In elderly remote healthcare, accurate measurement of plantar and cardiorespiratory functions is essential for early detection of mobility decline and chronic disease management, but traditional electrical wearable sensors typically cannot support integrated multimodal monitoring that such care requires. In this work, an all-in-one photonic solution is developed by embedding flexible polydimethylsiloxane polymer optical fiber (PDMS-POF) channels into an insole and chest belt, which are integrated with a smartphone via a custom 3D-printed connector to simultaneously capture gait, respiration, and heart rate. This system employs a series of signal processing techniques to automatically remove motion artifacts and optical noise, achieving clinical grade accuracy (≤ 0.6% error in stance/swing detection; ≤ 5% error in heart and respiratory rates during static activities). Building on this foundation, we introduce TodyNet-Pro, a temporal dynamic graph neural network designed to fuse multiple physiological signals and learn activity labels end to end using sparsity constraints and temporal weighting. In trials involving ten volunteers and validated against clinical standards, TodyNet-Pro achieved 95.5% accuracy in identifying seven static and dynamic activities, and over 80% accuracy in detecting four abnormal gait patterns, outperforming both standard deep learning models and recent studies on elderly activity recognition.

Sleep staging is a crucial method for the evaluation of sleep quality and the diagnosis of sleep disorders. In recent years, rapid progress has been made in sleep research through the application of fabric computing and neural networks. Flexible fabric sensors introduced by fabric computing minimize the discomfort of data collection devices on individuals, while neural network-based algorithms can automatically perform sleep staging based on the collected signals. However, there are two key challenges hinder the integration of automatic sleep staging networks with fabric computing: (1) signals in fabric-based environments exhibit strong heterogeneity due to the wide range of individuals, and (2) interactions between individuals and the fabric space introduce behavioral dynamics to the system. In this paper, we propose a masked autoencoder-based sleep staging neural networks (MAESleepNet), designed to integrate automatic sleep staging algorithm with fabric space. Specifically, MAESleepNet addresses the challenge of signal heterogeneity by learning fine-grained representations from local signals. Furthermore, MAESleepNet tackle the challenge of behavioral dynamics through stochastic masking and reconstruction pre-training. Experiments were conducted on three public datasets: (1) Sleep-EDF-20, (2) Sleep-EDF-78 and (3) SHHS. MAESleepNet achieves overall accuracies of 88.9%, 85.5%, and 87.3%, respectively, outperforming other state-of-the-art models. Furthermore, feature visualization and reconstruction visualization experiments were also conducted. The results demonstrates that MAESleepNet is an effective solution to the aforementioned challenges, paving the way for seamless integration into the fabric space.

Accurate early detection of obstructive sleep apnea (OSA) is critical for enabling timely auto-adjusting positive airway pressure (APAP) interventions. However, existing methods largely rely on binary classification (normal vs. apnea), failing to capture the transitional state preceding OSA onset and inducing therapy delays-hampered by open issues of ambiguous biomarkers and signal temporal misalignment. To address this, this study redefines sleep physiology into three distinct states: normal breathing, pre-apnea transitional (30 s pre-onset), and apnea. This study further proposes DREAM-OSA, a dual-modal transformer framework that specifically targets the transitional states, providing APAP with a sufficient advance response window (up to 10 s) to enable true early prediction of OSA events. It synergizes the complementary electroencephalogram (EEG) and respiratory signals through: 1) Modality-Specific Tokenization: EEG (decomposed into$\delta, \theta, \alpha, \beta, \gamma$bands) and respiratory signals are segmented into 1 s patches, encoded via dedicated 1DCNNs while preserving temporal-spectral structural information through learnable embeddings; and 2) Hierarchical Attention: Intra-modal self-attention captures temporal-spectral dynamics within each modality, while inter-modal cross-attention models bidirectional EEG-respiratory interactions. Evaluated on the MASS-SS1 dataset vs. the state-of-the-art methods towards real-time OSA early warning, DREAM-OSA achieves: overall accuracy up to 95.0%, and per-class F1-scores reaching 91.9% (normal), 94.9% (transitional), and 97.0% (apnea), demonstrating significantly more reliable detection of the transitional states, whereas its counterparts face performance bottleneck.

Remote healthcare monitoring is a crucial component in the field of medical Internet of Things (IoT), which effectively achieves remote monitoring, collection, and transmission of physiological data by combining AI algorithms with intelligent health monitoring systems to improve people’s quality of life and health. In this work, an AI-enabled scalable smartphone photonic sensing system is developed for remote healthcare monitoring using fiber optic sensors and a smartphone. The smartphone serves as both the light source and interrogator for the system, with the ability to connect to the network for integration with the IoT. Scalability is achieved through a multichannel framework, and by modifying the connector design, the system can incorporate more sensors to monitor multiple physiological parameters in real-time. In addition to acquiring basic respiratory and heartbeat signals and various gait parameters, the system successfully implemented the recognition of various gait patterns and fatigue monitoring using an adapted MobileNetV3 neural network structure. The accuracy of 98.5% for the gait pattern recognition task, and 94% and 95.8% for the mental and muscle fatigue monitoring tasks, respectively, demonstrates the system’s potential as a telemedicine tool. Additionally, the low cost, noninvasiveness, and portability of this innovative sensor system make it highly generalizable.

