In cancer research, identifying cancer subtypes and evaluating prognosis are crucial for personalized diagnosis and treatment of cancer. With the advancement of high-throughput sequencing technologies, multi-omics data has become essential for cancer classification and prognostic analysis. By integrating deep learning techniques, it is possible to more accurately identify cancer subtypes, providing a robust basis for personalized treatment of cancer patients. In this study, we propose a convolutional autoencoder prognostic model incorporating a channel attention mechanism (CA-CAE). The model utilizes multi-omics data to predict survival-associated cancer subtypes and identify prognostic genes. We applied CA-CAE to multiple cancer types, successfully identifying subtypes in 15 distinct cancer types and revealing significant survival differences among these subtypes. Moreover, compared to traditional statistical methods and other deep learning approaches, CA-CAE demonstrated superior performance in predicting survival outcomes.

This article presents a decentralized dynamic event-sampled control (ESC) strategy for multi-unit power systems (MUPSs) subject to mismatched interconnections. Initially, with the introduction of a dynamic event-sampling mechanism, the decentralized ESC problem of MUPSs is converted into a set of event-sampled optimal control problems of auxiliary subsystems. It is demonstrated that all the solutions of the event-sampled Hamilton-Jacobi-Bellman equations (ES-HJBEs) arising in these optimal control problems together constitute the decentralized dynamic ESC law. Then, in order to solve the ES-HJBEs, the critic neural networks (CNNs) within the adaptive dynamic programming framework are constructed. The CNNs’ weights are updated via simultaneously using the gradient descent method and concurrent learning. The benefits of such a tuning rule are that it not only makes full use of state data (including historical and instantaneous state data) but also no longer requires the persistence of excitation condition. Moreover, uniform ultimate boundedness of the closed-loop auxiliary subsystem states and the CNNs’ weight estimation errors are proved via Lyapunov approach. Finally, simulations of a three-unit power system are given to validate the present decentralized dynamic ESC scheme.

Oil–gas–water three-phase flow process exhibits randomness and transience. Under varying flow conditions and environmental factors, the process presents a dual challenge in state monitoring: data distribution discrepancies between source and target domains, along with the presence of unknown states in the target domain. Therefore, a cross-domain open-set state monitoring method based on multigrained adversarial network with hierarchical attribute causality (MANHAC) is presented in this work. A domain adversarial architecture is designed to distinguish unknown from known target flow states, in which the weighted thresholding method based on information entropy can adjust the decision boundary adaptively. Besides, to mitigate conditional distribution mismatch and negative transfer caused by forced global domain alignment, MANHAC introduces hierarchical causal attributes for describing different flow states, where attribute features are influenced by upstream cause attribute and optimized through competition with multiple discriminators. The proposed MANHAC employs a dual-alignment mechanism to achieve both global and fine-grained domain adaptation, which can effectively carry out cross-domain open set state identification. More importantly, it can describe unknown states by attribute vectors, providing more meaningful monitoring information. Dynamic experiments of three-phase flow demonstrate its effectiveness and superiority.

Transfer learning has been widely applied to intelligent fault diagnosis to address the challenge of insufficient labeled data. However, the efficacy of existing semi-supervised domain adaptation (SSDA) methods is often constrained by their critical dependence on pseudo-label quality and the adoption of indiscriminate global alignment strategies. To address the negative transfer induced by these limitations, a deep gated network (DGN) for targeted transfer learning is proposed in this article. First, an anchor-guided manifold clustering (AGMC) method is developed to generate high-quality pseudo-labels by exploiting both the local manifold structure and anchor supervision in the target domain. Subsequently, a feature extractor is constructed to learn discriminative representations from the source domain, integrate high-quality supervisory information from the target domain, and impose constraints on the feature space. Furthermore, a gated domain alignment strategy is designed to achieve precise class-level transfer. This strategy incorporates an integrated gating mechanism to selectively filter out domain-specific features while employing the local maximum mean discrepancy (LMMD) to align the conditional distributions across domains. Finally, the effectiveness and superiority of the proposed method are validated through transfer experiments on both cross-machine bearing and cross-condition two-phase flow datasets.

