Early diagnosis of chronic diseases represents one of the most significant challenges and opportunities in modern healthcare, with profound implications for improving patient outcomes and alleviating the substantial financial burdens placed on healthcare systems globally. An emerging technological paradigm that promises to address this challenge involves the development of “smart” biosensors. These biosensors are sophisticated analytical devices that integrate advanced machine learning algorithms, particularly artificial neural networks (ANNs), directly with sensing hardware to process complex, multivariate electrochemical and chemiresistive data in real-time. This review critically evaluates ANN-based biosensing systems for chronic disease management, conducting a comprehensive comparison of model architectures across various diagnostic and predictive applications and also highlighting persistent research gaps. While ANNs offer formidable pattern recognition capabilities, their practical performance and clinical utility remain fundamentally constrained by a series of challenges related to data availability, quality, security, and the inherent complexities of biological systems. Our analysis reveals that while substantial progress has been made, particularly in diabetes management and cancer screening via breath analysis, critical hurdles persist in sensor stability, model generalizability, and system integration. Ultimately, this review provides not merely a catalogue of technologies but a comprehensive, critical analysis aimed at equipping researchers with the insights needed to overcome the interdisciplinary barriers to achieving reliable, accessible, and early diagnosis of chronic diseases through next-generation intelligent biosensing platforms. We argue that the path forward requires a concerted focus on generating robust clinical validation data, developing interpretable and trustworthy models, and creating standardized frameworks for system evaluation and deployment.

This article introduces a novel hybrid method to enable the self-evolution of equipment digital twins (DTs), allowing them to continuously and accurately mirror their physical counterparts. Self-evolution is the process by which a DT autonomously updates its models using real-time sensor data, adapting to dynamic real-world behavior. To enhance this process, we propose a data-physics driven approach that synergistically integrates meta-learning and continual learning. Our method begins by designing an extended residual model using a Koopman autoencoder (KAE) neural network. This component bridges the gap between an imperfect analytical physics model and actual equipment behavior. Next, we employ the Reptile meta-learning algorithm to train offline a versatile foundation model on historical data, endowing it with strong adaptability for rapid learning from new information. A key innovation is a periodic event-triggered mechanism, which monitors the DT's simulation accuracy against a fixed time window. When a performance discrepancy is detected, it automatically triggers a self-evolution cycle. The foundation model is then updated through a fine-tuning strategy based on continual learning with random reinitialization. This fusion of offline meta-learning and online continual learning allows the DT to quickly adapt to new, unseen scenarios, ensuring it reflects the physical equipment's state in real-time. We validate the effectiveness and improved performance of our proposed framework through a comprehensive robot simulation case study.

Electrocardiography (ECG) is the gold standard for monitoring heart function and is crucial for preventing the worsening of cardiovascular diseases (CVDs). However, the inconvenience of ECG acquisition poses challenges for long-term continuous monitoring. Consequently, researchers have explored non-invasive and easily accessible photoplethysmography (PPG) as an alternative, converting it into ECG. Previous studies have focused on peaks or simple mapping to generate ECG, ignoring the inherent periodicity of cardiovascular signals. This results in an inability to accurately extract physiological information during the cycle, thus compromising the generated ECG signals' clinical utility. To this end, we introduce a novel PPG-to-ECG translation model called CATransformer, capable of adaptive modeling based on the cardiac cycle. Specifically, CATransformer automatically extracts the cycle using a cycle-aware module and creates multiple semantic views of the cardiac cycle. It leverages a transformer to capture detailed features within each cycle and the dynamics across cycles. Our method outperforms existing approaches, exhibiting the lowest RMSE across five paired PPG-ECG databases. Additionally, extensive experiments are conducted on four cardiovascular-related tasks to assess the clinical utility of the generated ECG, achieving consistent state-of-the-art performance. Experimental results confirm that CATransformer generates highly faithful ECG signals while preserving their physiological characteristics.

With the extraordinary success of generative artificial intelligence, large pretrained models (LPMs) have been widely used to achieve human-level performance. Despite the one-shot capability, it is always preferred to fine-tune the LPMs for domain-specific downstream tasks. Therefore, the federated learning system is leveraged to fine-tune the large pretrained models enabling concurrrently use multiple distributed clients as well as their local datasets. While the first-order fine-tuning methods suffer from high computational and memory costs due to the backward propagation, we are motivated to propose a federated zeroth-order fine-tuning method with only forward propagation. Moreover, we also leverage differential privacy to further preserve the data privacy of local clients. Experimental results illustrate that our proposed federated zeroth-order method can reduce the memory and retain a similar testing accuracy over the state-of-the-art benchmarks.

Tele-homecare has become a promising approach to meet the growing demand for elderly and disability care. In such a context, ensuring human-robot interaction safety during teleoperation poses a critical challenge. Existing teleoperation control approaches focus solely on the robot’s end-effector trajectory, failing to handle inevitable or even desirable contacts on other robot links. This paper proposes a teleoperated shared-control strategy to deal with this challenge. A lightweight exoskeleton is developed to teleoperate the robot and give force feedback to the operator. Additionally, an exoskeleton-based shared-control strategy is proposed to integrate operator commands with real-time proximity sensing information, allowing the robot to avoid collisions while executing tasks. To react to inevitable contact, the force feedback function is incorporated into the proposed strategy to enable the operator to experience intuitive contact. Comparative experiments and a demonstration are designed to evaluate the feasibility and reliability of the proposed strategy in a tele-homecare scenario. Compared to the traditional teleoperation strategy, the proposed method can greatly reduce the contact forces on the robot’s links, indicating the potential of the proposed strategy in advancing safety in tele-homecare systems.

