Search

Background Older adults with diabetes are at high risk of severe hypoglycemia (SH). Many machine-learning (ML) models predict short-term hypoglycemia are not specific for older adults and show poor precision-recall. We aimed to develop a multidimensional, electronic health record (EHR)-based ML model to predict one-year risk of SH requiring hospitalization in older adults with diabetes. Methods and findings We adopted a case-control design for a retrospective territory-wide cohort of 1,456,618 records from 364,863 unique older adults (age [greater than or equal to]65 years) with diabetes and at least 1 Hong Kong Hospital Authority attendance from 2013 to 2018. We used 258 predictors including demographics, admissions, diagnoses, medications, and routine laboratory tests in a one-year period to predict SH events requiring hospitalization in the following 12 months. The cohort was randomly split into training, testing, and internal validation sets in a 7:2:1 ratio. Six ML algorithms were evaluated including logistic-regression, random forest, gradient boost machine, deep neural network (DNN), XGBoost, and Rulefit. We tested our model in a temporal validation cohort in the Hong Kong Diabetes Register with predictors defined in 2018 and outcome events defined in 2019. Predictive performance was assessed using area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC) statistics, and positive predictive value (PPV). We identified 11,128 SH events requiring hospitalization during the observation periods. The XGBoost model yielded the best performance (AUROC = 0.978 [95% CI 0.972 to 0.984]; AUPRC = 0.670 [95% CI 0.652 to 0.688]; PPV = 0.721 [95% CI 0.703 to 0.739]). This was superior to an 11-variable conventional logistic-regression model comprised of age, sex, history of SH, hypertension, blood glucose, kidney function measurements, and use of oral glucose-lowering drugs (GLDs) (AUROC = 0.906; AUPRC = 0.085; PPV = 0.468). Top impactful predictors included non-use of lipid-regulating drugs, in-patient admission, urgent emergency triage, insulin use, and history of SH. External validation in the HKDR cohort yielded AUROC of 0.856 [95% CI 0.838 to 0.873]. Main limitations of this study included limited transportability of the model and lack of geographically independent validation. Conclusions Our novel-ML model demonstrated good discrimination and high precision in predicting one-year risk of SH requiring hospitalization. This may be integrated into EHR decision support systems for preemptive intervention in older adults at highest risk., Author(s): Mai Shi 1,2, Aimin Yang 1,3, Eric S. H. Lau 1,2, Andrea O. Y. Luk 1,2,3, Ronald C. W. Ma 1,2,3, Alice P. S. Kong 1,2,3, Raymond S. M. [...]

Stock price prediction is a popular yet challenging task and deep learning provides the means to conduct the mining for the different patterns that trigger its dynamic movement. In this paper, the task is to predict the close price for 25 companies enlisted at the Bucharest Stock Exchange, from a novel data set introduced herein. Towards this scope, two traditional deep learning architectures are designed in comparison: a long short-memory network and a temporal convolutional neural model. Based on their predictions, a trading strategy, whose decision to buy or sell depends on two different thresholds, is proposed. A hill climbing approach selects the optimal values for these parameters. The prediction of the two deep learning representatives used in the subsequent trading strategy leads to distinct facets of gain. [ABSTRACT FROM AUTHOR]

To enrich any model and its dynamics introduction of delay is useful, that models a precise description of real-life phenomena. Differential equations in which current time derivatives count on the solution and its derivatives at a prior time are known as delay differential equations (DDEs). In this study, we are introducing new techniques for finding the numerical solution of fractional delay differential equations (FDDEs) based on the application of neural minimization (NM) by utilizing Chebyshev simulated annealing neural network (ChSANN) and Legendre simulated annealing neural network (LSANN). The main purpose of using Chebyshev and Legendre polynomials, along with simulated annealing (SA), is to reduce mean square error (MSE) that leads to more accurate numerical approximations. This study provides the application of ChSANN and LSANN for solving DDEs and FDDEs. Proposed schemes can be effortlessly executed by using Mathematica or MATLAB software to get explicit solutions. Computational outcomes are depicted, for various numerical experiments, numerically and graphically with error analysis to demonstrate the accuracy and efficiency of the methods. [ABSTRACT FROM AUTHOR]

Slope one is a popular recommendation algorithm due to its simplicity and high efficiency for sparse data. However, it often suffers from under-fitting since the global information of all relevant users/items are considered. In this paper, we propose a new scheme called enhanced slope one recommendation through local information embedding. First, we employ clustering algorithms to obtain the user clusters as well as item clusters to represent local information. Second, we predict ratings using the local information of users and items in the same cluster. The local information can detect strong localized associations shared within clusters. Third, we design different fusion approaches based on the local information embedding. In this way, both under-fitting and over-fitting problems are alleviated. Experiment results on the real datasets show that our approaches defeats slope one in terms of both mean absolute error and root mean square error. [ABSTRACT FROM AUTHOR]

In recent years, with the development of computer science, deep learning is held as competent enough to solve the problem of inference and learning in high dimensional space. Therefore, it has received unprecedented attention from both the academia and the business community. Compared with CPU/GPU, FPGA has attracted much attention for its high-energy efficiency, short development cycle and reconfigurability in the aspect of deep learning algorithm. However, because of the limited research on OpenCL optimization on FPGA of deep learning algorithms, OpenCL tools and models applied to CPU/GPU cannot be directly used on FPGA. This makes it difficult for software programmers to use FPGA when implementing deep learning algorithms for a rewarding performance. To solve this problem, this paper proposed an OpenCL computational model based on FPGA template architecture to optimize the time-consuming convolution layer in deep learning. The comparison between the program applying the computational model and the corresponding optimization program provided by Xilinx indicates that the former is 8-40 times higher than the latter in terms of performance. [ABSTRACT FROM AUTHOR]

Message passing between components of a distributed physical system is non-instantaneous and contributes to determine the time scales of the emerging collective dynamics. In biological neuron networks this is due in part to local synaptic filtering of exchanged spikes, and in part to the distribution of the axonal transmission delays. How differently these two kinds of communication protocols affect the network dynamics is still an open issue due to the difficulties in dealing with the non-Markovian nature of synaptic transmission. Here, we develop a mean-field dimensional reduction yielding to an effective Markovian dynamics of the population density of the neuronal membrane potential, valid under the hypothesis of small fluctuations of the synaptic current. Within this limit, the resulting theory allows us to prove the formal equivalence between the two transmission mechanisms, holding for any synaptic time scale, integrate-and-fire neuron model, spike emission regimes and for different network states even when the neuron number is finite. The equivalence holds even for larger fluctuations of the synaptic input, if white noise currents are incorporated to model other possible biological features such as ionic channel stochasticity. [ABSTRACT FROM AUTHOR]

One of the main technical challenges of PET/MRI is to achieve an accurate PET attenuation correction (AC) estimation. In current systems, AC is accomplished by generating an MRI-based surrogate computed tomography (CT) from which AC-maps are derived. Nevertheless, all techniques currently implemented in clinical routine suffer from bias. We present here a convolutional neural network (CNN) that generated AC-maps from Zero Echo Time (ZTE) MR images. Seventy patients referred to our institution for 18FDG-PET/MR exam (SIGNA PET/MR, GE Healthcare) as part of the investigation of suspected dementia, were included. 23 patients were added to the training set of the manufacturer and 47 were used for validation. Brain computed tomography (CT) scan, two-point LAVA-flex MRI (for atlas-based AC) and ZTE-MRI were available in all patients. Three AC methods were evaluated and compared to CT-based AC (CTAC): one based on a single head-atlas, one based on ZTE-segmentation and one CNN with a 3D U-net architecture to generate AC maps from ZTE MR images. Impact on brain metabolism was evaluated combining voxel and regions-of-interest based analyses with CTAC set as reference. The U-net AC method yielded the lowest bias, the lowest inter-individual and inter-regional variability compared to PET images reconstructed with ZTE and Atlas methods. The impact on brain metabolism was negligible with average errors of -0.2% in most cortical regions. These results suggest that the U-net AC is more reliable for correcting photon attenuation in brain FDG-PET/MR than atlas-AC and ZTE-AC methods. [ABSTRACT FROM AUTHOR]