A novel binary channel fuzzy self-adjusted neural network (BCF-SANN) is proposed and researched for solving time-changing quadratic programming (QP) problems in this article. Unlike the fixed parameters of the typical zeroing neural network, the main parameters of the proposed BCF-SANN are time-changing, and its errors are adaptively quickly convergent. The biggest advantage of the novel neural network is that it combines a fuzzy self-adjusted controller, which takes the errors and derivatives of errors as fuzzy inputs and neural networks, further improving the convergence and robustness of the neural networks. To design the novel neural network, a time-changing QP problem is first established; then, using Lagrange's law, the time-changing QP problem is transformed into a time-changing matrix equation; and finally, based on the time-changing parameter neural dynamics method, a novel BCF-SANN is proposed. The detailed design process is given in this article, and the convergence and robustness of the proposed BCF-SANN are proved by theoretical analysis. Through comparative experiments, it is demonstrated that the proposed BCF-SANN has a faster convergence rate and stronger robustness than the traditional zeroing neural network and 1-D fuzzy recurrent neural network (RNN).

The brain signal classification is the basis for the implementation of brain-computer interfaces (BCIs). However, most existing brain signal classification methods are based on signal processing technology, which require a significant amount of manual intervention, such as channel selection and dimensionality reduction, and often struggle to achieve satisfactory classification accuracy. To achieve high classification accuracy and as little manual intervention as possible, a convolutional dynamically convergent differential neural network (ConvDCDNN) is proposed for solving the electroencephalography (EEG) signal classification problem. First, a single-layer convolutional neural network is used to replace the preprocessing steps in previous work. Then, focal loss is used to overcome the imbalance in the dataset. After that, a novel automatic dynamic convergence learning (ADCL) algorithm is proposed and proved for training neural networks. Experimental results on the BCI Competition 2003, BCI Competition III A, and BCI Competition III B datasets demonstrate that the proposed ConvDCDNN framework achieved state-of-the-art performance with accuracies of 100%, 99%, and 98%, respectively. In addition, the proposed algorithm exhibits a higher information transfer rate (ITR) compared with current algorithms.

Medical Internet of Things technology can effectively enable remote physiological data collection, monitoring, and transmission by integrating flexible sensors, wireless communication, and the human body in a wearable and embedded manner. Herein, a stretchable elastomer optical fiber has been developed and sandwiched with two PMMA optical fibers to form a fully flexible polymer optical fiber sensor. The optical fiber integrated with the system mainly consists of a microcomputer, a light-emitting diode driver, a light-emitting diode light source, a photodiode, and a Bluetooth module. One intelligent wearable photonic sensing system has been developed based on the Beer-Lambert law of the stretchable elastomer optical fiber for remote healthcare monitoring. Benefiting from the use of elastomer polymer materials, the sensing system features a maximum strain of more than 250%, a high tensile strain of up to 100%, and a durability of >500 tests. Also, based on the advantage of elastomer optical fiber, the sensing part can be flexibly pasted on the skin surface as a wearable device for real-time monitoring of multiple physiological parameters. In this study, we successfully realized the monitoring of breathing pattern, heart rate, pulse, facial micro-activity, and joint activity, and the recognition of articulatory activity and knee joint activity using a one-dimensional convolutional neural network, with an accuracy of more than 90% for each activity recognition. Such merits demonstrate its potential as a medical toolkit and indicate promise for remote healthcare monitoring.

A swarm-exploring neurodynamic network (SENN) based on a two-timescale model is proposed in this study for solving nonconvex nonlinear programming problems. First, by using a convergent-differential neural network (CDNN) as a local quadratic programming (QP) solver and combining it with a two-timescale model design method, a two-timescale convergent-differential (TTCD) model is exploited, and its stability is analyzed and described in detail. Second, swarm exploration neurodynamics are incorporated into the TTCD model to obtain an SENN with global search capabilities. Finally, the feasibility of the proposed SENN is demonstrated via simulation, and the superiority of the SENN is exhibited through a comparison with existing collaborative neurodynamics methods. The advantage of the SENN is that it only needs a single recurrent neural network (RNN) interact, while the compared collaborative neurodynamic approach (CNA) involves multiple RNN runs.