With the rise of the Internet of Things, wearable devices have enabled real-time health monitoring, particularly through physiological signals like electrocardiograms (ECG). The standard 12-lead ECG records the electrical activity of the heart from multiple perspectives, providing valuable insights into cardiac health. However, existing 12-lead ECG analysis methods often treat leads as channel-level arrangements or rely on spatial adjacency to predefine lead connections, limiting their ability to capture the complex spatial and functional relationships between leads fully. To address this limitation, we propose TGLLNet, a task-driven model that automatically learns interlead relationships to improve multilabel ECG classification. TGLLNet adaptively learns lead connectivity patterns and relational strengths, enhancing ECG representation and improving model generalizability across tasks. Specifically, TGLLNet employs a temporal graph construction module to convert ecg signals into temporal graphs and uses a residual pyramid graph convolution module for multilevel graph embeddings, utilizing a graph convolutional network with independently learnable adjacency matrices. Combined with a temporal context convolution module, TGLLNet captures spatio-temporal dependencies, significantly improving ECG representation. Experimental results on seven tasks from PTB-XL and CPSC2018 datasets demonstrate that TGLLNet outperforms existing methods, showing superior generalizability across different tasks. Our code is available athttps://github.com/rosemary333/TGLLnet.

The equipment digital twins (EDTs) for discrete manufacturing should be calibrated quickly to avoid irreversible physical damage to the equipment caused by biased control commands. Therefore, an online credibility assessment method for EDTs is urgently needed. However, existing assessment approaches consume too much time, and thus could not reveal dynamic faults in time. In this paper, the dynamic relationship between online evolution and actual applications of EDTs is investigated. Then, two steps are proposed to accelerate the assessment process significantly. One involves pre-constructing a performance-deviation-agent (PDA), and the other involves dynamically fitting the application-time-window (ATW) probability distribution. The methodology is applicable to discrete manufacturing processes. The dynamic credibility of EDT evolution process can be updated after every iteration of the model evolution. Sorting manufacturing equipment was used as a case study to demonstrate the effectiveness of this method. The time consumption was reduced by 90% compared with traditional assessment methods in the case.

The breakthroughs of next-generation information technologies have accelerated the advancement of industrial cyber-physical intelligence (ICPI), particularly in system intelligence and applications. However, this progress has also brought challenges in ensuring operational reliability and system intelligence. Anomaly detection, a critical component of fault-tolerant and intelligent ICPI, is usually addressed by treating it as a one-class classification and location problem. While autoencoder frameworks have shown promise in addressing this challenge, most existing methods usual struggle with precise anomaly identification or require resource-intensive region-based training. Furthermore, the dynamic nature of anomalies and the scarcity of labeled training data complicate the development and evaluation of anomaly detection models. In this article, an innovative tensor-based knowledge distillation approach (TKDA) is introduced, which integrates a pretrained teacher network, a tensor-decomposed student network, and a denoising module into a unified framework. Anomalies are identified and localized by analyzing differences in intermediate activation values between teacher and student networks during data processing. Extensive experiments demonstrate that TKDA addresses the limitations of low accuracy in anomaly location and inefficiency in computational processes, achieving significant improvements across diverse datasets, including F-MNIST, MNIST, CIFAR-10, MVTecAD, Retinal-OCT, and two medical datasets.

The gait index (GI), a valuable metric to assess human gait, incorporates clinically relevant parameters such as walking speed, knee angle, stride length, and stance-to-swing phase ratio. This index offers insights into an individual's gait pattern, aiding in the identification of subtle gait abnormalities and enabling continuous monitoring of gait changes over time. Building upon this foundation, the present study investigated the influence of specific gait parameters and demographic factors on the gait index, alongside their interaction effects. Analyzing data from 120 healthy individuals using beta regression models, we uncovered significant predictors and interaction effects shaping the Index. Our comparative assessment between Variable Dispersion Beta Regression (VDBR) and Fixed Dispersion Beta Regression (FDBR) models revealed VDBR's superiority over FPBR in capturing gait data heterogeneity. Our analysis revealed that while aging was correlated with decreased GI, gender and BMI exhibited limited individual impact. However, gait-specific predictors such as knee angle, stride length, walking speed, and stance-to-swing phase ratio significantly contributed to GI variability. Additionally, significant interaction effects were identified between knee angle and height normalized stride length, age and knee angle, and age and walking speed, highlighting the complex interplay between demographic and gait-related factors. These findings underscore the multifaceted nature of gait dynamics and offer valuable insights for clinicians, aiding in precise gait pattern assessment and informing the development of gait-related clinical practice, preventive care strategies, and rehabilitation programs. Overall, our research contributes to enhancing mobility and functionality in individuals with gait degradation by identifying significant predictors and interaction effects.

This meta-analysis and systematic review, conducted in accordance with PRISMA guidelines, explores the efficacy of cuff-less blood pressure (BP) monitoring methods, particularly focusing on photoplethysmogram-based technologies. This comprehensive analysis carefully searched prominent databases such as MEDLINE, PubMed, AMED, Embase, and IEEE-Xplore, encompassing 25 studies with a collective participant pool of 21 142 individuals. The study primarily investigates the accuracy and practicality of continuous BP estimation devices and algorithms, aiming to assess their suitability for daily or long-term, as well as their applicability and usability across a broad population. The mean disparities were 4.14 mmHg for systolic blood pressure (SBP) and 2.79 mmHg for diastolic blood pressure (DBP), highlighting a close congruence with established measurement techniques. An in-depth analysis into specific methodologies reveals that Pulse Waveform Analysis (PWA) demonstrates a more favorable performance compared to Pulse Wave Velocity (PWV) for both SBP and DBP, although these differences are not statistically significant. The findings indicate a promising future for wearable devices in short-term BP monitoring scenarios. Both PWA and PWV methods in wearable formats have shown considerable potential as effective tools for BP assessment. However, the study underscores the need for further research, particularly targeting hypertensive populations, to validate the long-term effectiveness and reliability of these wearables. Finally, this investigation is crucial for establishing the role of wearables in ongoing, reliable BP monitoring, especially when considered in conjunction with other health monitoring technologies.

