, increased workload was associated with elevated frontal delta and theta power but reduced parietal alpha power, whereas FC exhibited complex frequency- and region-dependent alterations. By implication, the employment of the EEG feature fusion emphasized their utility in serving as promising indicators for different workload conditions applications.

The emergence of diseases and drug-induced perturbations are oftentimes the cause of biological pathway deregulations. Identifying differentially expressed subpathways in organism-level networks of signaling pathways can be a computationally intensive undertaking, due to their complexity. In this direction, we present a subpathway analysis method which refines organism-level networks via a two step-approach. The method first constructs a core-network of differentially expressed genes and subsequently includes a set of topologically significant non-differentially expressed genes, that both exhibit correlated expression levels with their neighbors and facilitate the signal propagation within the core-network. The refined network is then searched for differentially expressed subpathways using a plethora of subpathway identification methods and checked for enrichment in functional terms such as drugs and diseases. The approach assesses the differential expression of the subpathways using a consensus approach, by detecting even weak signals of differential expression, while accounting for correlations which arise in gene expression data.

Autoencoders (AE) have been used successfully as unsupervised learners for inferring latent information, learning hidden features and reducing the dimensionality of the data. In this paper, we propose a new AE architecture: Gate-Layer AE (GLAE). The novelty of GLAE lies in its ability to encourage learning of the relationships among different input variables, which affords it with an inherent ability to recover missing variables from the available ones and to act as a concurrent multi-function approximator.GLAE uses a network architecture that associates each input with a binary gate acting as a switch that turns on or off the flow to each input unit, while synchronising its action with data flow to the network. We test GLAE with different coding sizes and compare its performance against the Classic AE, Denoising AE and Variational AE. The evaluation uses Electroencephalograph (EEG) data with an aim to reconstruct the EEG signal when some data are missing. The results demonstrate GLAE's superior performance in reconstructing EEG signals with up to 25% missing data in an input stream.

A major rise in the prevalence and impact of Alzheimer's disease (AD) is projected in the coming decades, resulting from increasing life expectancy, thus leading to substantially increased healthcare costs. While brain disfunctions at the time of diagnosis are irreversible, it is widely accepted that AD pathology develops decades before clinical symptoms onset. If incipient processes can be detected early in the disease progression, prospective intervention for preventing or slowing the disease can be designed. Currently, there is no noninvasive biomarker available to detect and monitor early stages of disease progression. The complex etiology of AD warrants a systems-based approach supporting the integration of multimodal and multilevel data, while network-based modeling provides the scaffolding for methods revealing complex systems-level disruptions initiated by the disease. In this work, we review current state-of-the-art, focusing on network-based biomarkers at molecular and brain functional connectivity levels. Particular emphasis is placed on outlining recent trends, which highlight the functional importance of modular substructures in molecular and connectivity networks and their potential biomarker value. Our perspective is rooted in network medicine and summarizes the pipelines for identifying network-based biomarkers, as well as the benefits of integrating genotype and brain phenotype information for a comprehensively noninvasive approach in the early diagnosis of AD. Finally, we propose a framework for integrating knowledge from molecular and brain connectivity levels, which has the potential to enable noninvasive diagnosis, provide support for monitoring therapies, and help understand heretofore unexamined deep level relations between genotype and brain phenotype.

Brain computer interface (BCI) technique has been demonstrated that human intentions or stimulus perception can be recognized using EEG signal recorded from the human scalp. When an intention is initiated in the brain or an external stimulus is perceived, the underlying relevant processing alters brain activity. This alteration in brain activity can be reflected in EEG signal. The intention or stimulus perception is therefore classified based on the alteration in brain activity. It might be difficult to classify brain activities in the perception of shape-analogous English letters because the similar shape could lead to less difference in brain activity. In order to explore classification feasibility and classification performance of shape-analogous letters using EEG signal, we performed an experiment of shape-analogous letter perception, in where participants perceived four letters (i.e., 'p', 'q', 'b' and 'd') while EEG signal was recorded. The F-score method was employed to assess the discriminative power for each feature, and a subgroup of features with high discriminative powers was then selected and fed into classifiers. Five classifiers (i.e., k-Nearest Neighbors (kNN), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Random Forest (RF) and AdaBoost (ADA)), which are either pervasive or advanced in the field of machine learning, were utilized to classify brain activities in perception of shape-analogous letters. For each classifier, its parameters and the number of used features were optimized. Based on the performance comparison among the classifiers, Random Forest (RF) classifier achieved a maximal accuracy of 74.1%, but it was not statistically significantly better than the SVM. Our study demonstrated that brain activities in perception of shape-analogous English letters can be classified based on EEG signal and showed that random forest classifier outperformed other classifiers according to the results of comparison.