Accurate assessment of user mental stress in human-machine system plays a crucial role in ensuring task performance and system safety. However, the underlying neural mechanisms of stress in human-machine tasks and assessment methods based on physiological indicators remain fundamental challenges. In this paper, we employ a virtual unmanned aerial vehicle (UAV) control experiment to explore the reorganization of functional brain network patterns under stress conditions. The results indicate enhanced functional connectivity in the frontal theta band and central beta band, as well as reduced functional connectivity in the left parieto-occipital alpha band, which is associated with increased mental stress. Evaluation of network metrics reveals that decreased global efficiency in the theta and beta bands is linked to elevated stress levels. Subsequently, inspired by the frequency-specific patterns in the stress brain network, a cross-band graph convolutional network (CBGCN) model is constructed for mental stress brain state recognition. The proposed method captures the spatial-frequency topological relationships of cross-band brain networks through multiple branches, with the aim of integrating complex dynamic patterns hidden in the brain network and learning discriminative cognitive features. Experimental results demonstrate that the neuro-inspired CBGCN model improves classification performance and enhances model interpretability. The study suggests that the proposed approach provides a potentially viable solution for recognizing stress states in human-machine system by using EEG signals.

Electroencephalogram (EEG) excels in portraying rapid neural dynamics at the level of milliseconds, but its spatial resolution has often been lagging behind the increasing demands in neuroscience research or subject to limitations imposed by emerging neuroengineering scenarios, especially those centering on consumer EEG devices. Current superresolution (SR) methods generally do not suffice in the reconstruction of high-resolution (HR) EEG as it remains a grand challenge to properly handle the connection relationship amongst EEG electrodes (channels) and the intensive individuality of subjects. This study proposes a deep EEG SR framework correlating brain structural and functional connectivities (Deep-EEGSR), which consists of a compact convolutional network and an auxiliary fully connected network for filter generation (FGN). Deep-EEGSR applies graph convolution adapting to the structural connectivity amongst EEG channels when coding SR EEG. Sample-specific dynamic convolution is designed with filter parameters adjusted by FGN conforming to functional connectivity of intensive subject individuality. Overall, Deep-EEGSR operates on low-resolution (LR) EEG and reconstructs the corresponding HR acquisitions through an end-to-end SR course. The experimental results on three EEG datasets (autism spectrum disorder, emotion, and motor imagery) indicate that: 1) Deep-EEGSR significantly outperforms the state-of-the-art counterparts with normalized mean squared error (NMSE) decreased by 1%-6% and the improvement of signal-to-noise ratio (SNR) up to 1.2 dB and 2) the SR EEG manifests superiority to the LR alternative in ASD discrimination and spatial localization of typical ASD EEG characteristics, and this superiority even increases with the scale of SR.

Artifact removal has been an open critical issue for decades in tasks centering on EEG analysis. Recent deep learning methods mark a leap forward from the conventional signal processing routines; however, those in general still suffer from insufficient capabilities 1) to capture potential temporal dependencies embedded in EEG and 2) to adapt to scenarios without a priori knowledge of artifacts. This study proposes an approach (namely DuoCL) to deep artifact removal with a dual-scale CNN (Convolutional Neural Network)-LSTM (Long Short-Term Memory) model, operating on the raw EEG in three phases: 1) Morphological Feature Extraction, a dual-branch CNN utilizes convolution kernels of two different scales to learn morphological features (individual sample); 2) Feature Reinforcement, the dual-scale features are then reinforced with temporal dependencies (inter-sample) captured by LSTM; and 3) EEG Reconstruction, the resulting feature vectors are finally aggregated to reconstruct the artifact-free EEG via a terminal fully connected layer. Extensive experiments have been performed to compare DuoCL to six state-of-the-art counterparts (e.g., 1D-ResCNN and NovelCNN). DuoCL can reconstruct more accurate waveforms and achieve the highest ${\mathsf{SNR}}$ & correlation (${\mathsf{CC}}$) as well as the lowest error (${\mathsf{RRMSE}}_{\mathsf{t}}$ & ${\mathsf{RRMSE}}_{\mathsf{f}}$). In particular, DuoCL holds potentials in providing a high-quality removal of unknown and hybrid artifacts.

Functional near-infrared spectroscopy (fNIRS) as an emerging optical neuroimaging technique has attracted the interest and attention of many investigators. With the growth of fNIRS data volume, effective data compression methods are urgent. Compressive sensing (CS) has been demonstrated a promising tool to deal with biomedical data. However, whether the compressibility of fNIRS data can discriminate different brain states is unclear. In this study, the fNIRS signals from fifteen attention-deficit/hyperactivity disorder (ADHD) children and fifteen typically developing (TD) children were recorded during an N-back task and a Go/NoGo task respectively. A block sparse Bayesian learning-based CS method was used to reconstruct the compressed fNIRS data. To assess the performance of the CS method, we adopted two metrics, structural similarity index (SSIM) and mean squared error (MSE), both of them effective in evaluating the compressibility of fNIRS data. Then, the two metrics were analyzed to discriminate the brain states of ADHD children and TD children during the two tasks using the multivariate pattern analysis (MVPA) method. As indicated by the results, the CS method could reconstruct the compressed fNIRS data with a high reconstruction quality at different compression ratio ([Formula: see text] and [Formula: see text]). Furthermore, the MVPA method could distinguish different brain states with high accuracy, and identify that the prefrontal cortex is a key brain region for distinguishing ADHD vs. TD or N-back vs. Go/NoGo. These findings indicated that CS is very promising for the storage and transmission of massive fNIRS data, and the compressibility of fNIRS data is a potential biomarker for the diagnosis of ADHD.