Existing algorithms of speech-based deception detection are severely restricted by the lack of sufficient number of labelled data. However, a large amount of easily available unlabelled data has not been utilized in reality. To solve this problem, this paper proposes a semi-supervised additive noise autoencoder model for deception detection. This model updates and optimizes the semi-supervised autoencoder and it consists of two layers of encoder and decoder, and a classifier. Firstly, it changes the activation function of the hidden layer in network according to the characteristics of the deception speech. Secondly, in order to prevent over-fitting during training, the specific ratio dropout is added at each layer cautiously. Finally, we directly connected the supervised classification task in the output of encoder to make the network more concise and efficient. Using the feature set specified by the INTERSPEECH 2009 Emotion Challenge, the experimental results on Columbia-SRI-Colorado (CSC) corpus and our own deception corpus show that the proposed model can achieve more advanced performance than other alternative methods with only a small amount of labelled data. [ABSTRACT FROM AUTHOR]

To understand the effect of attention on neuronal dynamics, we propose a multi-module network, with each module consisting of fully interconnected groups of excitatory and inhibitory neurons. This network shows transitive dynamics among quasi-attractors as its typical dynamics. When the release of acetylcholine onto the network is simulated by attention, the transitive dynamics change into stable dynamics in which the system converges to an attractor. We found that this network can reproduce three experimentally observed properties of attention-dependent response modulation, namely an increase in the firing rate, a decrease in the Fano factor of the firing rate, and a decrease in the correlation coefficients between the firing rates of pairs of neurons. Moreover, we also showed theoretically that the release of acetylcholine increases the sensitivity to bottom-up inputs by changing the response function. [ABSTRACT FROM AUTHOR]

When probed with complex stimuli that extend beyond their classical receptive field, neurons in primary visual cortex display complex and non-linear response characteristics. Sparse coding models reproduce some of the observed contextual effects, but still fail to provide a satisfactory explanation in terms of realistic neural structures and cortical mechanisms, since the connection scheme they propose consists only of interactions among neurons with overlapping input fields. Here we propose an extended generative model for visual scenes that includes spatial dependencies among different features. We derive a neurophysiologically realistic inference scheme under the constraint that neurons have direct access only to local image information. The scheme can be interpreted as a network in primary visual cortex where two neural populations are organized in different layers within orientation hypercolumns that are connected by local, short-range and long-range recurrent interactions. When trained with natural images, the model predicts a connectivity structure linking neurons with similar orientation preferences matching the typical patterns found for long-ranging horizontal axons and feedback projections in visual cortex. Subjected to contextual stimuli typically used in empirical studies, our model replicates several hallmark effects of contextual processing and predicts characteristic differences for surround modulation between the two model populations. In summary, our model provides a novel framework for contextual processing in the visual system proposing a well-defined functional role for horizontal axons and feedback projections. [ABSTRACT FROM AUTHOR]

Background: Mild cognitive impairment is a common systemic manifestation of chronic obstructive pulmonary disease (COPD). However, its pathophysiological origins are not understood. Since, cognitive function relies on efficient communication between distributed cortical and subcortical regions, we investigated whether people with COPD have disruption in white matter connectivity. Methods: Structural networks were constructed for 30 COPD patients (aged 54–84 years, 57% male, FEV1 52.5% pred.) and 23 controls (aged 51–81 years, 48% Male). Networks comprised 90 grey matter regions (nodes) interconnected by white mater fibre tracts traced using deterministic tractography (edges). Edges were weighted by the number of streamlines adjusted for a) streamline length and b) end-node volume. White matter connectivity was quantified using global and nodal graph metrics which characterised the networks connection density, connection strength, segregation, integration, nodal influence and small-worldness. Between-group differences in white matter connectivity and within-group associations with cognitive function and disease severity were tested. Results: COPD patients’ brain networks had significantly lower global connection strength (p = 0.03) and connection density (p = 0.04). There was a trend towards COPD patients having a reduction in nodal connection density and connection strength across the majority of network nodes but this only reached significance for connection density in the right superior temporal gyrus (p = 0.02) and did not survive correction for end-node volume. There were no other significant global or nodal network differences or within-group associations with disease severity or cognitive function. Conclusion: COPD brain networks show evidence of damage compared to controls with a reduced number and strength of connections. This loss of connectivity was not sufficient to disrupt the overall efficiency of network organisation, suggesting that it has redundant capacity that makes it resilient to damage, which may explain why cognitive dysfunction is not severe. This might also explain why no direct relationships could be found with cognitive measures. Smoking and hypertension are known to have deleterious effects on the brain. These confounding effects could not be excluded. [ABSTRACT FROM AUTHOR]

Most cetacean species are wide-ranging and highly mobile, creating significant challenges for researchers by limiting the scope of data that can be collected and leaving large areas un-surveyed. Aerial surveys have proven an effective way to locate and study cetacean movements but are costly and limited in spatial extent. Here we present a semi-automated pipeline for whale detection from very high-resolution (sub-meter) satellite imagery that makes use of a convolutional neural network (CNN). We trained ResNet, and DenseNet CNNs using down-scaled aerial imagery and tested each model on 31 cm-resolution imagery obtained from the WorldView-3 sensor. Satellite imagery was tiled and the trained algorithms were used to classify whether or not a tile was likely to contain a whale. Our best model correctly classified 100% of tiles with whales, and 94% of tiles containing only water. All model architectures performed well, with learning rate controlling performance more than architecture. While the resolution of commercially-available satellite imagery continues to make whale identification a challenging problem, our approach provides the means to efficiently eliminate areas without whales and, in doing so, greatly accelerates ocean surveys for large cetaceans. [ABSTRACT FROM AUTHOR]

With the rapid development of the internet, social media has become an essential tool for getting information, and attracted a large number of people join the social media platforms because of its low cost, accessibility and amazing content. It greatly enriches our life. However, its rapid development and widespread also have provided an excellent convenience for the range of fake news, people are constantly exposed to fake news and suffer from it all the time. Fake news usually uses hyperbole to catch people’s eyes with dishonest intention. More importantly, it often misleads the reader and causes people to have wrong perceptions of society. It has the potential for negative impacts on society and individuals. Therefore, it is significative research on detecting fake news. In the paper, we built a model named SMHA-CNN (Self Multi-Head Attention-based Convolutional Neural Networks) that can judge the authenticity of news with high accuracy based only on content by using convolutional neural networks and self multi-head attention mechanism. In order to prove its validity, we conducted experiments on a public dataset and achieved a precision rate of 95.5% with a recall rate of 95.6% under the 5-fold cross-validation. Our experimental result indicates that the model is more effective at detecting fake news. [ABSTRACT FROM AUTHOR]