Outlier detection is a significant challenge in Internet of Things (IoT)-based systems, which encompass a multitude of sensor nodes deployed for diverse applications. Ensuring accurate data transmission from these nodes to base stations is crucial, as outliers (fault/event/intrusion) can adversely impact data processing accuracy and overall Quality of Service. The principal component analysis (PCA) has gained popularity for outlier detection in IoT, with recursive PCA (R-PCA) being a widely used method. In this article, we explore popular PCA-based approaches and present an optimized, real-time, and reproducible enhancement to the existing R-PCA method. Our proposed improvement focuses on a data-driven approximation of the distribution of squared prediction error (SPE) scores, a fundamental component of PCA-based outlier detection. We address theoretical ambiguities in the assumptions underlying SPE scores in the existing R-PCA method. Through simulations, we demonstrate the inaccurate distributional assumption of SPE scores in the specified scheme. Additionally, we introduce a more suitable Satterthwaite-based approximation of the SPE score distribution, supported by quantile-quantile (Q-Q) plots. The effectiveness of the proposed approximation is validated through performance evaluation metrics, demonstrating its superiority over the Gaussian approximation used in R-PCA schemes. Furthermore, we provide an overview of our proposed scheme, which can be implemented in any PCA-based outlier detection system used by IoT practitioners and engineers. Our research contributes to advancing outlier detection methodologies in IoT-based systems, enabling more reliable anomaly detection and improved system performance.

Internet-of-Things (IoT), is the comprehensive interconnection systems of computational, networking and physical devices with the important goal of providing proactive and personalized services efficiently. The foundation of such services is big data integration and processing among various devices, which brings important challenges including data fusion, transferring and sharing of computational results. On the other hand, decentralized blockchain platforms provide novel technologies for reliable IoT data integration and processing. In addition, to facilitate decentralization and distribution of IoT big data, tensor-train (TT), as a tensor decomposition method, can play a vital role. Therefore, in this paper, a tensor-train-based permissioned-private (P2) blockchain is proposed to realize the organization, integration, sharing and applications of IoT data for intelligent IoT services. To demonstrate the performance of the proposed method, case studies with IoT data are carried out on permissioned-private chain platform to measure its performance.

Smart manufacturing has been transforming toward industrial digitalization integrated with various advanced technologies. Metaverse has been evolving as a next-generation paradigm of a digital space extended and augmented by reality. In the metaverse, users are interconnected for various virtual activities. In consideration of advanced possibilities that may be brought by the metaverse, it is envisioned that industrial metaverse should be integrated into smart manufacturing to upgrade industry for more visible, intelligent and efficient production in the future. Therefore, a conceptual model, named IMverse Model, and novel characteristics of the industrial metaverse for smart manufacturing are proposed in this article. Besides, an industrial metaverse architecture, named IMverse Architecture, is proposed involving several key enabling technologies. Typical innovative applications of the industrial metaverse throughout the whole product life cycle for smart manufacturing are presented with insights. Nonetheless, in prospect of future, the industrial metaverse still faces limitations and is far from implementation. Thus, challenges and open issues of the industrial metaverse for smart manufacturing are discussed, then outlook is provided for further research and application.

The emerging Metaverse has challenging requirements for the reliability of extended reality (XR) data transmission. Configurable communication is a promising technology to improve the XR communication performance, where the intelligent reflecting surface (IRS) is representative of the ability to control transmission channels. However, because of the absence of incentives and untrust among IRS and Metaverse users, there is no easy way to establish the configuration resource scheduling for XR communication. Existing trusted third party-based methods face single-point/collusion attacks, inefficiency in arbitration, and low intelligence problems. To solve these problems, we propose a swarm learning (SL)-based secure configurable resource trading mechanism for reliable 6G-Metaverse XR communication. First, an SL-based configurable resource trading framework is established, which includes two designed subchains for decentralized IRS resource management and intelligent allocation. Second, a smart contract-enabled configurable resource trading scheme is designed, where decentralized trust is built among IRS devices, Metaverse users, and base stations. Third, we propose a decentralized federated learning (FL)-driven IRS allocation scheme, which consists of XR communication-related data collection, model training, and resource configuration. Finally, experimental results demonstrate the effectiveness of the proposed SL-based configurable resource trading for reliable XR communication.

Throughout human history, deadly infectious diseases emerged occasionally. Even with the present-day advanced healthcare systems, the COVID-19 has caused more than six million deaths worldwide (as of 27 July 2022). Currently, researchers are working to develop tools for better and effective management of the pandemic. “Contact tracing” is one such tool to monitor and control the spread of the disease. However, manual contact tracing is labor-intensive and time-consuming. Therefore, manually tracking all potentially infected individuals is a great challenge, especially for an infectious disease like COVID-19. To date, many digital contact tracing applications were developed and used globally to restrain the spread of COVID-19. In this work, we perform a detailed review of the current digital contact tracing technologies. We mention some of their key limitations and propose a fully integrated system for contact tracing of infectious diseases using COVID-19 as a case study. Our system has four main modules—1) case maps; 2) exposure detection; 3) screening; and 4) health indicators that take multiple inputs like users’ self-reported information, measurement of physiological parameters, and information of the confirmed cases from the public health, and keeps a record of contact histories using Bluetooth technology. The system can potentially evaluate the users’ risk of getting infected and generate notifications to alert them about the exposure events, risk of infection, or abnormal health indicators. The system further integrates the Web-based information on confirmed COVID-19 cases and screening tools, which potentially increases the adoption rate of the system.