Workload is the amount of cognitive effort executed by a certain subject. Several attempts have been done in order to measure workload level. However, there exists a difficulty in analyzing workload: the problem of individuality, or variability among different individuals and how they respond to similar tasks. In order to have a more objective measure of workload level, the authors employed a more direct analysis upon the system that does cognitive work itself. The use of electroencephalogram (EEG) was employed to measure brain signals and process them to get an objective estimation of workload level. In this study, the authors evaluated the workload level related to complex training-based type task. Piloting simulation task was used to represent such type of task. The authors assess the EEG channel connections and found important connections for the estimation of workload level. This information can be used to build a more an EEG-based workload level estimator that is more efficient, i.e. less channels needed be used to accurately construct the estimation. The authors also found the significant brain signal frequency band that is related to the measure of workload in complex tasks. The problem of individual differences was also resolved using the proposed algorithm.

DEsubs is a network-based systems biology R package that extracts disease-perturbed subpathways within a pathway network as recorded by RNA-seq experiments. It contains an extensive and customized framework with a broad range of operation modes at all stages of the subpathway analysis, enabling so a case-specific approach. The operation modes include pathway network construction and processing, subpathway extraction, visualization and enrichment analysis with regard to various biological and pharmacological features. Its capabilities render DEsubs a tool-guide for both the modeler and experimentalist for the identification of more robust systems-level drug targets and biomarkers for complex diseases. AVAILABILITY AND IMPLEMENTATION: DEsubs is implemented as an R package following Bioconductor guidelines: http://bioconductor.org/packages/DEsubs/ CONTACT: [email protected] information: Supplementary data are available at Bioinformatics online.

MOTIVATION: In the era of network medicine and the rapid growth of paired time series mRNA/microRNA expression experiments, there is an urgent need for pathway enrichment analysis methods able to capture the time- and condition-specific 'active parts' of the biological circuitry as well as the microRNA impact. Current methods ignore the multiple dynamical 'themes'-in the form of enriched biologically relevant microRNA-mediated subpathways-that determine the functionality of signaling networks across time. RESULTS: To address these challenges, we developed time-vaRying enriCHment integrOmics Subpathway aNalysis tOol (CHRONOS) by integrating time series mRNA/microRNA expression data with KEGG pathway maps and microRNA-target interactions. Specifically, microRNA-mediated subpathway topologies are extracted and evaluated based on the temporal transition and the fold change activity of the linked genes/microRNAs. Further, we provide measures that capture the structural and functional features of subpathways in relation to the complete organism pathway atlas. Our application to synthetic and real data shows that CHRONOS outperforms current subpathway-based methods into unraveling the inherent dynamic properties of pathways. AVAILABILITY AND IMPLEMENTATION: CHRONOS is freely available at http://biosignal.med.upatras.gr/chronos/ CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

In a recent work, we have studied networks of two-dimensional and three-dimensional Hindmarsh-Rose oscillators and have discovered some very interesting oscillatory phenomena, called chimera states, in which synchronized neuronal ensembles coexist with completely asynchronous ones. In this paper, we summarize our work in connection with other studies on nonlocally coupled FitzHugh-Nagumo oscillators, examine the occurrence of chimera states in coupled bistable elements and point out that mixed oscillatory states also exist, in which desynchronized neurons are interspersed among neurons that oscillate in synchronous fashion. We also demonstrate, by a preliminary study, that it is possible to control these states by varying an external current parameter applied to the main potential variable in order to observe new phenomena that may be relevant in neuroscience applications.

An effective speech brain machine interface requires selecting the best cortical recording sites and signal features for decoding speech production, but also minimal clinical risk for the patient. Motivated by this need to reduce patient risk, the purpose of this study is to detect voice activity (speech onset and offset) automatically from spatial-spectral features of electrocorticographic signals using the optimal number of sensors (minimal invasiveness). ECoG signals were recorded while a subject performed two different syllable repetition tasks. We found that the optimal frequency resolution for detecting voice activity is 8 Hz using 31 sensors out of 55, achieving 98.2% accuracy by employing support vector machines (SVM) as a classifier, and that acceptable accuracy of 96.7% was achieved using 15 sensors, which would permit a less invasive surgery for the placement of electrodes. The proposed voice activity detector may be utilized as a part of an ECoG-based automated natural speech BMI system.