Objective: The Centers for Disease Control and Prevention (CDC) coordinates a labor-intensive process to measure the prevalence of autism spectrum disorder (ASD) among children in the United States. Random forests methods have shown promise in speeding up this process, but they lag behind human classification accuracy by about 5%. We explore whether more recently available document classification algorithms can close this gap. Materials and methods: Using data gathered from a single surveillance site, we applied 8 supervised learning algorithms to predict whether children meet the case definition for ASD based solely on the words in their evaluations. We compared the algorithms’ performance across 10 random train-test splits of the data, using classification accuracy, F1 score, and number of positive calls to evaluate their potential use for surveillance. Results: Across the 10 train-test cycles, the random forest and support vector machine with Naive Bayes features (NB-SVM) each achieved slightly more than 87% mean accuracy. The NB-SVM produced significantly more false negatives than false positives (P = 0.027), but the random forest did not, making its prevalence estimates very close to the true prevalence in the data. The best-performing neural network performed similarly to the random forest on both measures. Discussion: The random forest performed as well as more recently available models like the NB-SVM and the neural network, and it also produced good prevalence estimates. NB-SVM may not be a good candidate for use in a fully-automated surveillance workflow due to increased false negatives. More sophisticated algorithms, like hierarchical convolutional neural networks, may not be feasible to train due to characteristics of the data. Current algorithms might perform better if the data are abstracted and processed differently and if they take into account information about the children in addition to their evaluations. Conclusion: Deep learning models performed similarly to traditional machine learning methods at predicting the clinician-assigned case status for CDC’s autism surveillance system. While deep learning methods had limited benefit in this task, they may have applications in other surveillance systems. [ABSTRACT FROM AUTHOR]

Background and aim: Depression is often accompanied by thoughts of self-harm, which are a strong predictor of subsequent suicide attempt and suicide death. Few empirical data are available regarding the temporal correlation between depression symptoms and suicidal ideation. We investigated the anecdotal concern that suicidal ideation may increase during a period of depression improvement. Data: Longitudinal Patient Health Questionnaire (PHQ)-9 is a questionnaire of 9 multiple-choice questions to assess the frequency of depressive symptoms within the previous two weeks. We analyzed a chronic depression treatment population’s electronic health record (EHR) data, containing 610 patients’ longitudinal PHQ-9 scores (62% age 45 and older; 68% female) within 40 weeks. Methods: The irregular and sparse EHR data were transformed into continuous trajectories using Gaussian process regression. We first estimated the correlations between the symptoms (total score of the first 8 questions; PHQ-8) and suicide ideation (9th question score; Item 9) using the cross-correlation function. We then used an artificial neural network (ANN) to discover subtypes of depression patterns from the fitted depression trajectories. In addition, we conducted a separate analysis using the unfitted raw PHQ scores to examine PHQ-8’s and Item 9’s pattern changes. Results: Results showed that the majority of patients’ PHQ-8 and Item 9 scores displayed strong temporal correlations. We found five patterns in the PHQ-8 and the Item 9 trajectories. We also found 8% - 13% of the patients have experienced an increase in suicidal ideation during the improvement of their PHQ-8. Using a trajectory-based method for subtype pattern detection in depression progression, we provided a better understanding of temporal correlations between depression symptoms over time. [ABSTRACT FROM AUTHOR]

The ventral midbrain supports a variety of functions through the heterogeneity of neurons. Dopaminergic and GABA neurons within this region are particularly susceptible targets of amphetamine-class psychostimulants such as methamphetamine. While this has been evidenced through single-neuron methods, it remains unclear whether and to what extent the local neuronal network is affected and if so, by which mechanisms. Both GABAergic and dopaminergic neurons were heavily featured within the primary ventral midbrain network model system. Using spontaneous calcium activity, our data suggest methamphetamine decreased total network output via a D2 receptor-dependent manner. Over culture duration, functional connectivity between neurons decreased significantly but was unaffected by methamphetamine. However, across culture duration, exposure to methamphetamine significantly altered changes in network assortativity. Here we have established primary ventral midbrain networks culture as a viable model system that reveals specific changes in network activity, connectivity, and topology modulation by methamphetamine. This network culture system enables control over the type and number of neurons that comprise a network and facilitates detection of emergent properties that arise from the specific organization. Thus, the multidimensional properties of methamphetamine can be unraveled, leading to a better understanding of its impact on the local network structure and function. [ABSTRACT FROM AUTHOR]

Using fMRI, Hugdahl et al. (2015) reported the existence of a general-domain cortical network during active task-processing which was non-specific to the cognitive task being processed. They labelled this network the extrinsic mode network (EMN). The EMN would be predicted to be negatively, or anti-correlated with the classic default mode network (DMN), typically observed during periods of rest, such that while the EMN should be down-regulated and the DMN up-regulated in the absence of demands for task-processing, the reverse should occur when demands change from resting to task-processing. This would require alternating periods of task-processing and resting and analyzing data continuously when demands change from active to passive periods and vice versa. We were particularly interested in how the networks interact in the critical transition points between conditions. For this purpose, we used an auditory task with multiple cognitive demands in a standard fMRI block-design. Task-present (ON) blocks were alternated with an equal number of task-absent, or rest (OFF) blocks to capture network dynamics across time and changing environmental demands. To achieve this, we specified the onset of each block, and used a finite-impulse response function (FIR) as basis function for estimation of the fMRI-BOLD response. During active (ON) blocks, the results showed an initial rapid onset of activity in the EMN network, which remained throughout the period, and faded away during the first scan of the OFF-block. During OFF blocks, activity in the DMN network showed an initial time-lag where neither the EMN nor the DMN was active, after which the DMN was up-regulated. Studying network dynamics in alternating passive and active periods may provide new insights into brain network interaction and regulation. [ABSTRACT FROM AUTHOR]

A fundamental challenge in neuroscience is to understand what structure in the world is represented in spatially distributed patterns of neural activity from multiple single-trial measurements. This is often accomplished by learning a simple, linear transformations between neural features and features of the sensory stimuli or motor task. While successful in some early sensory processing areas, linear mappings are unlikely to be ideal tools for elucidating nonlinear, hierarchical representations of higher-order brain areas during complex tasks, such as the production of speech by humans. Here, we apply deep networks to predict produced speech syllables from a dataset of high gamma cortical surface electric potentials recorded from human sensorimotor cortex. We find that deep networks had higher decoding prediction accuracy compared to baseline models. Having established that deep networks extract more task relevant information from neural data sets relative to linear models (i.e., higher predictive accuracy), we next sought to demonstrate their utility as a data analysis tool for neuroscience. We first show that deep network’s confusions revealed hierarchical latent structure in the neural data, which recapitulated the underlying articulatory nature of speech motor control. We next broadened the frequency features beyond high-gamma and identified a novel high-gamma-to-beta coupling during speech production. Finally, we used deep networks to compare task-relevant information in different neural frequency bands, and found that the high-gamma band contains the vast majority of information relevant for the speech prediction task, with little-to-no additional contribution from lower-frequency amplitudes. Together, these results demonstrate the utility of deep networks as a data analysis tool for basic and applied neuroscience. [ABSTRACT FROM AUTHOR]

Solving the supply–demand imbalance is the most crucial issue for stable implementation of a public bike-sharing system. This gap can be reduced by increasing the accuracy of demand prediction by considering spatial and temporal properties of bike demand. However, only a few attempts have been made to account for both features simultaneously. Therefore, we propose a prediction framework based on graph convolutional networks. Our framework reflects not only spatial dependencies among stations, but also various temporal patterns over different periods. Additionally, we consider the influence of global variables, such as weather and weekday/weekend to reflect non-station-level changes. We compare our framework to other baseline models using the data from Seoul’s bike-sharing system. Results show that our approach has better performance than existing prediction models. [ABSTRACT FROM AUTHOR]