Multivariate time series data (Mv-TSD) portray the evolving processes of the system(s) under examination in a “multi-view” manner. Factorization methods are salient for Mv-TSD analysis with the potentials of structural feature construction correlating various data attributes. However, research challenges remain in the derivation of factors due to highly scattered data distribution of Mv-TSD and intensive interferences/outliers embedded in the source data. The proposed Enhanced Bayesian Factorization approach (Enhanced-BF) addresses the challenges in three phases: (1) variant scale partitioning applies to Mv-TSD according to degree of amplitude and obtains the blocks of variant scales; (2) hierarchical Bayesian model for tensor factorization automatically derives the factors of each block with interferences suppressed; (3) Bayesian unification model merges those block factors to construct the final structural features.Enhanced-BFhas been evaluated using a case study of brain data engineering with multivariate electroencephalogram (EEG). Experimental results indicate that the proposed method manifests robustness to the interferences and outperforms the counterparts in terms of operation efficiency and error when factorizing EEG tensor. Besides,Enhanced-BFexcels in factorization-based analysis of ongoing autism spectrum disorder (ASD) EEG: 3 times speed-up in factorization and$87.35\%$accuracy in ASD discrimination. The latent factors (“biomarkers”) can distinctly interpret the typical EEG characteristics of ASD subjects.

Rapid serial visual presentation (RSVP) is a type of electroencephalogram (EEG) pattern commonly used for target recognition. Besides delta- and theta-band responses already used for classification, RSVP task also evokes gamma-band responses having low amplitude and large individual difference. This paper proposes a filter bank spatio-temporal component analysis (FBSCA) method, extracting spatio-temporal features of the gamma-band responses for the first time, to enhance the RSVP classification performance. Considering the individual difference in time latency and responsive frequency, the proposed FBSCA method decomposes the gamma-band EEG data into sub-components in different time-frequency-space domains and seeks the weight coefficients to optimize the combinations of electrodes, common spatial pattern (CSP) components, time windows and frequency bands. Two state-of-the-art methods, i.e. hierarchical discriminant principal component analysis (HDPCA) and discriminative canonical pattern matching (DCPM), were used for comparison. The performance was evaluated in [Formula: see text] cross validations using a public dataset. Study results showed that the FBSCA method outperformed the other methods regardless of number of training trials. These results suggest that the proposed FBSCA method can enhance the RSVP classification.

Biomarkers, such as magnetic resonance imaging (MRI) and electroencephalogram have been used to help diagnose autism spectrum disorder (ASD). However, the diagnosis needs the assist of specialized medical equipment in the hospital or laboratory. To diagnose ASD in a more effective and convenient way, in this article, we propose an appearance-based gaze estimation algorithm-AttentionGazeNet, to accurately estimate the subject's 3-D gaze from a raw video. The experimental results show its competitive performance on the MPIIGaze dataset and the improvement of 14.7% for static head pose and 46.7% for moving head pose on the EYEDIAP dataset compared with the state-of-the-art gaze estimation algorithms. After projecting the obtained gaze vector onto the screen coordinate, we apply accumulated histogram to taking into account both spatial and temporal information of estimated gaze-point and head-pose sequences. Finally, classification is conducted on our self-collected autistic children video dataset (ACVD), which contains 405 videos from 135 different ASD children, 135 typically developing (TD) children in a primary school, and 135 TD children in a kindergarten. The classification results on ACVD shows the effectiveness and efficiency of our proposed method, with the accuracy 94.8%, the sensitivity 91.1% and the specificity 96.7% for ASD.