Industrial Cyber Physical Systems can use data and information gained from across a variety of different environments to enable robots that are reconfigurable. Custom grasping is a basic operation a robot must be able to carry out for a given task, i.e., finding the best grasping point for emergent behaviors. However, environmental disturbance and limited data degrade the precision and speed of many tailored machine learning models on robot grasping detection. This paper proposes a region-based method to enable fast custom grasping through fewer RGB-D data. The grasping detection problem is simplified as a two-stage prediction problem. At the first stage, a robust grasp candidate generation strategy is proposed based on the Sobel operator. At the second stage, a region-based predictor is designed to locate the best grasping point-pair for an emergent task. The predictor is trained by a modified consistency based self-training method to realize semi-supervised learning. Experimental results show that the success rate of custom grasping of new emergent object can be increased by 3.4% on average using the proposed method. By introducing data augmentation strategies in training, the success rate is further increased by 9.2% on average. A robot is able to grasp new object with 91.5% success rate using less than 100 training samples. The number of training samples required for the proposed method is less than to 1% of which for the previous works. Note to Practitioners—This research was motivated by the problem of robot reconfigurability for various industrial automation processes and focuses mainly on the recognition of grasping point-pair of emergent object for different task. Existing approaches on robotic grasping detection are tailored to a given object and require expensive training with large amount of labeled data. This paper presents a region-based few shot learning approach that enables the robot to detect the best grasping point-pair autonomously and quickly. We show how to generate candidate point-pairs with image distortion and background disturbance. We then demonstrate how the best grasping point-pair can be located with much less training cost. Experiments suggest that this approach is feasible in robot automation for handling a class of objects. In future research, we will construct behavior learning module to enable evolving cyber-physical robotic system for more purposes.

Differential-private federated learning (DP-FL) has emerged to prevent privacy leakage when disclosing encoded sensitive information in model parameters. However, the existing DP-FL frameworks usually preserve privacy homogeneously across clients, while ignoring the different privacy attitudes and expectations. Meanwhile, DP-FL is hard to guarantee that uncontrollable clients (i.e., stragglers) have truthfully added the expected DP noise. To tackle these challenges, we propose a heterogeneous differential-private federated learning framework, named HDP-FL, which captures the variation of privacy attitudes with truthful incentives. First, we investigate the impact of the HDP noise on the theoretical convergence of FL, showing a tradeoff between privacy loss and learning performance. Then, based on the privacy-utility tradeoff, we design a contract-based incentive mechanism, which encourages clients to truthfully reveal private attitudes and contribute to learning as desired. In particular, clients are classified into different privacy preference types and the optimal privacy-price contracts in the discrete-privacy-type model and continuous-privacy-type model are derived. Our extensive experiments with real datasets demonstrate that HDP-FL can maintain satisfactory learning performance while considering different privacy attitudes, which also validate the truthfulness, individual rationality, and effectiveness of our incentives.

Cyber-physical-social systems (CPSS), an emerging cross-disciplinary research area, combines cyber-physical systems (CPS) with social networking for the purpose of providing personalized services for humans. CPSS big data, recording various aspects of human lives, should be processed to mine valuable information for CPSS services. To efficiently deal with CPSS big data, artificial intelligence (AI), an increasingly important technology, is used for CPSS data processing and analysis. Meanwhile, the rapid development of edge devices with fast processors and large memories allows local edge computing to be a powerful real-time complement to global cloud computing. Therefore, to facilitate the processing and analysis of CPSS big data from the perspective of multi-attributes, a cloud-edge-aided quantized tensor-train distributed long short-term memory (QTT-DLSTM) method is presented in this article. First, a tensor is used to represent the multi-attributes CPSS big data, which will be decomposed into the QTT form to facilitate distributed training and computing. Second, a distributed cloud-edge computing model is used to systematically process the CPSS data, including global large-scale data processing in the cloud, and local small-scale data processed at the edge. Third, a distributed computing strategy is used to improve the efficiency of training via partitioning the weight matrix and large amounts of input data in the QTT form. Finally, the performance of the proposed QTT-DLSTM method is evaluated using experiments on a public discrete manufacturing process dataset, the Li-ion battery dataset, and a public social dataset.

Deep learning has been proved to be one of the most effective method in feature encoding for different intelligent applications such as video-based human action recognition. However, its nonconvex optimization mechanism leads large memory consumption, which hinders its deployment on the smart embedded systems with limited computational resources. To overcome this challenge, we propose a novel deep model compression technique for smart embedded systems, which realizes both the memory size reduction and inference complexity decrease within a small drop of accuracy. First, we propose an improved naive Bayes inference-based channel parameter pruning to obtain a sparse model with higher accuracy. Then, to improve the inference efficiency, the improved Tucker decomposition method is proposed, where an improved genetic algorithm is used to optimize the Tucker ranks. Finally, to elevate the effectiveness of our proposed method, extensive experiments are conducted. The experimental results show that our method can achieve the state-of-the-art performance compared with existing methods in terms of accuracy, parameter compression, and floating-point operations reduction.

Human behavior understanding is playing more and more important role in human-centered Industrial Internet of Video Things (IIoVT) system with the deep combination of artificial intelligence and video-based industrial Internet of Things. However, it requires expensively computational resources, including high-performance computing units and large memory, to train a deep computation model with a large number of parameters, which limits its effectiveness and efficiency for IIoVT applications. In this article, a tensor-train mechanism based deep model is presented for video human behavior understanding to meet the requirement of IIoVT applications. It can get competitive performance in accuracy and training efficiency with potentiality for combination of artificial intelligence and prefront IIoVT system. On the one hand, to achieve desirable accuracy, we improved the conventional CNN and adopted the recurrent neural network mechanism to enhance the video representation over time, which takes the correlation between consecutive deep feature into consideration. On the other hand, to enhance the inference capacity between the spatial and temporal features, we carry out the self-critical reinforcement learning mechanism in parameter learning stage. Meanwhile, to further reduce parameter storage size to meet requirement for the deployment of deep neural network and edge device, the tensor-train mechanism is used, which transforms the parameter matrix to a tensor space and carry out tensor decomposition mechanism to decrease the number of parameter generated in parameter training. Finally, we conduct extensive experiments to evaluate our scheme, and the results demonstrate that our method can improve the training efficiency and save the memory space for the deep computation model with better accuracy.