A variety of simple models has been proposed to understand the collective motion of animals. These models can be insightful but may lack important elements necessary to predict the motion of each individual in the collective. Adding more detail increases predictability but can make models too complex to be insightful. Here we report that deep attention networks can obtain a model of collective behavior that is simultaneously predictive and insightful thanks to an organization in modules. When using simulated trajectories, the model recovers the ground-truth interaction rule used to generate them, as well as the number of interacting neighbours. For experimental trajectories of large groups of 60-100 zebrafish, Danio rerio, the model obtains that interactions between pairs can approximately be described as repulsive, attractive or as alignment, but only when moving slowly. At high velocities, interactions correspond only to alignment or alignment mixed with repulsion at close distances. The model also shows that each zebrafish decides where to move by aggregating information from the group as a weighted average over neighbours. Weights are higher for neighbours that are close, in a collision path or moving faster in frontal and lateral locations. The network also extracts that the number of interacting individuals is dynamical and typically in the range 8–22, with 1–10 more important ones. Our results suggest that each animal decides by dynamically selecting information from the collective. [ABSTRACT FROM AUTHOR]

The present study aimed to develop a denoising convolutional neural network metal artifact reduction hybrid reconstruction (DnCNN-MARHR) algorithm for decreasing metal objects in digital tomosynthesis (DT) for arthroplasty by using projection data. For metal artifact reduction (MAR), we implemented a DnCNN-MARHR algorithm based on a training network (mini-batch stochastic gradient descent algorithm with momentum) to estimate the residual reference (140 keV virtual monochromatic [VM]) and object (70 kV with metal artifacts) images. For this, we used projection data and subtracted the estimated residual images from the object images, involving hybrid and subjectively reconstructed image usage (back projection and maximum likelihood expectation maximization [MLEM]). The DnCNN-MARHR algorithm was compared with the dual-energy material decomposition reconstruction algorithm (DEMDRA), VM, MLEM, established and commonly used filtered back projection (FBP), and a simultaneous algebraic reconstruction technique-total variation (SART-TV) with MAR processing. MAR was compared using artifact index (AI) and texture analysis. Artifact spread functions (ASFs) for images that were out-of-plane and in-focus were evaluated using a prosthesis phantom. The overall performance of the DnCNN-MARHR algorithm was adequate with regard to the ASF, and the derived images showed better results, without being influenced by the metal type (AI was almost equal to the best value for the DEMDRA). In the ASF analysis, the DnCNN-MARHR algorithm generated better MAR compared with that obtained employing usual algorithms for reconstruction using MAR processing. In addition, comparison of the difference (mean square error) between DnCNN-MARHR and the conventional algorithm resulted in the smallest VM. The DnCNN-MARHR algorithm showed the best performance with regard to image homogeneity in the texture analysis. The proposed algorithm is particularly useful for reducing artifacts in the longitudinal direction, and it is not affected by tissue misclassification. [ABSTRACT FROM AUTHOR]

Neural correlates of mind wandering: The ability to detect mind wandering as it occurs is an important step towards improving our understanding of this phenomenon and studying its effects on learning and performance. Current detection methods typically rely on observable behaviour in laboratory settings, which do not capture the underlying neural processes and may not translate well into real-world settings. We address both of these issues by recording electroencephalography (EEG) simultaneously from 15 participants during live lectures on research in orthopedic surgery. We performed traditional group-level analysis and found neural correlates of mind wandering during live lectures that are similar to those found in some laboratory studies, including a decrease in occipitoparietal alpha power and frontal, temporal, and occipital beta power. However, individual-level analysis of these same data revealed that patterns of brain activity associated with mind wandering were more broadly distributed and highly individualized than revealed in the group-level analysis. Mind wandering detection: To apply these findings to mind wandering detection, we used a data-driven method known as common spatial patterns to discover scalp topologies for each individual that reflects their differences in brain activity when mind wandering versus attending to lectures. This approach avoids reliance on known neural correlates primarily established through group-level statistics. Using this method for individual-level machine learning of mind wandering from EEG, we were able to achieve an average detection accuracy of 80–83%. Conclusions: Modelling mind wandering at the individual level may reveal important details about its neural correlates that are not reflected when using traditional observational and statistical methods. Using machine learning techniques for this purpose can provide new insight into the varieties of neural activity involved in mind wandering, while also enabling real-time detection of mind wandering in naturalistic settings. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) tools have been applied to diagnose or predict disease risk from medical images with recent data disclosure actions, but few of them are designed for mobile terminals due to the limited computational power and storage capacity of mobile devices. In this work, a novel AI diagnostic system is proposed for cholelithiasis recognition on mobile devices with Android platform. To this aim, a data set of CT images of cholelithiasis is firstly collected from The Third Hospital of Shandong Province, China, and then we technically use histogram equalization to preprocess these CT images. As results, a lightweight convolutional neural network is obtained in a constructive way to extract cholelith features and recognize gallstones. In terms of implementation, we compile Java and C++ to adapt to the application of deep learning algorithm on mobile devices with Android platform. Noted that, the training task is completed offline on PC, but cholelithiasis recognition tasks are performed on mobile terminals. We evaluate and compare the performance of our MobileNetV2 with MobileNetV1, Single Shot Detector (SSD), YOLOv2 and original SSD (with VGG-16) as feature extractors for object detection. It is achieved that our MobileNetV2 achieve similar accuracy rate, about 91% with the other four methods, but the number of parameters used is reduced from 36.1M (SSD 300, SSD512), 50.7M (Yolov2) and 5.1M (MobileNetV1) to 4.3M (MobileNetV2). The complete process on testing mobile devices, including Virtual machine, Xiaomi 7 and Htc One M8 can be controlled within 4 seconds in recognizing cholelithiasis as well as the degree of the disease. [ABSTRACT FROM AUTHOR]

The hierarchical modular organization of functional networks in the brain is crucial for supporting diverse cognitive functions. Functional disorders in the brain are associated with an abnormal hierarchical modular organization. The default mode network (DMN) is a complex dynamic network that is linked to specialized cognitive functions and clinically relevant information. In this study, we hypothesize that hierarchical functional segregation and integration of the DMN within attention-deficit/hyperactivity disorder (ADHD) is abnormal. We investigated topological metrics of both segregation and integration in different hierarchical subnetworks of the DMN between patients with ADHD and healthy controls. We found that the hierarchical functional integration and segregation of the DMN decreased and increased, respectively, in ADHD. Our results also indicated that the abnormalities in the DMN are intrinsically caused by changes in functional segregation and integration in its higher-level subnetworks. To better understand the temporally dynamic changes in the hierarchical functional integration and segregation of the DMN within ADHD, we further analyzed the dynamic transitions between functional segregation and integration. We found that the adaptive reorganizational ability of brain network states decreased in ADHD patients, which indicated less adaptive regulation between the DMN subnetworks in ADHD for supporting correspondingly normal cognitive function. From the perspective of hierarchical functional segregation and integration, our results further provide evidence to support dysfunctional brain cognitive functions within ADHD linked to brain network segregation and integration. [ABSTRACT FROM AUTHOR]