Extracting the latent factors of big time series data is an important means to examine the dynamic complex systems under observation. These low-dimensional and “small” representations reveal the key insights to the overall mechanisms, which can otherwise be obscured by the notoriously high dimensionality and scale of big data as well as the enormously complicated interdependencies amongst data elements. However, grand challenges still remain: (1) to incrementally derive the multi-mode factors of the augmenting big data and (2) to achieve this goal under the circumstance of insufficient a priori knowledge. This study develops an incrementally parallel factorization solution (namely I-PARAFAC) for huge augmenting tensors (multi-way arrays) consisting of three phases over a cutting-edge GPU cluster: in the “giant-step” phase, a variational Bayesian inference (VBI) model estimates the distribution of the close neighborhood of each factor in a high confidence level without the need for a priori knowledge of the tensor or problem domain; in the “baby-step” phase, a massively parallel Fast-HALS algorithm (namely G-HALS) has been developed to derive the accurate subfactors of each subtensor on the basis of the initial factors; in the final fusion phase, I-PARAFAC fuses the known factors of the original tensor and those accurate subfactors of the “increment” to achieve the final full factors. Experimental results indicate that: (1) the VBI model enables a blind factor approximation, where the distribution of the close neighborhood of each final factor can be quickly derived (10 iterations for the test case). As a result, the model of a low time complexity significantly accelerates the derivation of the final accurate factors and lowers the risks of errors; (2) I-PARAFAC significantly outperforms even the latest high performance counterpart when handling augmenting tensors, e.g., the increased overhead is only proportional to the increment while the latter has to repeatedly factorize the whole tensor, and the overhead in fusing subfactors is always minimal; (3) I-PARAFAC can factorize a huge tensor (volume up to 500 TB over 50 nodes) as a whole with the capability several magnitudes higher than conventional methods, and the runtime is in the order of 1/n to the number of compute nodes; (4) I-PARAFAC supports correct factorization-based analysis of a real 4-order EEG dataset captured from a variety of epilepsy patients. Overall, it should also be noted that counterpart methods have to derive the whole tensor from the scratch if the tensor is augmented in any dimension; as a contrast, the I-PARAFAC framework only needs to incrementally compute the full factors of the huge augmented tensor.

Summary Brain healthcare, when supported by Internet of Things, can perform online and accurate analysis of brain big data for the classification of multivariate Electroencephalogram (EEG), which is a prerequisite for the recent boom in neurofeedback applications and clinical practices. However, it remains a grand research challenge due to (1) the embedded intensive noises and the intrinsic nonstationarity determined by the evolution of brain states; and (2) the lack of a user‐friendly computing platform to sustain the complicated analytics. This study presents the design of an online EEG classification system aided by Cloud centering on a lightweight Convolutional Neural Network (CNN). The system incrementally trains the CNN on Cloud and enables hot deployment of the trained classifier without the need to restart the gateway to adapt to the users' needs. The classifier maintains a High Convolutional Layer to gain the ability of processing high‐dimensional EEG segments. The number of hidden layers is minimized to ensure the efficiency of training. The lightweight CNN adopts an “hourglass” block of fully connected layers to reduce the number of neurons quickly toward the output end. A case study of depression evaluation has been performed against raw EEG datasets to distinguish between (1) Healthy and Major Depression Disorder with an accuracy, sensitivity, and specificity of [98.59% ± 0.28%], [97.77% ± 0.63%], and [99.51% ± 0.19%], respectively; and (2) Effective and Noneffective treatment outcome with an accuracy, sensitivity, and specificity of [99.53% ± 0.002%], [99.50% ± 0.01%], and [99.58% ± 0.02%], respectively. The results show that the classification can be completed several magnitudes faster when EEG is collected on the gateway (several milliseconds vs. 4 seconds).

Online electroencephalograph (EEG) classification is a core service of recently booming brain e-health, but its performance often becomes unstable because (1) conventional end-to-end models (e.g., deep neural network, DNN) largely remain static, while brain states of diseases are highly dynamic and exhibits significant individuality; and (2) EEG analytics are too complicated and have to be sustained by advanced computing services. This study adopts an automatic machine learning method to construct a dual-CNN (convolutional neural network) of high performance in terms of both accuracy and efficiency. The model can optimize its hyperparameters continuously on its own initiative. Experimental results in the evaluation of depression using real EEG datasets indicate that (1) the proposed method executes 3.5 times faster compared with a conventional counterpart; (2) the dual-CNN gains a significant performance improvement (versus CapsuleNet and Resnet-16) in identifying Major Depression Disorder (MDD) with accuracy, sensitivity, and specificity up to 98.81, 98.36, and 99.31 percent respectively; and those for treatment outcome are 99.52, 99.63, and 99.37 percent respectively, and (3) classification can be completed several hundred times faster than EEG being collected upon a COTS computer.

High frequency oscillations (HFOs) are considered as biomarker for epileptogenicity. Reliable automation of HFOs detection is necessary for rapid and objective analysis, and is determined by accurate computation of the baseline. Although most existing automated detectors measure baseline accurately in channels with rare HFOs, they lose accuracy in channels with frequent HFOs. Here, we proposed a novel algorithm using the maximum distributed peak points method to improve baseline determination accuracy in channels with wide HFOs activity ranges and calculate a dynamic baseline. Interictal ripples (80-200[Formula: see text]Hz), fast ripples (FRs, 200-500[Formula: see text]Hz) and baselines in intracerebral EEGs from seven patients with intractable epilepsy were identified by experienced reviewers and by our computer-automated program, and the results were compared. We also compared the performance of our detector to four well-known detectors integrated in RIPPLELAB. The sensitivity and specificity of our detector were, respectively, 71% and 75% for ripples and 66% and 84% for FRs. Spearman's rank correlation coefficient comparing automated and manual detection was [Formula: see text] for ripples and [Formula: see text] for FRs ([Formula: see text]). In comparison to other detectors, our detector had a relatively higher sensitivity and specificity. In conclusion, our automated detector is able to accurately calculate a dynamic iEEG baseline in different HFO activity channels using the maximum distributed peak points method, resulting in higher sensitivity and specificity than other available HFO detectors.