Video surveillance techniques like scene segmentation are playing an increasingly important role in multimedia Internet-of-Things (IoT) systems. However, existing deep learning-based methods face challenges in both accuracy and memory when deployed on edge computing devices with limited computing resources. To address these challenges, a tensor-train video scene segmentation scheme that compares the local background information in regional scene boundary boxes in adjacent frames is proposed. Compared to the existing methods, the proposed scheme can achieve competitive performance in both segmentation accuracy and parameter compression rate. In detail, first, an improved faster region convolutional neural network (faster-RCNN) model is proposed to recognize and generate a large number of region boxes with foreground and background to achieve boundary boxes. Then, the foreground boxes with sparse objects are removed and the rest are considered as optional background boxes used to measure the similarity between two adjacent frames. Second, to accelerate the training efficiency and reduce memory size, a general and efficient training way using tensor-train decomposition to factor the input-to-hidden weight matrix is proposed. Finally, experiments are conducted to evaluate the performance of the proposed scheme in terms of accuracy and model compression. Our results demonstrate that the proposed model can improve the training efficiency and save the memory space for the deep computation model with good accuracy. This work opens the potential for the use of artificial intelligence methods in edge computing devices for multimedia IoT systems.

Industrial Internet of Things (IIoT), as an important industrial branch of the Internet of Things (IoT), has an essential purpose to improve intelligent industrial production. For this purpose, IIoT big data should be efficiently processed to mine valuable information. In handing the IIoT big data, cloud-edge computing is getting more attention to reduce the interaction latency to meet the real-time requirement, especially in the field of prognostic and health management (PHM). It is expected that artificial intelligence (AI) technologies will significantly change the manner of processing IIoT big data. Therefore, new methods about PHM, combining cloud-edge computing with AI technologies, are required to process the IIoT big data for intelligent industrial manufacturing. As an essential element of PHM, predicting the remaining useful life (RUL) of industrial equipment plays an increasingly crucial role, especially for industrial intelligence. However, traditional methods pay much attention on prediction accuracy and neglect the influence of computing time. In this article, by combining cloud-edge computing with AI technology, a new data-driven method, namely, cloud-edge-based lightweight temporal convolutional networks (LTCNs), for RUL prediction is proposed. First, to meet the real-time requirement, a cloud-edge computing and AI-based framework for RUL prediction is presented. Second, a new model structure named LTCN is proposed and applied in the framework. Real-time prediction results will be obtained in the edge plane and higher accuracy prediction results will be obtained through historical information in the cloud plane. Third, an incremental learning approach based on updating partial parameters of LTCN is discussed to improve the accuracy of prediction models with newly collected data. Experiments show that our method can improve the prediction accuracy and reduce the computational time of RUL.

Integration of each aspect of the manufacturing process with the new generation of information technology such as the Internet of Things, big data, and cloud computing makes industrial manufacturing systems more flexible and intelligent. Industrial big data, recording all aspects of the industrial production process, contain the key value for industrial intelligence. For industrial manufacturing, an essential and widely used electronic device is the lithium-ion battery (LIB). However, accurately predicting the remaining useful life (RUL) of LIB is urgently needed to reduce unexpected maintenance and avoid accidents. Due to insufficient amount of degradation data, the prediction accuracy of data-driven methods is greatly limited. Besides, mathematical models established by model-driven methods to represent degradation process are unstable because of external factors like temperature. To solve this problem, a new LIB RUL prediction method based on improved convolution neural network (CNN) and long short-term memory (LSTM), namely Auto-CNN-LSTM, is proposed in this article. This method is developed based on deep CNN and LSTM to mine deeper information in finite data. In this method, an autoencoder is utilized to augment the dimensions of data for more effective training of CNN and LSTM. In order to obtain continuous and stable output, a filter to smooth the predicted value is used. Comparing with other commonly used methods, experiments on a real-world dataset demonstrate the effectiveness of the proposed method.

Industrial Internet-of-Things (IIoT) has revolutionized almost every aspect of industrial manufacturing through industrial intelligence by incorporating production equipment, mobile terminals, and smart devices with wireless or wired networks. However, industrial visual information, such as images, videos, graphs, and texts, generated and collected from the industrial processes, contains various kinds of hidden value for industrial intelligence. Therefore, for the trend of providing ubiquitous industrial intelligence, new paradigms of perception and processing technologies of visual information such as recognition methods are required. However, industrial visual information is heterogeneous and complex with multiattributes, which presents significant challenges on visual information perception and processing technologies such as multiattributes recognition method. In this article, to provide industrial intelligence, a tensor-based visual feature recognition method is used to recognize the object from the perspective of multiattributes with the combination of attributes. To demonstrate its practical implementation, a case study about the industrial intelligence on the faulty location and diameter of bearings in the IIoT is described. Also, experiments on object recognition are carried out on the public image set COIL-100 to demonstrate the performance of the proposed method.

The industrial Internet of Things (IIoT) is growing quickly due to increasing deployment and integration of smart sensors, instruments, and devices, and software using wired or wireless networks. Through this integrated hardware-software approach, industrial practices will improve significantly, resulting in industrial intelligence for more efficient manufacturing. To realize such industrial intelligence, significant developments in IIoT big data processing and analysis are required to uncover and use hidden essential and valuable information of the production process. But large-scale, streaming, multiattribute IIoT data from production processes are noisy and have redundancies. Therefore, a suitable data processing technique such as tensor-train that can handle these IIoT data is needed. However, existing tensor-train decomposition methods are inefficient and cannot meet the processing demands of the large-scale IIoT big data. In this article, we propose an advanced (improved and highly efficient) distributed tensor-train (ADTT) decomposition method with its incremental computational method for processing IIoT big data. Finally, experiments are carried out on a typical and publicly available IIoT dataset - the bearing test data to verify and measure the performances of the proposed ADTT method.