Sensor-based human activity recognition aims at detecting various physical activities performed by people with ubiquitous sensors. Different from existing deep learning-based method which mainly extracting black-box features from the raw sensor data, we propose a hierarchical multi-view aggregation network based on multi-view feature spaces. Specifically, we first construct various views of feature spaces for each individual sensor in terms of white-box features and black-box features. Then our model learns a unified representation for multi-view features by aggregating views in a hierarchical context from the aspect of feature level, position level and modality level. We design three aggregation modules corresponding to each level aggregation respectively. Based on the idea of non-local operation and attention, our fusion method is able to capture the correlation between features and leverage the relationship across different sensor position and modality. We comprehensively evaluate our method on 12 human activity benchmark datasets and the resulting accuracy outperforms the state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Quantitative analysis of the contribution rate of China’s hydraulic science and technology and analysis of the underlying reasons behind changes provide an important foundation upon which the government can formulate water policies. This paper abandons the assumption of a scale economy and separates the changes of benefits brought about by the scale from scientific and technological progress, thus changing the C-D production function from linear to nonlinear. Based on a feedforward neural network, it calculates the coefficient of the output elasticity, the economic contribution rate of China’s hydraulic science and technology and the scale economies for each year using relevant data from 1981 to 2016. The results show that (1) the average contribution rate of capital investment from 1981 to 2016 was 47.3%, and the average contribution rate of labor from 1981 to 2016 was 9.1%. It is not obvious that the significant increase in the labor force has contributed to the growth of China’s water conservancy industry. (2) The average contribution rate of scale economies in 1981–2016 was 26.7%, and the contribution rate of scale economies is negatively correlated with the capital contribution rate. (3) The average contribution rate of China’s hydraulic science and technology was 43.6% from 1981 to 2016, and the average contribution rate of the total factor productivity after removing scale economies from 1981 to 2016 was 16.9%. During the period of the 6th Five-Year Plan(1981~1985), the contribution rate of water conservancy science and technology was relatively high. Since that time, it has remained at 40%. In recent years, as water conservancy reforms in key areas have made positive progress, scientific and technological progress has increased the growth of water conservancy benefits annually. [ABSTRACT FROM AUTHOR]

Despite its clinical importance, detection of highly divergent or yet unknown viruses is a major challenge. When human samples are sequenced, conventional alignments classify many assembled contigs as “unknown” since many of the sequences are not similar to known genomes. In this work, we developed ViraMiner, a deep learning-based method to identify viruses in various human biospecimens. ViraMiner contains two branches of Convolutional Neural Networks designed to detect both patterns and pattern-frequencies on raw metagenomics contigs. The training dataset included sequences obtained from 19 metagenomic experiments which were analyzed and labeled by BLAST. The model achieves significantly improved accuracy compared to other machine learning methods for viral genome classification. Using 300 bp contigs ViraMiner achieves 0.923 area under the ROC curve. To our knowledge, this is the first machine learning methodology that can detect the presence of viral sequences among raw metagenomic contigs from diverse human samples. We suggest that the proposed model captures different types of information of genome composition, and can be used as a recommendation system to further investigate sequences labeled as “unknown” by conventional alignment methods. Exploring these highly-divergent viruses, in turn, can enhance our knowledge of infectious causes of diseases. [ABSTRACT FROM AUTHOR]

Short-term metro passenger flow forecasting is an essential component of intelligent transportation systems (ITS) and can be applied to optimize the passenger flow organization of a station and offer data support for metro passenger flow early warning and system management. LSTM neural networks have recently achieved remarkable recent in the field of natural language processing (NLP) because they are well suited for learning from experience to predict time series. For this purpose, we propose an empirical mode decomposition (EMD)-based long short-term memory (LSTM) neural network model for predicting short-term metro inbound passenger flow. The EMD algorithm decomposes the original sequential passenger flow into several intrinsic mode functions (IMFs) and a residual. Selected IMFs that are strongly correlated with the original data can be obtained via feature selection. The selected IMFs and the original data are integrated into inputs for LSTM neural networks, and a single LSTM prediction model and an EMD-LSTM hybrid forecasting model are developed. Finally, historical real automatic fare collection (AFC) data from metro passengers are collected from Chengdu Metro to verify the validity of the proposed EMD-LSTM prediction model. The results indicate that the proposed EMD-LSTM hybrid forecasting model outperforms the LSTM, ARIMA and BPN models. [ABSTRACT FROM AUTHOR]

The resting-state brain is often considered a nonlinear dynamic system transitioning among multiple coexisting stable states. Despite the increasing number of studies on the multistability of the brain system, the processes of state transitions have rarely been systematically explored. Thus, we investigated the state transition processes of the human cerebral cortex system at rest by introducing a graph-theoretical analysis of the state transition network. The energy landscape analysis of brain state occurrences, estimated using the pairwise maximum entropy model for resting-state fMRI data, identified multiple local minima, some of which mediate multi-step transitions toward the global minimum. The state transition among local minima is clustered into two groups according to state transition rates and most inter-group state transitions were mediated by a hub transition state. The distance to the hub transition state determined the path length of the inter-group transition. The cortical system appeared to have redundancy in inter-group transitions when the hub transition state was removed. Such a hub-like organization of transition processes disappeared when the connectivity of the cortical system was altered from the resting-state configuration. In the state transition, the default mode network acts as a transition hub, while coactivation of the prefrontal cortex and default mode network is captured as the global minimum. In summary, the resting-state cerebral cortex has a well-organized architecture of state transitions among stable states, when evaluated by a graph-theoretical analysis of the nonlinear state transition network of the brain. [ABSTRACT FROM AUTHOR]

Due to the fast speed of data generation and collection from advanced equipment, the amount of data obviously overflows the limit of available memory space and causes difficulties achieving high learning accuracy. Several methods based on discard-after-learn concept have been proposed. Some methods were designed to cope with a single incoming datum but some were designed for a chunk of incoming data. Although the results of these approaches are rather impressive, most of them are based on temporally adding more neurons to learn new incoming data without any neuron merging process which can obviously increase the computational time and space complexities. Only online versatile elliptic basis function (VEBF) introduced neuron merging to reduce the space-time complexity of learning only a single incoming datum. This paper proposed a method for further enhancing the capability of discard-after-learn concept for streaming data-chunk environment in terms of low computational time and neural space complexities. A set of recursive functions for computing the relevant parameters of a new neuron, based on statistical confidence interval, was introduced. The newly proposed method, named streaming chunk incremental learning (SCIL), increases the plasticity and the adaptabilty of the network structure according to the distribution of incoming data and their classes. When being compared to the others in incremental-like manner, based on 11 benchmarked data sets of 150 to 581,012 samples with attributes ranging from 4 to 1,558 formed as streaming data, the proposed SCIL gave better accuracy and time in most data sets. [ABSTRACT FROM AUTHOR]

We present a novel approach based on deep learning for decoding sensory information from non-invasively recorded Electroencephalograms (EEG). It can either be used in a passive Brain-Computer Interface (BCI) to predict properties of a visual stimulus the person is viewing, or it can be used to actively control a BCI application. Both scenarios were tested, whereby an average information transfer rate (ITR) of 701 bit/min was achieved for the passive BCI approach with the best subject achieving an online ITR of 1237 bit/min. Further, it allowed the discrimination of 500,000 different visual stimuli based on only 2 seconds of EEG data with an accuracy of up to 100%. When using the method for an asynchronous self-paced BCI for spelling, an average utility rate of 175 bit/min was achieved, which corresponds to an average of 35 error-free letters per minute. As the presented method extracts more than three times more information than the previously fastest approach, we suggest that EEG signals carry more information than generally assumed. Finally, we observed a ceiling effect such that information content in the EEG exceeds that required for BCI control, and therefore we discuss if BCI research has reached a point where the performance of non-invasive visual BCI control cannot be substantially improved anymore. [ABSTRACT FROM AUTHOR]

In protein tertiary structure prediction, model quality assessment programs (MQAPs) are often used to select the final structural models from a pool of candidate models generated by multiple templates and prediction methods. The 3-dimensional convolutional neural network (3DCNN) is an expansion of the 2DCNN and has been applied in several fields, including object recognition. The 3DCNN is also used for MQA tasks, but the performance is low due to several technical limitations related to protein tertiary structures, such as orientation alignment. We proposed a novel single-model MQA method based on local structure quality evaluation using a deep neural network containing 3DCNN layers. The proposed method first assesses the quality of local structures for each residue and then evaluates the quality of whole structures by integrating estimated local qualities. We analyzed the model using the CASP11, CASP12, and 3D-Robot datasets and compared the performance of the model with that of the previous 3DCNN method based on whole protein structures. The proposed method showed a significant improvement compared to the previous 3DCNN method for multiple evaluation measures. We also compared the proposed method to other state-of-the-art methods. Our method showed better performance than the previous 3DCNN-based method and comparable accuracy as the current best single-model methods; particularly, in CASP11 stage2, our method showed a Pearson coefficient of 0.486, which was better than those of the best single-model methods (0.366–0.405). A standalone version of the proposed method and data files are available at . [ABSTRACT FROM AUTHOR]