Summary Brain data processing has been embracing the big data era driven by the rapid advances of neuroscience as well as the experimental techniques for recording neuronal activities. Processing of massive brain data has become a constant in neuroscience research and practice, which is vital in revealing the hidden information to better understand the brain functions and malfunctions. Brain data are routinely non‐linear and non‐stationary in nature, and existing algorithms and approaches to neural data processing are generally complicated in order to characterize the non‐linearity and non‐stationarity. Brain big data processing has pressing needs for appropriate computing technologies to address three grand challenges: (1) efficiency, (2) scalability and (3) reliability. Recent advances of computing technologies are making non‐linear methods viable in sophisticated applications of massive brain data processing. General‐purpose Computing on the Graphics Processing Unit (GPGPU) technology fosters an ideal environment for this purpose, which benefits from the tremendous computing power of modern graphics processing units in massively parallel architecture that is frequently an order of magnitude larger than the modern multi‐core CPUs. This article first recaps significant speed‐ups of existing algorithms aided by GPGPU in neuroimaging and processing electroencephalogram (EEG), functional magnetic resonance imaging (fMRI), magnetoencephalography (MEG) and etc. The article then demonstrates a series of successful approaches to processing EEG data in various dimensions and scales in a massively parallel manner: (1)decomposition:a massively parallel Ensemble Local Mean Decomposition (ELMD) algorithm aided by GPGPU can decompose EEG series, which forms the basis of further time‐frequency transformation, in real‐time without sacrificing the precision of processing; (2)synchronization measurement:a parallelized Nonlinear Interdependence (NLI) method for global synchronization measurement of multivariate EEG with speed‐up of more than 1000 times, and it was successful in localization of epileptic focus; and (3)dimensionality reduction:a large‐scale Parallel Factor Analysis which excels in run‐time performance and scales far better by hundreds of times than conventional approach does, and it supports fast factorization of EEG with more than 1000 channels. Through these practices, the massively parallel computing technology manifests great potentials in addressing the grand challenges of brain big data processing. Copyright © 2016 John Wiley & Sons, Ltd.

It has long been an important issue in various disciplines to examine massive multidimensional data superimposed by a high level of noises and interferences by extracting the embedded multi-way factors. With the quick increases of data scales and dimensions in the big data era, research challenges arise in order to (1) reflect the dynamics of large tensors while introducing no significant distortions in the factorization procedure and (2) handle influences of the noises in sophisticated applications. A hierarchical parallel processing framework over a GPU cluster, namely H-PARAFAC, has been developed to enable scalable factorization of large tensors upon a “divide-and-conquer” theory for Parallel Factor Analysis (PARAFAC). The H-PARAFAC framework incorporates a coarse-grained model for coordinating the processing of sub-tensors and a fine-grained parallel model for computing each sub-tensor and fusing sub-factors. Experimental results indicate that (1) the proposed method breaks the limitation on the scale of multidimensional data to be factorized and dramatically outperforms the traditional counterparts in terms of both scalability and efficiency, e.g., the runtime increases in the order of n2 when the data volume increases in the order of n3, (2) H-PARAFAC has potentials in refraining the influences of significant noises, and (3) H-PARAFAC is far superior to the conventional window-based counterparts in preserving the features of multiple modes of large tensors.

It has long been an important issue in various disciplines to examine massive multi-dimensional data by extracting the embedded multi-way factors. With the quick increases in both scales and dimensions of data under analysis, research challenges arise in order to reflect the dynamics of large-scale tensors while introducing no significant distortions in the factorization procedure in sophisticated applications. A massively parallel computing framework, namely H-PARAFAC, has been developed to enable Parallel Factor Analysis (PARAFAC) of massive tensors upon a "divide-and-conquer" theory (a modified alternating least squares approach). The hierarchical framework incorporates a coarse-grained model for coordinating the processing of sub tensors and a fine-grained parallel model for computing each sub tensor and fusing sub-factors. Experiments have been performed on a GPU cluster, and the results indicate that (1) the proposed method breaks the limitation on the size of data to be factorized, and (2) it dramatically outperforms the traditional counterparts in terms of both scalability and efficiency, e.g., The runtime increases linearly with the data volume increases in the order of n3.