The accurate prediction of bearing remaining useful life (RUL) is crucial for industrial safety and cost reduction. However, existing deep learning methods often lack interpretability and struggle to integrate local-global and structural-temporal information, limiting their effectiveness. To address this issue, an observable and controllable RUL prediction method is proposed, which visualizes the bearing degradation process through spatio-temporal dynamic graph (STDG), and fully extracts structural information and temporal features to achieve precise RUL prediction. First, canonical variable fluctuation analysis (CVFA) is employed to maximize past-future data correlation, enabling information distillation by analyzing static deviations and dynamic fluctuations. Second, restraint graph attention network (RGAT) is proposed to extract local structural information, generating dynamic factors that feedback into the spatio-temporal static graph (STSG) to construct the STDG for observation and control of experts. Thirdly, to address degradation variability and long sequence dependency, the reversible instance normalization Transformer (RevIN-Former) explores the temporal evolution of the STDG from a global perspective, achieving accurate RUL prediction. Finally, the effectiveness of the proposed method is validated through three accelerated life experiments on bearings, which reduces the RUL prediction error by 88% compared to other methods.

Recently, transfer learning (TL) approaches have been extensively applied in industrial cross-domain fault diagnosis, most of which depend on the consistency assumption of the source and target fault categories. In practice, it is common to utilize multiple source domains for transfer learning, but each of them may not include all fault categories in the target domain, which are referred to as incomplete multisource domains. For the challenge of fault diagnosis under incomplete multisource domains, a cross-domain bilateral transfer learning (CDBTL) method is proposed in this article. First, a cross-domain bilateral transfer strategy is developed, where the source and target domains are reconstructed from each other and their distribution differences are reduced by minimizing the reconstruction error to avoid negative transfer. Then, for the source domain with label information, CDBTL maximizes the between-class distance of different fault categories and minimizes the within-class distance of the same fault category to ensure the discriminative nature of its feature representation. Afterwards, the common projection matrix is learned through the mutual cooperation of projection matrices between different incomplete source domains and target domain to compensate for the missing fault categories in a single source domain. The key to discriminate CDBTL from many exiting TL algorithms is that it relaxes the restriction of consistent fault categories in the source and target domains, and skillfully integrates the knowledge of multiple incomplete source domains. Extensive experiments on Tennessee Eastman process demonstrate the superiority of CDBTL in solving cross-domain fault diagnosis problem, whose accuracy is averagely improved by 17.99% compared with eleven existing algorithms.Note to Practitioners—The changing industrial operating modes (domains) may result in different data distributions and fault categories between the historical mode (source domain) and current mode (target domain). Traditional machine learning methods usually fail to diagnose under the above domain and category inconsistencies. The key work in this article is to develop a cross-domain bilateral transfer learning (CDBTL) algorithm to realize cross-domain fault diagnosis under incomplete multisource domains. The proposed algorithm can avoid negative transfer while reducing inter-domain differences, and utilize the mutual cooperation of multiple source domains to fully cover the fault categories in the target domain. The constructed CDBTL model can be combined with various classifiers, and the learned classifiers can be directly applied to fault diagnosis in advanced manufacturing industry under incomplete multisource domains.

Gas–liquid two-phase flow is a complex dynamic and nonlinear process that is widely encountered in many process industries. Accurate flow state identification is crucial for ensuring operation safety and economic benefits. However, obtaining training samples for certain flow states can be difficult due to safety requirements and high costs. Therefore, a zero-shot learning (ZSL) based flow state identification strategy is proposed from the perspective of attribute description and attribute transfer, in which the common attribute space is constructed by the semantic description of flow state categories. The attribute-relevant features are extracted by the proposed supervised deep slow and steady feature analysis (SD-S$^{\mathbf{2}}$FA) under the supervision of attributes. In SD-S$^{\mathbf{2}}$FA, an extended Siamese network is designed to extract slow and steady features (S$^{\mathbf{2}}$Fs), in which three 1D convolutional neural networks (1D-CNN) represent the nonlinear feature embedding function, and the Siamese architecture can capture the long-term temporal coherence. Since the state attributes are shared by all the flow states, the identification for unseen flow states can be realized by attribute prediction and attribute transfer. The effectiveness and superiority of the proposed method is demonstrated through the gas–liquid two-phase flow experiment.

The fault detection capability of traditional data-driven process monitoring methods is highly dependent on the quality of process data. However, affected by measurement noise, harsh operation scenarios and other factors, the process data are inevitably contaminated by uncertainty in real processes. In this article, a vertices packaging-based interval independent component analysis (VP-I2CA) method is proposed to monitor the uncertain non-Gaussian processes. First, a variable bandwidth-kernel density estimation-based measurement error estimation method is developed to describe the uncertainty-contaminated process data in interval form using limited reliable data samples. Then, VP-I2CA is developed to estimate the demixing matrix based on hypermatrices constructed by vertices encoding, which includes all possible combinations between the bounds of interval data by explicitly considering the existence of uncertainty. In order to reduce computational complexity, the idea of data packaging is introduced to represent the hyper-independent components in interval form with a series of values by packaging all possible feature information hidden in uncertain process data. Afterwards, four monitoring statistics are constructed to monitor the systematic and nonsystematic parts of process operation variation. The proposed algorithm is verified in both a six-variable numerical simulation system and a continuous stirred tank reactor system.