Predicting RNA-binding protein (RBP) specificity is important for understanding gene expression regulation and RNA-mediated enzymatic processes. It is widely believed that RBP binding specificity is determined by both the sequence and structural contexts of RNAs. Existing approaches, including traditional machine learning algorithms and more recently, deep learning models, have been extensively applied to integrate RNA sequence and its predicted or experimental RNA structural probabilities for improving the accuracy of RBP binding prediction. Such models were trained mostly on the large-scale in vitro datasets, such as the RNAcompete dataset. However, in RNAcompete, most synthetic RNAs are unstructured, which makes machine learning methods not effectively extract RBP-binding structural preferences. Furthermore, RNA structure may be variable or multi-modal according to both theoretical and experimental evidence. In this work, we propose ThermoNet, a thermodynamic prediction model by integrating a new sequence-embedding convolutional neural network model over a thermodynamic ensemble of RNA secondary structures. First, the sequence-embedding convolutional neural network generalizes the existing k-mer based methods by jointly learning convolutional filters and k-mer embeddings to represent RNA sequence contexts. Second, the thermodynamic average of deep-learning predictions is able to explore structural variability and improves the prediction, especially for the structured RNAs. Extensive experiments demonstrate that our method significantly outperforms existing approaches, including RCK, DeepBind and several other recent state-of-the-art methods for predictions on both in vitro and in vivo data. The implementation of ThermoNet is available at . [ABSTRACT FROM AUTHOR]

Across diverse biological systems—ranging from neural networks to intracellular signaling and genetic regulatory networks—the information about changes in the environment is frequently encoded in the full temporal dynamics of the network nodes. A pressing data-analysis challenge has thus been to efficiently estimate the amount of information that these dynamics convey from experimental data. Here we develop and evaluate decoding-based estimation methods to lower bound the mutual information about a finite set of inputs, encoded in single-cell high-dimensional time series data. For biological reaction networks governed by the chemical Master equation, we derive model-based information approximations and analytical upper bounds, against which we benchmark our proposed model-free decoding estimators. In contrast to the frequently-used k-nearest-neighbor estimator, decoding-based estimators robustly extract a large fraction of the available information from high-dimensional trajectories with a realistic number of data samples. We apply these estimators to previously published data on Erk and Ca2+ signaling in mammalian cells and to yeast stress-response, and find that substantial amount of information about environmental state can be encoded by non-trivial response statistics even in stationary signals. We argue that these single-cell, decoding-based information estimates, rather than the commonly-used tests for significant differences between selected population response statistics, provide a proper and unbiased measure for the performance of biological signaling networks. [ABSTRACT FROM AUTHOR]

Past research on the process of extinguishing a fire typically used a traditional linear water jet falling point model and the results ignored external factors, such as environmental conditions and the status of the fire engine, even though the water jet falling point location prediction was often associated with these parameters and showed a nonlinear relationship. This paper constructed a BP (Back Propagation) neural network model. The fire gun nozzle characteristics were included as model inputs, and the water discharge point coordinates were the model outputs; thus, the model could precisely predict the water discharge point with small error and high precision to determine an accurate firing position and allow for the timely adjustment of the spray gun. To improve the slow convergence and local optimality problems of the BP neural network (BPNN), this paper further used a genetic algorithm to optimize the BPNN (GA-BPNN). The BPNN can be used to optimize the weights in the network to train them for global optimization. A genetic algorithm was introduced into the neural network approach, and the water jet landing prediction model was further improved. The simulation results showed that the prediction accuracy of the GA-BP model was better than that of the BPNN alone. The established model can accurately predict the location of the water jet, making the prediction results more useful for firefighters. [ABSTRACT FROM AUTHOR]

Parkinson’s disease (PD) affects 2–3% of the population over the age of 65 with loss of dopaminergic neurons in the substantia nigra impacting the functioning of basal ganglia-thalamocortical circuits. The precise role played by the thalamus is unknown, despite its critical role in the functioning of the cerebral cortex, and the abnormal neuronal activity of the structure in PD. Our objective was to more clearly elucidate how functional connectivity and morphology of the thalamus are impacted in PD (n = 32) compared to Controls (n = 20). To investigate functional connectivity of the thalamus we subdivided the structure into two important regions-of-interest, the first with putative connections to the motor cortices and the second with putative connections to prefrontal cortices. We then investigated potential differences in the size and shape of the thalamus in PD, and how morphology and functional connectivity relate to clinical variables. Our data demonstrate that PD is associated with increases in functional connectivity between motor subdivisions of the thalamus and the supplementary motor area, and between prefrontal thalamic subdivisions and nuclei of the basal ganglia, anterior and dorsolateral prefrontal cortices, as well as the anterior and paracingulate gyri. These results suggest that PD is associated with increased functional connectivity of subdivisions of the thalamus which may be indicative alterations to basal ganglia-thalamocortical circuitry. [ABSTRACT FROM AUTHOR]

Lately, discourse structure has received considerable attention due to the benefits its application offers in several NLP tasks such as opinion mining, summarization, question answering, text simplification, among others. When automatically analyzing texts, discourse parsers typically perform two different tasks: i) identification of basic discourse units (text segmentation) ii) linking discourse units by means of discourse relations, building structures such as trees or graphs. The resulting discourse structures are, in general terms, accurate at intra-sentence discourse-level relations, however they fail to capture the correct inter-sentence relations. Detecting the main discourse unit (the Central Unit) is helpful for discourse analyzers (and also for manual annotation) in improving their results in rhetorical labeling. Bearing this in mind, we set out to build the first two steps of a discourse parser following a top-down strategy: i) to find discourse units, ii) to detect the Central Unit. The final step, i.e. assigning rhetorical relations, remains to be worked on in the immediate future. In accordance with this strategy, our paper presents a tool consisting of a discourse segmenter and an automatic Central Unit detector. [ABSTRACT FROM AUTHOR]

The mechanism(s) of action of most commonly used pharmacological blockers of voltage-gated ion channels are well understood; however, this knowledge is rarely considered when interpreting experimental data. Effects of blockade are often assumed to be equivalent, regardless of the mechanism of the blocker involved. Using computer simulations, we demonstrate that this assumption may not always be correct. We simulate the blockade of a persistent sodium current (INaP), proposed to underlie rhythm generation in pre-Bötzinger complex (pre-BötC) respiratory neurons, via two distinct pharmacological mechanisms: (1) pore obstruction mediated by tetrodotoxin and (2) altered inactivation dynamics mediated by riluzole. The reported effects of experimental application of tetrodotoxin and riluzole in respiratory circuits are diverse and seemingly contradictory and have led to considerable debate within the field as to the specific role of INaP in respiratory circuits. The results of our simulations match a wide array of experimental data spanning from the level of isolated pre-BötC neurons to the level of the intact respiratory network and also generate a series of experimentally testable predictions. Specifically, in this study we: (1) provide a mechanistic explanation for seemingly contradictory experimental results from in vitro studies of INaP block, (2) show that the effects of INaP block in in vitro preparations are not necessarily equivalent to those in more intact preparations, (3) demonstrate and explain why riluzole application may fail to effectively block INaP in the intact respiratory network, and (4) derive the prediction that effective block of INaP by low concentration tetrodotoxin will stop respiratory rhythm generation in the intact respiratory network. These simulations support a critical role for INaP in respiratory rhythmogenesis in vivo and illustrate the importance of considering mechanism when interpreting and simulating data relating to pharmacological blockade. [ABSTRACT FROM AUTHOR]

