With the rapid development of the new energy sector, production equipment for battery cells faces increasing challenges in maintaining efficiency and quality. Among these, the laser die cutting and winding machine plays a pivotal role in transforming electrode sheets into finished cells. Its performance directly affects the dimensional precision and internal structural consistency of the cells, which are critical to product quality and production-line efficiency. To tackle the challenges in monitoring and maintaining this critical equipment, we propose a time-series data-driven method based on the Transformer architecture, named multiscale dynamic dilated attention, which effectively predicts sensor offset trajectories and provides early shutdown fault warnings when correction sensors approach their operational limits. Furthermore, this model incorporates adjustable segment sizes and counts, and assigns distinct dilation rates to individual attention heads, this design enables a dynamic tradeoff among receptive field, modeling capacity, and computational cost, allowing fine-grained control over long-range dependence modeling while reducing the canonical self-attention complexity from$O(\mathit {L}^{2})$to approximately$O(\mathit {L})$. Extensive experiments and real-world applications demonstrate that the proposed method achieves state-of-the-art performance in both prediction accuracy and practical deployment, while exhibiting excellent adaptability across diverse operating conditions.

Non-rigid registration of intraoperative tissue is essential for surgical navigation and scene reconstruction in minimally invasive surgery. However, accurate registration remains challenging due to significant tissue deformation and partial overlaps caused by laparoscope movement. We propose an Overlap-Aware Online-Adaptive Non-Rigid Registration Method (OANRM) to address these challenges. The framework introduces a Hierarchical Matching Network (HMNet) that simultaneously predicts overlapping regions and their correspondences through a novel similarity-based approach. Our method uniquely incorporates an online adaptation mechanism that continuously fine-tunes the network parameters using unsupervised losses, enabling robust performance across varying surgical scenarios without requiring additional training data. A Transform Displacement Deformation Prediction (TDDP) module further enhances the framework by handling non-overlapping regions through integrating Random Sample Consensus with distance-based interpolation. The method is validated on both artificial datasets with controlled deformations and clinical datasets from real surgical procedures. Experimental results demonstrate that OANRM achieves state-of-the-art performance, significantly outperforming existing methods in handling complex tissue deformations and varying overlap ratios. https://github.com/AIGCer0807/OANRM.

This article proposes a generalizable nonintrusive load monitoring (NILM) framework to address nonindependent and identically distributed (Non-I.I.D.) data challenges in heterogeneous energy consumption sectors. The proposed framework uses adversarial feature augmentation based on the observed states of appliances in source sectors, and implements unsupervised domain adaptation for NILM tasks in target. A ConvNet feature extractor is built to extract the features of source samples, which are then, integrated with observed labels for adversarial data augmentation in the feature space. The generated features are used to establish a domain-invariant NILM algorithm in an unlabeled manner. Feature augmentation and domain invariant feature extractors are employed to learn effective feature mapping between the source and target sectors, thereby, accomplishing NILM tasks in Non-I.I.D. samples without additional labels. The experimental results validate the effectiveness of the proposed framework in three scenarios consisting of multiple datasets, with the best performance compared to the five state-of-the-art models.

An accurate short-term prediction of wind power is crucial for grid reliability and optimized power generation allocation. Currently, deep learning-based methods for wind power prediction (WPP) focus excessively on exploiting spatial-temporal correlations of multivariate time series and neglect potential confounding factors. However, these factors can introduce spurious correlations, diminishing the reliability and robustness of the methods solely based on correlations. Specifically, traditional deep learning methods lack generalizability across varied power output scenarios due to the interference from confounding factors, such as spatial distribution differences and turbine disturbances. Therefore, this paper proposes a novel method, called Temporal Causal Attention Network (TCAN) for short-term WPP, leveraging the physically meaningful internal-external causality to model interactions among factors, and enhancing generalizability across power output scenarios. In this method, a multiscale temporal transformer is proposed to capture the internal causality of a dynamic time series, and a causal graphs-based mechanism is introduced to identify the external causal path among different factors. In addition, a causal attention-based information updating method is proposed to measure the impact of cause nodes on effect nodes. The real data of eight wind turbines from Asia's largest wind power producer are used to evaluate the proposed method. Results show that compared to the second-best method in the experiment, the average RMSE, MAAPE, and R-Squared for the proposed method for 10-min WPP are enhanced by 35.34%, 32.73%, and 4.73%, respectively.