In this paper, we perform spatio-spectral analysis of the human cortex with implanted electrocorticographic (ECoG) electrodes during the voice production process. For this study, the ECoG signals were recorded while the subject performed two-syllable tasks. Additionally, assuming that the speech activity of a subject is expressed as ECoG signal activity disparately distributed over the space of the electrodes, we examined the spectral information in response to the electrode locations. The study was based on spectral features (power spectral density) estimated for each electrode. Quantitative analysis based on the Relief algorithm was followed to estimate the degree of importance of each electrode for describing the voice activity. The experimental results showed that the spectral analysis with resolution of 8 Hz offers the highest voice discrimination performance (94.2%) using support vector machines as classifier. Finally, our analysis showed that during voice activity the frequency bands [168, 208] Hz are mostly affected.

The aim of this paper is to introduce a novel semisupervised scheme for abnormality detection and segmentation in medical images. Semisupervised learning does not require pathology modeling and, thus, allows high degree of automation. In abnormality detection, a vector is characterized as anomalous if it does not comply with the probability distribution obtained from normal data. The estimation of the probability density function, however, is usually not feasible due to large data dimensionality. In order to overcome this challenge, we treat every image as a network of locally coherent image partitions (overlapping blocks). We formulate and maximize a strictly concave likelihood function estimating abnormality for each partition and fuse the local estimates into a globally optimal estimate that satisfies the consistency constraints, based on a distributed estimation algorithm. The likelihood function consists of a model and a data term and is formulated as a quadratic programming problem. The method is applied for automatically segmenting brain pathologies, such as simulated brain infarction and dysplasia, as well as real lesions in diabetes patients. The assessment of the method using receiver operating characteristic analysis demonstrates improvement in image segmentation over two-group analysis performed with Statistical Parametric Mapping (SPM).

Proteomic analysis using mass spectrometry data is a powerful tool for biomarker discovery. However, proteomic data suffers from two crucial problems i/ are inherently very noisy and ii/ the number of features that finally characterize each spectrum is usually very large. In the present study, a well-established framework of data preprocessing steps was developed to deal with the problem of noise, incorporating smoothing, normalization, peak detection, and peak alignment algorithms. In addition, to alleviate the problem of feature dimensionality, a novel iterative peak selection method was developed for choosing peaks (features) from the pre- processed spectra, based on sequential global thresholding followed by particle swarm optimization. These features were fed into a probabilistic neural network algorithm, in order to discriminate healthy from prostate cancer cases and, thus, to determine, through the algorithm's optimal design, biomarkers related to prostate cancer.

BACKGROUND: Nowadays modern biology aims at unravelling the strands of complex biological structures such as the protein-protein interaction (PPI) networks. A key concept in the organization of PPI networks is the existence of dense subnetworks (functional modules) in them. In recent approaches clustering algorithms were applied at these networks and the resulting subnetworks were evaluated by estimating the coverage of well-established protein complexes they contained. However, most of these algorithms elaborate on an unweighted graph structure which in turn fails to elevate those interactions that would contribute to the construction of biologically more valid and coherent functional modules. RESULTS: In the current study, we present a method that corroborates the integration of protein interaction and microarray data via the discovery of biologically valid functional modules. Initially the gene expression information is overlaid as weights onto the PPI network and the enriched PPI graph allows us to exploit its topological aspects, while simultaneously highlights enhanced functional association in specific pairs of proteins. Then we present an algorithm that unveils the functional modules of the weighted graph by expanding a kernel protein set, which originates from a given 'seed' protein used as starting-point. CONCLUSION: The integrated data and the concept of our approach provide reliable functional modules. We give proofs based on yeast data that our method manages to give accurate results in terms both of structural coherency, as well as functional consistency.

Synthetic aperture radar (SAR) images are inherently affected by multiplicative speckle noise, which is due to the coherent nature of the scattering phenomenon. This paper proposes a novel Bayesian-based algorithm within the framework of wavelet analysis, which reduces speckle in SAR images while preserving the structural features and textural information of the scene. First, we show that the subband decompositions of logarithmically transformed SAR images are accurately modeled by alpha-stable distributions, a family of heavy-tailed densities. Consequently, we exploit this a priori information by designing a maximum a posteriori (MAP) estimator. We use the alpha-stable model to develop a blind speckle-suppression processor that performs a nonlinear operation on the data and we relate this nonlinearity to the degree of non-Gaussianity of the data. Finally, we compare our proposed method to current state-of-the-art soft thresholding techniques applied on real SAR imagery and we quantify the achieved performance improvement.