The neural computational model GrowthEstimate is introduced with focusing on new perspectives for the practical estimation of weight specific growth rate (SGR, % day–1). It is developed using recurrent neural networks of reservoir computing type, for estimating SGR based on the known data of three key biological factors relating to growth. These factors are: (1) weight (g) for specifying the age of the growth stage; (2) digestive efficiency through the pyloric caecal activity ratio of trypsin to chymotrypsin (T/C ratio) for specifying genetic differences in food utilization and growth potential, basically resulting from food consumption under variations in food quality and environmental conditions; and (3) protein growth efficiency through the condition factor (CF, 100 × g cm–3), as higher dietary protein level affecting higher skeletal growth (length) and resulting in lower CF. The computational model was trained using four datasets of different salmonids with size variations. It was evaluated with 15% of each dataset, resulting in an acceptable range of SGR outputs. Additional tests with different species indicated similarity between the estimated SGR outputs and the real SGR values, and the same ranking of wild population growth. The developed model GrowthEstimate is exceptionally useful for the precise and comparable growth estimation of living resources at individual levels, especially in natural ecosystems where the studied individuals, environmental conditions, food availability and consumption rates cannot be controlled. It is a revelation and will help to minimize uncertainty in wild stock assessment process. This will improve our knowledge in nutritional ecology, through the biochemical effects of climate change and environmental impact on the growth performance quality of aquatic living resources in the wild, as well as in aquaculture. The original GrowthEstimate software is available at GitHub repository (). All other relevant data are within the paper. It will be improved for generality for future use, and required co-operations of the biodata collections of different species from different climate zones. Therefore, a co-operation will be available. [ABSTRACT FROM AUTHOR]

Many neuroimaging and electrophysiology studies have suggested that semantic integration as a high-level cognitive process involves various cortical regions and is modulated by attention. However, the cortical network specific to semantic integration and the modulatory mechanism of attention remain unclear. Here, we designed an fMRI experiment using “bimodal stimulus” to extract information regarding the cortical activation related to the effects of semantic integration with and without attention, and then analyzed the characteristics of the cortical network and the modulating effect of attention on semantic integration. To further investigate the related cortical regions, we constructed a functional brain network for processing attended AV stimuli to evaluate the nodal properties using a graph-based method. The results of the fMRI and graph-based analyses showed that the semantic integration with attention activated the anterior temporal lobe (ATL), temporoparietal junction (TPJ), and frontoparietal cortex, with the ATL showing the highest nodal degree and efficiency; in contrast, semantic integration without attention involved a relatively small cortical network, including the posterior superior temporal gyrus (STG), Heschl’s gyrus (HG), and precentral gyrus. These results indicated that semantic integration is a complex cognitive process that occurs not only in the attended condition but also in the unattended condition, and that attention could modulate the distribution of cortical networks related to semantic integration. We suggest that semantic integration with attention is a conscious process and needs a wide cortical network working together, in which the ATL plays the role of a central hub; in contrast, semantic integration without attention is a pre-attentive process and involves a relatively smaller cortical network, in which the HG may play an important role. Our study will provide valuable insights into semantic integration and will be useful for investigations on multisensory integration and attention mechanism at multiple processing stages and levels within the cortical hierarchy. [ABSTRACT FROM AUTHOR]

Single-point incremental forming (SPIF) is a technology that allows incremental manufacturing of complex parts from a flat sheet using simple tools; further, this technology is flexible and economical. Measuring the forming force using this technology helps in preventing failures, determining the optimal processes, and implementing on-line control. In this paper, an experimental study using SPIF is described. This study focuses on the influence of four different process parameters, namely, step size, tool diameter, sheet thickness, and feed rate, on the maximum forming force. For an efficient force predictive model based on an adaptive neuro-fuzzy inference system (ANFIS), an artificial neural network (ANN) and a regressions model were applied. The predicted forces exhibited relatively good agreement with the experimental results. The results indicate that the performance of the ANFIS model realizes the full potential of the ANN model. [ABSTRACT FROM AUTHOR]

Complex network theory has been successful at unveiling the topology of the brain and showing alterations to the network structure due to brain disease, cognitive function and behavior. Functional connectivity networks (FCNs) represent different brain regions as the nodes and the connectivity between them as the edges of a graph. Graph theoretic measures provide a way to extract features from these networks enabling subsequent characterization and discrimination of networks across conditions. However, these measures are constrained mostly to binary networks and highly dependent on the network size. In this paper, we propose a novel graph-to-signal transform that overcomes these shortcomings to extract features from functional connectivity networks. The proposed transformation is based on classical multidimensional scaling (CMDS) theory and transforms a graph into signals such that the Euclidean distance between the nodes of the network is preserved. In this paper, we propose to use the resistance distance matrix for transforming weighted functional connectivity networks into signals. Our results illustrate how well-known network structures transform into distinct signals using the proposed graph-to-signal transformation. We then compute well-known signal features on the extracted graph signals to discriminate between FCNs constructed across different experimental conditions. Based on our results, the signals obtained from the graph-to-signal transformation allow for the characterization of functional connectivity networks, and the corresponding features are more discriminative compared to graph theoretic measures. [ABSTRACT FROM AUTHOR]

Big web data from sources including online news and Twitter are good resources for investigating deep learning. However, collected news articles and tweets almost certainly contain data unnecessary for learning, and this disturbs accurate learning. This paper explores the performance of word2vec Convolutional Neural Networks (CNNs) to classify news articles and tweets into related and unrelated ones. Using two word embedding algorithms of word2vec, Continuous Bag-of-Word (CBOW) and Skip-gram, we constructed CNN with the CBOW model and CNN with the Skip-gram model. We measured the classification accuracy of CNN with CBOW, CNN with Skip-gram, and CNN without word2vec models for real news articles and tweets. The experimental results indicated that word2vec significantly improved the accuracy of the classification model. The accuracy of the CBOW model was higher and more stable when compared to that of the Skip-gram model. The CBOW model exhibited better performance on news articles, and the Skip-gram model exhibited better performance on tweets. Specifically, CNN with word2vec models was more effective on news articles when compared to that on tweets because news articles are typically more uniform when compared to tweets. [ABSTRACT FROM AUTHOR]

We provide a novel computational framework on how biological and artificial agents can learn to flexibly couple and decouple neural task modules for cognitive processing. In this way, they can address the stability-plasticity dilemma. For this purpose, we combine two prominent computational neuroscience principles, namely Binding by Synchrony and Reinforcement Learning. The model learns to synchronize task-relevant modules, while also learning to desynchronize currently task-irrelevant modules. As a result, old (but currently task-irrelevant) information is protected from overwriting (stability) while new information can be learned quickly in currently task-relevant modules (plasticity). We combine learning to synchronize with task modules that learn via one of several classical learning algorithms (Rescorla-Wagner, backpropagation, Boltzmann machines). The resulting combined model is tested on a reversal learning paradigm where it must learn to switch between three different task rules. We demonstrate that our combined model has significant computational advantages over the original network without synchrony, in terms of both stability and plasticity. Importantly, the resulting models’ processing dynamics are also consistent with empirical data and provide empirically testable hypotheses for future MEG/EEG studies. [ABSTRACT FROM AUTHOR]