Analysis of neural data with multiple modes and high density has recently become a trend with the advances in neuroscience research and practices. There exists a pressing need for an approach to accurately and uniquely capture the features without loss or destruction of the interactions amongst the modes (typically) of space, time, and frequency. Moreover, the approach must be able to quickly analyze the neural data of exponentially growing scales and sizes, in tens or even hundreds of channels, so that timely conclusions and decisions may be made. A salient approach to multi-way data analysis is the parallel factor analysis (PARAFAC) that manifests its effectiveness in the decomposition of the electroencephalography (EEG). However, the conventional PARAFAC is only suited for offline data analysis due to the high complexity, which computes to be$O(n^{2})$with the increasing data size. In this study, a large-scale PARAFAC method has been developed, which is supported by general-purpose computing on the graphics processing unit (GPGPU). Comparing to the PARAFAC running on conventional CPU-based platform, the new approach dramatically excels by${>}360$times in run-time performance, and effectively scales by${>}400$times in all dimensions. Moreover, the proposed approach forms the basis of a model for the analysis of electrocochleography (ECoG) recordings obtained from epilepsy patients, which proves to be effective in the epilepsy state detection. The time evolutions of the proposed model are well correlated with the clinical observations. Moreover, the frequency signature is stable and high in the ictal phase. Furthermore, the spatial signature explicitly identifies the propagation of neural activities among various brain regions. The model supports real-time analysis of ECoG in${>}1{,}000$channels on an inexpensive and available cyber-infrastructure.

How to improve social communication skills for children, especially those with social communication difficulties such as attention deficit/hyperactivity disorder, has long been a challenge faced by researchers and therapists. Recent research indicates that computer-assisted approaches may be effective in addressing this issue. This study aimed to understand children's behaviors and then provide appropriate support to improve their social communication skills. We have established an intelligent system, inside which a child can freely play interactive social skills games with virtual characters. The virtual characters can adjust their own behaviors by adapting to the child's cognitive state (e.g., focus of attention) and affective state (e.g., happiness or surprise). The child's behavior is identified in real-time by recognition of multimodal sensory information, which includes head pose and eye gaze estimation, gesture detection, and affective state detection supported by a series of algorithms proposed in this study. Furthermore, this intelligent system has been enabled in a nonintrusive manner using a novel approach of multicamera surveillance to provide the child with natural interaction with the system. Experimental results show the system can estimate a user's attention and affective states with correctness rates of 93% and 91.3%, respectively. The results obtained suggest that the methods have strong potential as alternative methods for sensing human behavior and providing appropriate support.

The exiting covariance matching method is not suited for real-time applications due to its demand for exhaustive search. Aiming at this problem, we developed a novel approach based on fuzzy genetic algorithm (GA) to boost the computing efficiency of covariance matching. The approach employs GA in searching for optimal solution in a large image region. To avoid premature convergence or local optimum which often occur in traditional GAs, we use a fuzzy inference system to adaptively estimate the crossover and mutation probabilities to gain convergence in a much higher speed than using a conventional GA. Experimental results show that the proposed approach can significantly improve the processing speed of covariance matching, while keeping the matching results almost unchanged. The runtime performance of the proposed approach is faster than its counterparts using exhaustive search with eight times and more.

This paper addresses a virtual force field between mobile robots and obstacles to keep them away with a desired distance. An online learning method of hybrid force/position control is proposed for obstacle avoidance in a robot environment. An Elman neural network is proposed to compensate the effect of uncertainties between the dynamic robot model and the obstacles. Moreover, this paper uses an Elman fuzzy adaptive controller to adjust the exact distance between the robot and the obstacles. The effectiveness of the proposed method is demonstrated by simulation examples.

Thermal models of buildings are often used to identify energy savings within a building. Given that a significant proportion of that energy is typically used to maintain building temperature, establishing the optimal control of the buildings thermal system is important. This requires an understanding of the thermal dynamics of the building, which is often obtained from physical thermal models. However, these models require detailed building parameters to be specified and these can often be difficult to determine. In this paper, we propose an evolutionary approach to parameter identification for thermal models that are formulated as an optimization task. A state-of-the-art evolutionary algorithm, i.e., SaNSDE+, has been developed. A fitness function is defined, which quantifies the difference between the energy-consumption time-series data that are derived from the identified parameters and that given by simulation with a set of predetermined target model parameters. In comparison with a conventional genetic algorithm, fast evolutionary programming, and two state-of-the-art evolutionary algorithms, our experimental results show that the proposed SaNSDE+ has significantly improved both the solution quality and the convergence speed, suggesting this is an effective tool for parameter identification for simulated building thermal models.

Covariance matching is an excellent algorithm of target tracking. In this paper, forgetting factor and random sampling methods are proposed to improve the robustness and efficiency of covariance tracking. First, a distance function between covariance matrixes is weighted by using a forgetting factor based on a fuzzy membership function to overcome the disturbances from similar targets. Then a random sampling method is applied to reduce the computing time in covariance matching and to facilitate real-time object tracking. Experiment results show that the algorithm proposed in this paper can effectively mitigate the clutter and occlusion problems at a high computing speed.