Smart city, a focus of many researchers from academia and industry, is a successful example of Cyber-Physical-Social Systems (CPSS). Based on the rapid and efficient processing of large-scale data, Smart city, an example of CPSS, has revolutionized the service provision model by providing proactive services for humans. However, to operationalize the services provided in smart cities, a comprehensive analysis of heterogeneous and large-scale big data is required. Further, to speed up data processing and improve the adaptability and extensibility of big data, CPSS big data processing should be realized in the form of blocks and avoid redundant computing on historical data. In this paper, as an extension of multi-order distributed and incremental High-Order Singular Value Decomposition (HOSVD) computing, Ring-based Tree algorithm and Tree-based Tree algorithm are proposed for the problems of increasing scale of processable data and computational efficiency. The experimental and simulation results demonstrate that the proposed algorithms have high performance in terms of error, improvement factor, and improvement factor ratio. At last, to demonstrate the performance of our improved algorithms, a case study about CPSS big data processing is provided.

Big service is an extremely important application of service computing to provide predictive and needed services to humans. To operationalize big services, the heterogeneous data collected from Cyber-Physical-Social Systems (CPSS) must be processed efficiently. However, because of the rapid rise in the volume of data, faster and more efficient computational techniques are required. Therefore, in this paper, we propose a multi-order distributed high-order singular value decomposition method (MDHOSVD) with its incremental computational algorithm. To realize the MDHOSVD, a tensor blocks unfolding integration regulation is proposed. This method allows for the efficient analysis of large-scale heterogeneous data in blocks in an incremental fashion. Using simulation and experimental results from real-life, the high-efficiency of the proposed data processing and computational method, is demonstrated. Further, a case study about cyber-physical-social system data processing is illustrated. The proposed MDHOSVD method speeds up data processing, scales with data volume, improves the adaptability and extensibility over data diversity and converts low-level data into actionable knowledge.

Powered by the technologies that have originated from manufacturing, the fourth revolution of healthcare technologies is happening (Healthcare 4.0). As an example of such revolution, new generation homecare robotic systems (HRS) based on the cyber-physical systems (CPS) with higher speed and more intelligent execution are emerging. In this article, the new visions and features of the CPS-based HRS are proposed. The latest progress in related enabling technologies is reviewed, including artificial intelligence, sensing fundamentals, materials and machines, cloud computing and communication, as well as motion capture and mapping. Finally, the future perspectives of the CPS-based HRS and the technical challenges faced in each technical area are discussed.

The eleven papers presented in this special issue provide a snapshot of the latest advances in the field of enabling technologies in health engineering and health informatics for the new revolution of Healthcare 4.0, hoping to further enable, drive and accelerate the research, development, and application of key technologies into healthcare systems.

C yber- P hysical- S ocial S ystems (CPSS) integrating the cyber, physical, and social worlds is a key technology to provide proactive and personalized services for humans. In this paper, we studied CPSS by taking h uman- i nteraction-aware b ig d ata (HIBD) as the starting point. However, the HIBD collected from all aspects of our daily lives are of high-order and large-scale, which bring ever-increasing challenges for their cleaning, integration, processing, and interpretation. Therefore, new strategies for representing and processing of HIBD become increasingly important in the provision of CPSS services. As an emerging technique, tensor is proving to be a suitable and promising representation and processing tool of HIBD. In particular, tensor networks, as a significant tensor decomposition technique, bring advantages of computing, storage, and applications of HIBD. Furthermore, T ensor- T rain (TT), a type of tensor network, is particularly well suited for representing and processing high-order data by decomposing a high-order tensor into a series of low-order tensors. However, at present, there is still need for an efficient Tensor-Train decomposition method for massive data. Therefore, for larger-scale HIBD, a highly-efficient computational method of Tensor-Train is required. In this paper, a d istributed T ensor- T rain (DTT) decomposition method is proposed to process the high-order and large-scale HIBD. The high performance of the proposed DTT such as the execution time is demonstrated with a case study on a typical form of CPSS data, C omputed T omography (CT) image data.

The deep convolutional computation model (DCCM) enabled remarkable progress in feature learning of industrial big data in Internet of Things. However, as a typical static deep learning model, it is difficult to learn features for incremental industrial big data. To solve this problem, we propose an incremental DCCM by developing two incremental algorithms, i.e., parameter-incremental algorithm and structure-incremental algorithm. The parameter-incremental algorithm aims to incrementally train the fully connected layers together with fine tuning for incorporating the new knowledge into the prior one. Then, the structure-incremental algorithm is used to transfer the previous knowledge by introducing an updating rule of the tensor convolutional, pooling, and fully connected layers. Furthermore, the dropout strategy is extended into the tensor fully connected layer to improve the robustness of the proposed model. Finally, extensive experiments are carried out on the representative datasets including CIFRA and CUAVE to justify the proposed model in terms of adaption, preservation, and convergence efficiency.

Transfer learning, which focuses on finding a favorable representation for instances of different domains based on auxiliary data, can mitigate the divergence between domains through knowledge transfer. Recently, increasing efforts on transfer learning have employeddeepneuralnetworks (DNN) to learn more robust and higher level feature representations to better tackle cross-media disparities. However, only a few articles consider the correction and semantic matching between multi-layer heterogeneous domain networks. In this article, we propose adeep semantic mapping model forheterogeneous multimediatransferlearning (DHTL) using co-occurrence data. More specifically, we integrate the DNN withcanonicalcorrelationanalysis (CCA) to derive a deep correlation subspace as the joint semantic representation for associating data across different domains. In the proposed DHTL, a multi-layer correlation matching network across domains is constructed, in which the CCA is combined to bridge each pair of domain-specific hidden layers. To train the network, a joint objective function is defined and the optimization processes are presented. When the deep semantic representation is achieved, the shared features of the source domain are transferred for task learning in the target domain. Extensive experiments for three multimedia recognition applications demonstrate that the proposed DHTL can effectively find deep semantic representations for heterogeneous domains, and it is superior to the several existing state-of-the-art methods for deep transfer learning.