Classifying bank accounts by using transaction data is encouraging in cracking down on illegal financial activities. However, few research simultaneously use heterogenous features, which are embedded in the time series data. In this paper, a two route convolution neural network TRHD-CNN model, fed with two types of heterogeneous feature matrices, is proposed for classifying the bank accounts. TRHD-CNN adopts divide and conquer strategy to extract characteristics from two types of data source independently. The strategy is proved able in mining complementary classification characteristics. We firstly transfer the original log data into a directed and dynamic transaction network. On the basis of that, two feature generation methods are devised for extracting information from local topological structure and time series transaction respectively. A DirectedWalk method is developed in this paper to learning the network vector of vertices used for embedding the neighbor relationship of bank account. The extensive experimental results, conducted on a real bank transaction dataset that contains illegal pyramid selling accounts, show the significant advantage of TRHD-CNN over the existing methods. TRHD-CNN can provide recall scores up to 5.15% higher than competing methods. In addition, the two-route architecture of TRHD-CNN is easy to extend to multi-route scenarios and other fields. [ABSTRACT FROM AUTHOR]

Psychological and personality factors, socioeconomic status, and brain properties all contribute to chronic pain but have essentially been studied independently. Here, we administered a broad battery of questionnaires to patients with chronic back pain (CBP) and collected repeated sessions of resting-state functional magnetic resonance imaging (fMRI) brain scans. Clustering and network analyses applied on the questionnaire data revealed four orthogonal dimensions accounting for 56% of the variance and defining chronic pain traits. Two of these traits—Pain-trait and Emote-trait—were associated with back pain characteristics and could be related to distinct distributed functional networks in a cross-validation procedure, identifying neurotraits. These neurotraits showed good reliability across four fMRI sessions acquired over five weeks. Further, traits and neurotraits all related to the income, emphasizing the importance of socioeconomic status within the personality space of chronic pain. Our approach is a first step in providing metrics aimed at unifying the psychology and the neurophysiology of chronic pain applicable across diverse clinical conditions. [ABSTRACT FROM AUTHOR]

In the last decades, huge efforts have been made in the bioinformatics community to develop machine learning-based methods for the prediction of structural features of proteins in the hope of answering fundamental questions about the way proteins function and their involvement in several illnesses. The recent advent of Deep Learning has renewed the interest in neural networks, with dozens of methods being developed taking advantage of these new architectures. However, most methods are still heavily based pre-processing of the input data, as well as extraction and integration of multiple hand-picked, and manually designed features. Multiple Sequence Alignments (MSA) are the most common source of information in de novo prediction methods. Deep Networks that automatically refine the MSA and extract useful features from it would be immensely powerful. In this work, we propose a new paradigm for the prediction of protein structural features called rawMSA. The core idea behind rawMSA is borrowed from the field of natural language processing to map amino acid sequences into an adaptively learned continuous space. This allows the whole MSA to be input into a Deep Network, thus rendering pre-calculated features such as sequence profiles and other features calculated from MSA obsolete. We showcased the rawMSA methodology on three different prediction problems: secondary structure, relative solvent accessibility and inter-residue contact maps. We have rigorously trained and benchmarked rawMSA on a large set of proteins and have determined that it outperforms classical methods based on position-specific scoring matrices (PSSM) when predicting secondary structure and solvent accessibility, while performing on par with methods using more pre-calculated features in the inter-residue contact map prediction category in CASP12 and CASP13. Clearly demonstrating that rawMSA represents a promising development that can pave the way for improved methods using rawMSA instead of sequence profiles to represent evolutionary information in the coming years. Availability: datasets, dataset generation code, evaluation code and models are available at: . [ABSTRACT FROM AUTHOR]

Emotion recognition plays an important role in human-computer interaction. Previously and currently, many studies focused on speech emotion recognition using several classifiers and feature extraction methods. The majority of such studies, however, address the problem of speech emotion recognition considering emotions solely from the perspective of a single language. In contrast, the current study extends monolingual speech emotion recognition to also cover the case of emotions expressed in several languages that are simultaneously recognized by a complete system. To address this issue, a method, which provides an effective and powerful solution to bilingual speech emotion recognition, is proposed and evaluated. The proposed method is based on a two-pass classification scheme consisting of spoken language identification and speech emotion recognition. In the first pass, the language spoken is identified; in the second pass, emotion recognition is conducted using the emotion models of the language identified. Based on deep learning and the i-vector paradigm, bilingual emotion recognition experiments have been conducted using the state-of-the-art English IEMOCAP (four emotions) and German FAU Aibo (five emotions) corpora. Two classifiers along with i-vector features were used and compared, namely, fully connected deep neural networks (DNN) and convolutional neural networks (CNN). In the case of DNN, 64.0% and 61.14% unweighted average recalls (UARs) were obtained using the IEMOCAP and FAU Aibo corpora, respectively. When using CNN, 62.0% and 59.8% UARs were achieved in the case of the IEMOCAP and FAU Aibo corpora, respectively. These results are very promising, and superior to those obtained in similar studies on multilingual or even monolingual speech emotion recognition. Furthermore, an additional baseline approach for bilingual speech emotion recognition was implemented and evaluated. In the baseline approach, six common emotions were considered, and bilingual emotion models were created, trained on data from the two languages. In this case, 51.2% and 51.5% UARs for six emotions were obtained using DNN and CNN, respectively. The results using the baseline method were reasonable and promising, showing the effectiveness of using i-vectors and deep learning in bilingual speech emotion recognition. On the other hand, the proposed two-pass method based on language identification showed significantly superior performance. Furthermore, the current study was extended to also deal with multilingual speech emotion recognition using corpora collected under similar conditions. Specifically, the English IEMOCAP, the German Emo-DB, and a Japanese corpus were used to recognize four emotions based on the proposed two-pass method. The results obtained were very promising, and the differences in UAR were not statistically significant compared to the monolingual classifiers. [ABSTRACT FROM AUTHOR]

While there have been many studies using machine learning (ML) algorithms to predict process outcomes and device performance in semiconductor manufacturing, the extensively developed technology computer-aided design (TCAD) physical models should play a more significant role in conjunction with ML. While TCAD models have been effective in predicting the trends of experiments, a machine learning statistical model is more capable of predicting the anomalous effects that can be dependent on the chambers, machines, fabrication environment, and specific layouts. In this paper, we use an analytics-statistics mixed training (ASMT) approach using TCAD. Under this method, the TCAD models are incorporated into the machine learning training procedure. The mixed dataset with the experimental and TCAD results improved the prediction in terms of accuracy. With the application of ASMT to the BOSCH process, we show that the mean square error (MSE) can be effectively decreased when the analytics-statistics mixed training (ASMT) scheme is used instead of the classic neural network (NN) used in the baseline study. In this method, statistical induction and analytical deduction can be combined to increase the prediction accuracy of future intelligent semiconductor manufacturing. [ABSTRACT FROM AUTHOR]

Neural networks are required to meet significant metabolic demands associated with performing sophisticated computational tasks in the brain. The necessity for efficient transmission of information imposes stringent constraints on the metabolic pathways that can be used for energy generation at the synapse, and thus low availability of energetic substrates can reduce the efficacy of synaptic function. Here we study the effects of energetic substrate availability on global neural network behavior and find that glucose alone can sustain excitatory neurotransmission required to generate high-frequency synchronous bursting that emerges in culture. In contrast, obligatory oxidative energetic substrates such as lactate and pyruvate are unable to substitute for glucose, indicating that processes involving glucose metabolism form the primary energy-generating pathways supporting coordinated network activity. Our experimental results are discussed in the context of the role that metabolism plays in supporting the performance of individual synapses, including the relative contributions from postsynaptic responses, astrocytes, and presynaptic vesicle cycling. We propose a simple computational model for our excitatory cultures that accurately captures the inability of metabolically compromised synapses to sustain synchronous bursting when extracellular glucose is depleted. [ABSTRACT FROM AUTHOR]