MOTIVATION: Currently the most popular approach to analyze genome-wide expression data is clustering. One of the major drawbacks of most of the existing clustering methods is that the number of clusters has to be specified a priori. Furthermore, by using pure unsupervised algorithms prior biological knowledge is totally ignored Moreover, most current tools lack an effective framework for tight integration of unsupervised and supervised learning for the analysis of high-dimensional expression data and only very few multi-class supervised approaches are designed with the provision for effectively utilizing multiple functional class labeling. RESULTS: The paper adapts a novel Self-Organizing map called supervised Network Self-Organized Map (sNet-SOM) to the peculiarities of multi-labeled gene expression data. The sNet-SOM determines adaptively the number of clusters with a dynamic extension process. This process is driven by an inhomogeneous measure that tries to balance unsupervised, supervised and model complexity criteria. Nodes within a rectangular grid are grown at the boundary nodes, weights rippled from the internal nodes towards the outer nodes of the grid, and whole columns inserted within the map The appropriate level of expansion is determined automatically. Multiple sNet-SOM models are constructed dynamically each for a different unsupervised/supervised balance and model selection criteria are used to select the one optimum one. The results indicate that sNet-SOM yields competitive performance to other recently proposed approaches for supervised classification at a significantly reduced computational cost and it provides extensive exploratory analysis potentiality within the analysis framework. Furthermore, it explores simple design decisions that are easier to comprehend and computationally efficient.

Ultrasonic images are generally affected by multiplicative speckle noise, which is due to the coherent nature of the scattering phenomenon. Speckle filtering is thus a critical pre-processing step in medical ultrasound imagery, provided that the features of interest for diagnosis are not lost. We present a novel speckle removal algorithm within the framework of wavelet analysis. First, we show that the subband decompositions of logarithmically transformed ultrasound images are best described by alpha-stable distributions, a family of heavy-tailed densities. Consequently, we design a Bayesian estimator that exploits this a priori information. Using the alpha-stable model we develop a noise-removal processor that performs a nonlinear operation on the data. Finally, we compare our proposed technique to current state-of-the-art speckle reduction methods. Our algorithm effectively reduces speckle, it preserves step edges, and it enhances fine signal details, better than existing methods.

A novel speckle suppression method for medical ultrasound images is presented. First, the logarithmic transform of the original image is analyzed into the multiscale wavelet domain. We show that the subband decompositions of ultrasound images have significantly non-Gaussian statistics that are best described by families of heavy-tailed distributions such as the alpha-stable. Then, we design a Bayesian estimator that exploits these statistics. We use the alpha-stable model to develop a blind noise-removal processor that performs a nonlinear operation on the data. Finally, we compare our technique with current state-of-the-art soft and hard thresholding methods applied on actual ultrasound medical images and we quantify the achieved performance improvement.

The problem of maximizing the performance of the detection of ischemia episodes is a difficult pattern classification problem. The motivation for developing the supervising network self-organizing map (sNet-SOM) model is to exploit this fact for designing computationally effective solutions both for the particular ischemic detection problem and for other applications that share similar characteristics. Specifically, the sNet-SOM utilizes unsupervised learning for the "simple" regions and supervised for the "difficult" ones in a two stage learning process. The unsupervised learning approach extends and adapts the self-organizing map (SOM) algorithm of Kohonen. The basic SOM is modified with a dynamic expansion process controlled with an entropy based criterion that allows the adaptive formation of the proper SOM structure. This extension proceeds until the total number of training patterns that are mapped to neurons with high entropy reduces to a size manageable numerically with a capable supervised model. The second learning phase has the objective of constructing better decision boundaries at the ambiguous regions. At this phase, a special supervised network is trained for the computationally reduced task of performing the classification at the ambiguous regions only. The utilization of sNet-SOM with supervised learning based on the radial basis functions and support vector machines has resulted in an improved accuracy of ischemia detection especially in the last case. The highly disciplined design of the generalization performance of the support vector machine allows designing the proper model for the number of patterns transferred to the supervised expert.

The paper presents chaotic systems of difference equations that can effectively encrypt information. Two classes of systems are presented: The first one (Class 1) is optimized for secure communications over reliable channels, while the second (Class 2) tolerates transmission noise at the expense of reduced parameter space size. The nonlinearity of these systems is achieved by designing proper piecewise linear functions and by using module operations. The utilization of additional nonlinear terms can improve the enciphering efficiency. The encrypting performance of the algorithms is evaluated analytically and by simulation experiments. Also, the case of an imperfect transmission channel that inserts noise in the transmitted signal is addressed and the design is modified in order to offer reliable secure transmission over channels with very small signal to noise ratios.