Recent years have witness a considerable advance of Internet of Things with the tremendous progress of communication theories and sensing technologies. A large number of data, usually referring to big data, have been generated from Internet of Things. In this paper, we present a double-projection deep computation model (DPDCM) for big data feature learning, which projects the raw input into two separate subspaces in the hidden layers to learn interacted features of big data by replacing the hidden layers of the conventional deep computation model (DCM) with double-projection layers. Furthermore, we devise a learning algorithm to train the DPDCM. Cloud computing is used to improve the training efficiency of the learning algorithm by crowdsourcing the data on cloud. To protect the private data, a privacy-preserving DPDCM (PPDPDCM) is proposed based on the BGV encryption scheme. Finally, experiments are carried on Animal-20 and NUS-WIDE-14 to estimate the performance of DPDCM and PPDPDCM by comparing with DCM. Results demonstrate that DPDCM achieves a higher classification accuracy than DCM. More importantly, PPDPDCM can effectively improve the efficiency for training parameters, proving its potential for big data feature learning.

Due to the rapid advances of information technologies, Big Data, recognized with 4Vs characteristics (volume, variety, veracity, and velocity), bring significant benefits as well as many challenges. A major benefit of Big Data is to provide timely information and proactive services for humans. The primary purpose of this paper is to review the current state-of-the-art of Big Data from the aspects of organization and representation, cleaning and reduction, integration and processing, security and privacy, analytics and applications, then present a novel framework to provide high-quality so called Big Data-as-a-Service. The framework consists of three planes, namely sensing plane, cloud plane and application plane, to systemically address all challenges of the above aspects. Also, to clearly demonstrate the working process of the proposed framework, a tensor-based multiple clustering on bicycle renting and returning data is illustrated, which can provide several suggestions for rebalancing of the bicycle-sharing system. Finally, some challenges about the proposed framework are discussed.

Cyber-physical-social systems (CPSS), integrating cyber, physical, and social spaces together, bring both conveniences and challenges to humans. For practical applications and user convenience, it is essential that the Big Data produced in CPSS be processed in real time. Therefore, Big Data computation should avoid redundant computations on historical data when dealing with periodic incoming data. In this paper, we propose a columnwise high-order singular value decomposition (HOSVD) algorithm to realize dimensionality reduction, extraction, and noise reduction for tensor-represented Big Data. First, the distributed HOSVD (DHOSVD) is proposed using the columnwise Jacobi-based approach to realize the distributed computation of HOSVD. Second, big streaming data are continuously produced and the intermediate results could be recorded for the next computational step. Third, we propose a similar columnwise incremental HOSVD (IHOSVD) scheme to support online computation on temporally incremental data streaming. The performance of the two HOSVD-based schemes will illustrate the scalability of our efficient real-time Big Data processing methods.

Currently, a large number of industrial data, usually referred to big data, are collected from Internet of Things (IoT). Big data are typically heterogeneous, i.e., each object in big datasets is multimodal, posing a challenging issue on the convolutional neural network (CNN) that is one of the most representative deep learning models. In this paper, a deep convolutional computation model (DCCM) is proposed to learn hierarchical features of big data by using the tensor representation model to extend the CNN from the vector space to the tensor space. To make full use of the local features and topologies contained in the big data, a tensor convolution operation is defined to prevent overfitting and improve the training efficiency. Furthermore, a high-order backpropagation algorithm is proposed to train the parameters of the deep convolutional computational model in the high-order space. Finally, experiments on three datasets, i.e., CUAVE, SNAE2, and STL-10 are carried out to verify the performance of the DCCM. Experimental results show that the deep convolutional computation model can give higher classification accuracy than the deep computation model or the multimodal model for big data in IoT.

A simple, low-power and wearable health analyzer for early identification and management of some diseases is presented. To achieve this goal, we propose a walking pattern analysis system that uses features, such as speed, energy, turn ratio, and bipedal behavior to characterize and classify individuals in distinct walking-ages. A database is constructed from 74 healthy young adults in the age range from 18 to 60 years using the combination of inertial signals from an accelerometer and a gyroscope on a level path including turns. An efficient advanced signal decomposition method called improved complete ensemble empirical mode decomposition with adaptive noise (improved CEEMDAN) was used for feature extraction. Analyzes show that the gait of healthy able-bodied individuals exhibits a natural bipedal asymmetry to a certain level depending on the activity-type and age, which relate to individual's functional attributes rather than pathological gait. The analysis of turn ratio, a measure of activity-transition energy change and stability, indicated turning to be less locally stable than straight-line walking making it a more reliable measure for determining falls and other health issues. Extracted features were used to analyze two distinct walking-age groups of the healthy young adults based on their walking pattern, classifying 18-45 years old individuals in one group and 46-60 years old in the other group. Our proposed simple, inexpensive walking analyzer system can be easily used as an ambulatory screening tool by clinicians to identify at risk population at the early onset of some diseases.

This paper describes a walking-age pattern analysis and identification system using a 3-D accelerometer and a gyroscope. First, a walking pattern database from 79 volunteers of ages ranging from 10 to 83 years is constructed. Second, using feature extraction and clustering, three distinct walking-age groups, children of ages 10 and below, adults in 20-60s, and elders in 70s and 80s, were identified. For this study, low-pass filtering, empirical mode decomposition, and K-means were used to process and analyze the experimental results. Analysis shows that volunteers' walking-ages can be categorized into distinct groups based on simple walking pattern signals. This grouping can then be used to detect persons with walking patterns outside their age groups. If the walking pattern puts an individual in a higher "walking age" grouping, then this could be an indicator of potential health/walking problems, such as weak joints, poor musculoskeletal support system or a tendency to fall.