Convolutional neural networks (CNNs) have been widely used in fault diagnosis due to their superiority in feature extraction. Traditional CNNs are a type of closed-box techniques with little interpretability, and their effectiveness is greatly affected when fault mechanisms and modes are extremely complex. To cope with such issue, this article presents a way to integrate outlier-type prior knowledge into CNNs based on an attention mechanism for fault diagnosis. First, outliers of the image-like data obtained by a sliding window processing from the raw data are formally defined as prior knowledge. Then, the defined outlier-type prior knowledge is integrated into any layer of CNNs by a parameter-free attention mechanism. Compared with existing similar methods, the proposal realizes a novel and flexible definition of prior knowledge and achieves deep fusion of prior knowledge and CNNs with low computational cost. The performance of the proposal was evaluated on the Tennessee Eastman process dataset and the real wind turbine blade icing dataset, which indicates that the proposal could not only realize accurate results but also had good model interpretability in terms of achieving high accuracy. The acquisition of outlier-type prior knowledge was discussed and the results demonstrate the effectiveness of the proposed prior knowledge integration method.

2D-to-3D human pose lifting is fundamental for 3D human pose estimation (HPE), for which graph convolutional networks (GCNs) have proven inherently suitable for modeling the human skeletal topology. However, the current GCN-based 3D HPE methods update the node features by aggregating their neighbors’ information without considering the interaction of joints in different joint synergies. Although some studies have proposed importing limb information to learn the movement patterns, the latent synergies among joints, such as maintaining balance are seldom investigated. We propose the Hop-wise GraphFormer with Intragroup Joint Refinement (HopFIR) architecture to tackle the 3D HPE problem. HopFIR mainly consists of a novel hop-wise GraphFormer (HGF) module and an intragroup joint refinement (IJR) module. The HGF module groups the joints by k-hop neighbors and applies a hop-wise transformer-like attention mechanism to these groups to discover latent joint synergies. The IJR module leverages the prior limb information for peripheral joint refinement. Extensive experimental results show that HopFIR outperforms the SOTA methods by a large margin, with a mean per-joint position error (MPJPE) on the Human3.6M dataset of 32.67 mm. We also demonstrate that the state-of-the-art GCN-based methods can benefit from the proposed hop-wise attention mechanism with a significant improvement in performance: SemGCN [42] and MGCN [49] are improved by 8.9% and 4.5%, respectively.

This article proposes a non-intrusive load monitoring (NILM) framework based on a deep convolutional neural network (DCNN) to profile each household applianceon/offstatus and the residential power consumption. It uses only load trajectory, which can overcome the limitations of existing voltage-current trajectory NILM techniques. The DCNN architecture with a load trajectory as the input enables the NILM to directly analyze the electricity consumption at the appliance-level. Meanwhile, the temporal feature transferring procedure improves load monitoring performance and extends its application range include monitoring appliances based on multiple and combined characteristics. Furthermore, the power variation augmentation technique enhances the load signature uniqueness. The fusion of temporal and power variation features provides rich identification information for NILM and improves the accuracy of appliance identification. Experimental results demonstrate that the proposed NILM framework is effective and superior for enhancing demand side management and energy efficiency.

Depression is a mental disorder that causes pain to people and society and is also the largest cause of disability in the world. Intelligent early screening of depression is of great benefit for patients to obtain better diagnoses and treatment. However, previous low-precision detection methods based on facial vision heavily rely on computing resources, which hinders the wide application of automatic depression diagnoses. Therefore, this article proposes an intelligent method for multiscene automatic depression symptom detection, which uses an efficient and convenient cloud-edge collaboration framework combined with negative emotion monitoring and cross-scale facial feature analysis. We deploy a shallow model (EdgeER) on the edge server and a deep model (C-DepressNet) on the cloud server. EdgeER is used to quickly detect negative user emotions and screen user data. C-DepressNet is used to analyze degrees of depression with high precision. The experimental results show that our cloud-edge collaboration framework has superior performance in depression detection accuracy and service response times.