Ensemble empirical-mode decomposition (EEMD) is a novel adaptive time-frequency analysis method, which is particularly suitable for extracting useful information from noisy nonlinear or nonstationary data. Unfortunately, since the EEMD is highly compute-intensive, the method does not apply in real-time applications on top of commercial-off-the-shelf computers. Aiming at this problem, a parallelized EEMD method has been developed using general-purpose computing on the graphics processing unit (GPGPU), namely, G-EEMD. A spectral entropy facilitated by G-EEMD was, therefore, proposed to analyze the EEG data for estimating the depth of anesthesia (DoA) in a real-time manner. In terms of EEG data analysis, G-EEMD has dramatically improved the run-time performance by more than 140 times compared to the original serial EEMD implementation. G-EEMD also performs far better than another parallelized implementation of EEMD bases on conventional CPU-based distributed computing technology despite the latter utilizes 16 high-end computing nodes for the same computing task. Furthermore, the results obtained from a pharmacokinetics/pharmacodynamic (PK/PD) model analysis indicate that the EEMD method is slightly more effective than its precedent alternative method (EMD) in estimating DoA, the coefficient of determination R(2) by EEMD is significantly higher than that by EMD (p < 0.05, paired t-test) and the prediction probability P(k) by EEMD is also slighter higher than that by EMD (p < 0.2, paired t-test).

Artificial Neural Networks (ANNs) have been applied to machine condition monitoring. This paper first addresses a ANN trained by Group Search Optimizer (GSO), which is a novel population based optimization algorithm inspired by animal social foraging behaviour. The global search performance of GSO has been proven to be competitive to other evolutionary algorithms, such as Genetic Algorithms (GAs) and Particle Swarm Optimizer (PSO). Herein, the parameters of a 3-layer feed-forward ANN, including connection weights and bias are tuned by the GSO algorithm. Secondly the GSO based ANN is applied to model and analysis ultrasound data recorded from grinding machines to distinguish different conditions. The real experimental results show that the proposed method is capable to indicate the malfunction of machine condition from the ultrasound data.

To achieve the desired quality in plastic injection molding, advanced monitoring techniques are often recommended in the workshop. Unfortunately, the signal in plastic injection modeling process such as nozzle pressure that is relevant to part quality is not easy to obtain because of the cost of sensors. The sensor-based modeling idea is therefore adopted. In this paper, a new method for predicting the parts weight in plastic injection molding using least squares support vector regression (LS-SVR) is proposed, which is composed of two steps. The first step is to estimate the nozzle pressure with the hydraulic system pressure using an LS-SVR model. The second step is to predict product weight using the estimated nozzle pressure, which is done using another LS-SVR model. The experimental results show that the new method is very effective.

Cost-sensitive classification is an attractive topic in data mining. Although genetic programming (GP) technique has been applied to general classification, to our knowledge, it has not been exploited to address cost-sensitive classification in the literature, where the costs of misclassification errors are non-uniform. To investigate the applicability of GP to cost-sensitive classification, this paper first reviews the existing methods of cost-sensitive classification in data mining. We then apply GP to address cost-sensitive classification by means of two methods through: a) manipulating training data, and b) modifying the learning algorithm. In particular, a constrained genetic programming (CGP), a GP-based cost-sensitive classifier, has been introduced in this study. CGP is capable of building decision trees to minimize not only the expected number of errors, but also the expected misclassification costs through a novel constraint fitness function. CGP has been tested on the heart disease dataset and the German credit dataset from the UCI repository. Its efficacy with respect to cost has been demonstrated by comparisons with non-cost-sensitive learning methods and cost-sensitive learning methods in terms of the costs.

It is very important to use a reliable and inexpensive sensor to obtain useful information about manufacturing processing, such as cutting force for monitoring automated machining. In this paper, the feed-cutting force is estimated using inexpensive current sensors installed on the ac servomotor of a computerized numerical control (CNC) turning center, with the results applied to the intelligent tool wear monitoring system. The mathematical model is used to disclose the implicit dependency of feed-cutting force on feed-motor current and feed speed. Afterwards, a neuro-fuzzy network is used to identify the cutting force with current measurement only. This hybrid math-fuzzy approach will reduce the modeling uncertainty and measurement cost. Finally, the estimated cutting force is applied in the tool-wear monitoring process. Successful experiments demonstrate robustness and effectiveness of the suggested method in the wide range of tool-wear monitoring applications.

Wavelet transforms and fuzzy techniques are used to monitor tool breakage and wear conditions in real time according to the measured spindle and feed motor currents, respectively. First, continuous and discrete wavelet transforms are used to decompose the spindle and feed ac servo motor current signals to extract signal features so as to detect the breakage of drills successfully. Next, the models of the relationships between the current signals and the cutting parameters are established under different tool wear states. Subsequently, fuzzy classification methods are used to detect tool wear states based on the above models. Finally, the two methods above are integrated to establish an intelligent tool condition monitoring system for drilling operations. The monitoring system can detect tool breakage and tool wear conditions using very simple current sensors. Experimental results show that the proposed system can reliably detect tool conditions in drilling operations in real time and is viable for industrial applications.