A high slew-rate, two-stage, single-ended amplifier for optical imaging applications is reported. Two auxiliary circuits have been added to the core amplifier to boost the speed of the positive and negative slews. The amplifier is designed using parameters form IBM 0.13 μm technology. It achieves 41.2 dB DC gain, 723 MHz unity-gain bandwidth and 540 V/μs positive and 325 V/μs symmetric slew-rates for a load capacitance of 2 pF. A 357% symmetric improvement is achieved for slew rate while power is only increased by ∼3%. The core amplifier dissipates 98 μW from a 1.2 V supply.

Wireless Body Area Networks (WBANs) are mainly characterized by deployment of biomedical sensors around human body which transmit vital signs measurements about healthy status to the coordinator. Depending on the relevance between symptoms and diseases, it may not be necessary for every sensor to transmit its measurements for diagnoses. This paper shows how the relevance can be exploited on the Medium Access Control (MAC) layer by utilizing the mutual information. A theoretical framework is developed for sensor scheduling under an operation cost constraint. It is shown that the compact subset of sensors can be found to provide necessary information for timely and correct diagnoses. Based on the theoretical framework, an algorithm combining sensor selection and information gain is then designed. Simulation results show that the algorithm achieves high performance in terms of energy, latency and collision rate.

In this paper, we focus on the problem of constructing an information gain model for stroke prevention in Ubiquitous Healthcare (U-Health) Wireless Body Area Networks (WBANs). We have constructed an information-based probabilistic relation model among the key indicators and sequenced their data gathering priority and precedence in the WBAN. Then, we constructed a cost function over the energy expenditure involved in their data gathering, and expressed the relationship between utility gain and energy loss as a constrained optimization problem. We also designed an algorithm to carry out the proposed approach. Through simulation study, we demonstrated the validity of some aspects of the approach.

In this paper, the reconstruction problem of compressive sensing algorithm that is exploited for image compression, is investigated. Considering the Total Variation (TV) minimization algorithm, and by adding some new constraints compatible with typical image properties, the performance of the reconstruction is improved. Using DCT and contourlet transforms, sparse expansion of the image are exploited to provide new constraints to remove irrelevant vectors from the feasible set of the optimization problem while keeping the problem as a standard Second Order Cone Programming (SOCP) one. Experimental results show that, the proposed method, with new constraints, outperforms the conventional TV minimization method by up to 2 dB in PSNR.

With the increase of the aging society worldwide, hospitals, medical practitioners and health insurers are now increasingly seeking ways to reduce the cost of healthcare while maitaining its quality. One solution subject to attention from gouvernement and healthcare providers is the U-Health Smart Home that aims to provide non-intrusive and non-invasive monitoring and assistance to the elderly directly in their own home. At POSTECH, the U-Health Smart Home project is focused on building a smart home along with an autonomic system to monitor the home as well as the inhabitants to provide intelligent support and assistance in any situation at anytime. This paper presents our initial results from this project. The first contribution is a general framework for the U-Health smart home and the second one is an initial semantic model that can be used in the autonomic system to provide autonomic support to the elderly.

The effects of white light on the electrical performance of polymer thin-film transistors (PTFTs) based on regioregular poly(3-octylthiophene) (P3OT) are investigated. Upon illumination, a significant increase in the PFET's drain current is observed with a maximum photosensitivity of 10/sup 4/ in the subthreshold operation and a broad-band responsivity with a maximum value of 160 mA/W at irradiance of 1.7 mW/cm/sup 2/ and at low gate biases. The photosensitivity decreases with the increase in the absolute gate bias. The simultaneous control of the device with both the gate voltage and illumination is possible at low irradiances of <0.7 mW/cm/sup 2/. It is found that the illumination effectively decreases the threshold voltage of the device, but it does not change the field-effect mobility. Using a trap model, it is shown that the narrow layers close to the drain and source contacts with high concentrations of defects are two possible regions for photogeneration of excitons and separation of charges. Using the theory of space-charge limited conduction, the extracted band mobility for P3OT is 0.08 cm/sup 2//V/spl middot/s, while a mobility of 8/spl times/10/sup -5/cm/sup 2//V/spl middot/s is found for the regions next to the source and drain contacts. The PTFT's high photosensitivity at zero gate voltage suggests a simple design of low-voltage, high-sensitivity two-terminal photodetectors for applications in large-area flexible optoelectronics.

A new mixer circuit is proposed by using the novel gate-controlled lateral PNP (LPNP) device. A major advantage of this four terminal device is that its collector current I/sub c/ can also be controlled by applying a voltage V/sub G/ to the additional input gate terminal. The associated large-signal and small-signal characteristics, such as current gain h/sub FE/ and transconductance g/sub M/, show very strong non-linear features as a function of the gate voltage. Therefore, this new device is particularly suitable for a mixer circuit for possible applications to portable wireless communications. The primary experimental results show that the mixing outputs for an intermediate frequency (IF) of 0.5 MHz using this new mixer circuit have conversion gain of 5 dB to 12 dB for the input radio frequency (RF) signal ranges of up to 400 MHz. The signal-to-noise ratio of its mixing IF output is about 50 dB. Another potential feature of this mixer circuit is that it can be easily integrated with other required signal processing circuits using a BiCMOS technology, and with no additional processing steps.

A new resonant-tunnel diode (RTD) large-signal DC model suitable for PSPICE simulation is presented in this paper. For better accuracy, the model equations are deliberately chosen through the combination of Gaussian and/or exponential functions, and it can be easily implemented in PSPICE using the FUNCTION statement. Most of the associated parameters required in this new model have explicit relations to the measured I-V curves, and can be easily extracted. This new DC model has been successfully applied to simulating single RTD devices, and a RTD-based three-state memory circuit. Compared to other RTD DC models, the presented model gives better accuracy and has less convergence problems. In addition, the new model can be used to simulate hysteresis effect, and can easily incorporate AC effects.>