With the application of the Internet of Medical Things (IoMT) in minimally invasive surgery (MIS), surgeons now have a better chance at hard-to-treat cases by carrying out more complicated MIS workflows. However, a scheduled surgical workflow is often required to be updated based on the patient’s internal tissue states. Perioperative complications could occur if in-time adjustments are lacking in the operating rooms when needed. To help manage the uncertainty of live surgical workflows in the IoMT environment, we propose a MIS safety improvement framework. It helps surgeons in predicting surgical workflows with limited MIS video frames by embedding our proposed model GuidedNet. To predict future surgical activities, we first build three isomorphic neural networks to capture the spatiotemporal information. Then, we establish a guidance fusion module to handle the contextual information. It guides the GuidedNet to recognize the surgical stage. Moreover, we build a novel joint loss function to train the GuidedNet to predict the future surgical stage. We evaluate the approach on a large data set that contains 80 cholecystectomy videos (Cholec-80) and compare it with the state of the art. Experiments show that the GuidedNet can assist surgeons in carrying out MIS as well as guide the next stage of surgery for improving surgical safety. Comparing to the state of the art, our approach can obtain better predict accuracy (up to 79%) with less computing resource consumption. The result also shows that our approach has a high application prospect in video classification in other Internet of Things scenarios.

The consistency of interval-valued fuzzy preference relations (IFPRs) is a prerequisite for the application of IFPRs in real problems. To support the application of IFPRs, various types of consistency of IFPRs have been developed. They all satisfy some fixed mathematical conditions under the assumption that decision makers are perfectly rational. In practice, this assumption is usually violated because decision makers generally have bounded rationality. Considering the bounded rationality of decision makers, this article develops a new type of consistency of IFPRs called triangular bounded consistency, which is based on the historical preferences of decision makers. A triangular framework is designed to describe the three IFPRs of any three alternatives. The strict transitivity of IFPRs is defined as the restricted max-max transitivity of IFPRs, which is reconstructed in the triangular framework. In this situation, under the assumption that the preferences of a decision maker are consistent in similar circumstances, the triangular bounded consistency of IFPRs is defined by use of the historical IFPRs of decision makers. Based on the developed consistency, the process of determining the optimal estimations of missing IFPRs in an incomplete IFPR matrix is developed. A problem of selecting suppliers of simulation systems is analyzed using the multicriteria group decision-making (MCGDM) process with the triangular bounded consistency of IFPRs to demonstrate the application of the developed consistency in MCGDM.

Three-dimensional display and virtual reality technology have been applied in minimally invasive surgery to provide doctors with a more immersive surgical experience. One of the most popular systems based on this technology is the Da Vinci surgical robot system. The key to build the in vivo 3-D virtual reality model with a monocular endoscope is an accurate estimation of depth and motion. In this article, a fully unsupervised learning method for depth and motion estimation using the continuous monocular endoscopic video is proposed. After the detection of highlighted regions, EndoMotionNet and EndoDepthNet are designed to estimate ego-motion and depth, respectively. The timing information between consecutive frames is considered with a long short-term memory layer by EndoMotionNet to enhance the accuracy of ego-motion estimation. The estimated depth value of the previous frame is used to estimate the depth of the next frame by EndoDepthNet with a multimode fusion mechanism. The custom loss function is defined to improve the robustness and accuracy of the proposed unsupervised-learning-based method. Experiments with the public datasets verify that the proposed unsupervised-learning-based continuous depth and motion estimation method can effectively improve the accuracy of depth and motion estimation, especially after processing the frame.

Upper gastrointestinal (UGI) cancer has been identified as one of the ten most common causes of cancer deaths globally. UGI cancer screening is critical to improving the survival rate of UGI cancer patients. While many approaches to UGI cancer screening rely on single-modality data such as gastroscope imaging, limited studies have been dedicated to UGI cancer screening exploiting multisource and multimodal medical data, which could potentially lead to improved screening results. In this paper, we propose semantic-level cancer-screening network (SCNET), a framework for UGI cancer screening based on semantic-level multimodal upper gastrointestinal data fusion. Specifically, the proposed SCNET consists of a gastrointestinal image recognition flow and a textual medical record processing flow. High-level features of upper gastrointestinal data are extracted by identifying effective feature channels according to the correlation between the textual features and the spatial structure of the image features. The final screening results are obtained after the data fusion step. The experimental results show that the improvement of our approach over the state-of-the-art ones reached 4.01% in average. The source code of SCNET is available at https://github.com/netflymachine/SCNET.

Automatic acetowhite lesion segmentation in colposcopy images (cervigrams) is essential in assisting gynecologists for the diagnosis of cervical intraepithelial neoplasia grades and cervical cancer. It can also help gynecologists determine the correct lesion areas for further pathological examination. Existing computer-aided diagnosis algorithms show poor segmentation performance because of specular reflections, insufficient training data and the inability to focus on semantically meaningful lesion parts. In this paper, a novel computer-aided diagnosis algorithm is proposed to segment acetowhite lesions in cervigrams automatically. To reduce the interference of specularities on segmentation performance, a specular reflection removal mechanism is presented to detect and inpaint these areas with precision. Moreover, we design a cervigram image classification network to classify pathology results and generate lesion attention maps, which are subsequently leveraged to guide a more accurate lesion segmentation task by the proposed lesion-aware convolutional neural network. We conducted comprehensive experiments to evaluate the proposed approaches on 3045 clinical cervigrams. Our results show that our method outperforms state-of-the-art approaches and achieves better Dice similarity coefficient and Hausdorff Distance values in acetowhite legion segmentation.

Abstract Endoscopy is an important tool for gastric cancer screening. Due to the lack of effective decision support system for endoscopy, the detection of gastric cancer in the clinic is usually with low sensitivity. In this paper, we propose a Genetic Algorithm optimized Neural Network (GAoNN) approach for gastric cancer detection based on endoscopy reports mining. Considering the fact that gastric cancer sensitivity can significantly improve the 5‐year survival rate of patients, both the prediction accuracy and the sensitivity are employed to construct a multiobjective optimization model for enhancing the classification performance of GAoNN. In particular, we extended an effective genetic algorithm Nondominated Sorting Genetic Algorithm II (NSGA‐II) to train a neural network and reduced the complexity in training hyperparameters and improved the efficiency by substituting the computationally intensive stochastic gradient descent (SGD) algorithm in a neural network. Specifically, we designed the novel crossover and mutation operators and modified the nondominated ranking and crowding distance sorting procedures in NSGA‐II for GAoNN. Through testing on 8,546 real‐world endoscopy reports, we show that GAoNN achieves a prediction accuracy up to 83.74%, which is better than several competitors by significantly increasing sensitivity to 83.14%. GAoNN also reduces the training time by 30.94% when compared with conventional SGD‐based training, which indicates the feasibility of GAoNN in clinical practice.

Thyroid nodule has been a common and serious threaten to human health. With the identification and diagnosis of thyroid nodules in the general population, large volumes of examination reports in clinical practice have been accumulated. They provide data basics of analyzing radiologists' behavior of diagnosing thyroid nodules. To conduct data-driven analysis of radiologists' behavior, an experimental framework is designed based on belief rule base, which is essentially a white box for knowledge representation and uncertain reasoning. Under the framework, with 2744 examination reports of thyroid nodules in the period from January 2012 to February 2019 that have been collected from a tertiary hospital located in Hefei, Anhui, China, experimental results are obtained from conducting missing validation, self-validation, and mutual validation. Three principles are then concluded from the results and corresponding analysis. The first is that missing features on some criteria are considered as benign ones by default, the second is that there is generally inconsistency between the recorded features on criteria and the overall diagnosis, and the third is that different radiologists have different diagnostic preferences. These three principles reflect three diagnostic behavioral characteristics of radiologists, namely reliability, inconsistency, and independence. Based on the three principles and radiologists' behavioral characteristics, managerial insights in a general case are concluded to make the findings in this study available in other situations.

With the fast advances in Internet technologies, social networks have become a major platform for social interaction, lifestyle demonstration, and message dissemination. Effective community detection in social networks helps to assess public sentiment, identify community leaders, and produce personalized recommendation. While different community detection approaches have been proposed in the literature, the trust model based detection schemes model user interactions as trust transfer, which helps to capture the implicit relation in the network. Unfortunately, trust model based detection schemes face acold startproblem, i.e., they cannot accurately model newly joined users as these users have few interactions for a duration after joining the network. In this paper, we propose TLCDA, a novel trust model based community detection algorithm. By enhancing the traditional trust computation with inter-node relation strength and similarity in social networks, TLCDA detects communities through coarse-grained K-Mediods clustering. Our evaluation on real social networks shows that the communities detected by TLCDA exhibit superior preference cohesion while satisfying the topology cohesion.

The essential role of the particular families of capacities and the capacity identification methods is to help the decision maker to deal with the exponential complexity inherent in the construction process of the capacity. The 2-additive capacities appear to be the most popular among the particular families of capacities since they permit to model interactions between criteria while preserving simplicity. Besides the preference with respect to the decision criteria, most of the capacity identification methods also need to provide the desired overall evaluations of the decision alternatives in the learning set, which is a time-consuming task for the decision maker. In this paper, we propose some models to identify 2-additive capacities only from a kind of refined preference information with respect to the decision criteria called the multicriteria correlation preference information (MCCPI). The MCCPI is a group of 2-D preference information which can be described and obtained by the refined diamond diagram. The common principle of the proposed models is to minimize the different kinds of deviations between the MCCPI and the most desired 2-additive capacity(ies). A multicriteria decision making example is presented to show the feasibility of the proposed methods, and a 2-D scale of the MCCPI is also introduced in the further discussion of the illustrative example.

Abstract In this paper, we address the cloud service trustworthiness evaluation problem, which in essence is a multi‐attribute decision‐making problem, by proposing a novel evaluation model based on the fuzzy gap measurement and the evidential reasoning approach. There are many sources of uncertainties in the process of cloud service trustworthiness evaluation. In addition to the intrinsic uncertainties, cloud service providers face the problem of discrepant evaluation information given by different users from different perspectives. To address these problems, we develop a novel fuzzy gap evaluation approach to assess cloud service trustworthiness and to provide evaluation values from different perspectives. From the evaluation values, the perception–importance, delivery–importance, and perception–delivery gaps are generated. These three gaps reflect the discrepancy evaluation of cloud service trustworthiness in terms of perception utility, delivery utility, and importance utility, respectively. Finally, the gap measurement of each perspective is represented by a belief structure and aggregated using the evidential reasoning approach to generate final evaluation results for informative and robust decision making. From this hybrid two‐stage evaluation process, cloud service providers can get improvement suggestions from intermediate information derived from the gap measurement, which is the main advantage of this evaluation process.

The rapid growth of computational power demand from scientific, business, and Web applications has led to the emergence of cloud-oriented data centers. These centers use pay-as-you-go execution environments that scale transparently to the user. Load prediction is a significant cost-optimal resource allocation and energy saving approach for a cloud computing environment. Traditional linear or nonlinear prediction models that forecast future load directly from historical information appear less effective. Load classification before prediction is necessary to improve prediction accuracy. In this paper, a novel approach is proposed to forecast the future load for cloud-oriented data centers. First, a hidden Markov model (HMM) based data clustering method is adopted to classify the cloud load. The Bayesian information criterion and Akaike information criterion are employed to automatically determine the optimal HMM model size and cluster numbers. Trained HMMs are then used to identify the most appropriate cluster that possesses the maximum likelihood for current load. With the data from this cluster, a genetic algorithm optimized Elman network is used to forecast future load. Experimental results show that our algorithm outperforms other approaches reported in previous works.