Gas-water two-phase flow has typically stable flow statuses and constantly changing transition flow statuses. Accurate identification and real-time monitoring of flow status are conducive to the in-depth study of two-phase flow and the safe operation of industrial process. A monitoring strategy based on sparse local Fisher discriminant analysis (SLFDA) is proposed in this article. First, multisensor signals are obtained to reflect flow process information. Second, the least absolute shrinkage and selection operator is used to find the sparse discriminant directions to determine the key variables relevant to the flow process from multiple sensor signals. Then, the weight coefficient matrixes of SLFDA keep the original structure of the same flow status data and make the data of different flow statuses more separated, which distinguish different flow statuses to the maximum extent. Finally, two monitoring indexes including the discriminant index and the stability index are established to analyze the dynamic flow process, which enable a concurrent monitoring of both flow evolution and instability to realize fine-scale description of flow process. SLFDA can monitor various flow statuses through only one projection discriminant matrix by transforming high-dimensional signals into features representing the flow characteristics, which avoids model traversal and improves monitoring efficiency. Further study of flow status features provides meaningful physical interpretation and in-depth process analysis with consideration of actual flow process. The application on the data of gas-water two-phase flow in horizontal pipe demonstrates the feasibility and efficacy of the method.

Closed-loop control is commonly used in industrial processes to track setpoints or regulate process disturbances. Process dynamics resulting from closed-loop control are reflected in data mainly in two aspects, namely serial correlation and variation of response speed. Concurrent analysis of both aspects from data has not been fully investigated in the literature. In this work, a combined strategy of canonical variate analysis and slow feature analysis is proposed to monitor process dynamics resulting from closed-loop control by exploring both serial correlations and variation speed of process data. First, the canonical subspaces reflecting serial correlation are modeled by maximizing correlation between the past and future values of the process data. Then, both the serially correlated canonical subspace and its residual subspace are further explored to extract the slow features, which are representations of process variation speed. The proposed method provides a meaningful physical interpretation and in-depth process analysis with considerations of process dynamics under closed-loop control. Besides, it provides a concurrent monitoring of both process faults and operating condition deviations, resulting in fine-scale identification of different operation statuses. To demonstrate the feasibility and effectiveness, the proposed strategy is tested in a simulated typical chemical process under closed-loop control, namely the three-phase flow process.

Since the turn of the century, hundreds of image watermarking algorithms based on histogram features have been reported. However, all the existing watermark embedding was based on binary embedding, and they only store two situations on two continuous bins, i.e., {0,1}. In this paper, we improve the existing embedding algorithms and propose a novel high-capacity watermarking algorithm by using two bins histogram modification. The new algorithm is based on the ternary numeral system, and watermark information is divided into three cases {0,1,2}. Specifically, we extract the histogram of the cover image first, then select the appropriate embedding range by two predefined thresholds, and last form the bin groups contained reasonable number of pixels. In the generation of watermark, we transfer the watermark information into a digital string W = w1w2... wt, wi∈ (0,1,2),i = 1,2,...,t. In the embedding algorithm, if wi= 2, we will adjust the height of two continuous bins, and let b'/a' ≥ T; if wi= 1, let a'/b' ≥ T; if wi= 0, let |a'-b'| ≤ 1, here a' and b' represent the height of the front and back bin, and T is the threshold. Experimental results show that the embedding capacity of the proposed watermarking algorithm is 60% higher than that of the existing algorithm. In addition, the new scheme can resist traditional geometric attacks.

Social media is rocking the world in recent year, which makes modeling social media contents important. However, the heterogeneity of social media data is the main constraint. This paper focuses on inferring emotions from large-scale social media data. Tweets on social media platform, always containing heterogeneous information from different combinations of modalities, are utilized to construct a cross-media dataset. How to integrate cross-media information and solve the problem of modality deficiency are main challenges. To address those challenges, this paper proposes a Cross-media Auto-Encoder(CAE) to infer emotions on cross-media data, and CAE is designed to reconstruct missing modalities and integrate heterogeneous representations. In our experiments, We employ a dataset of 226,113 tweets to infer emotions of tweets, and our method outperforms several machine learning methods (+11.11% in terms of F1-measure). Feature contribution analysis also verifies the importance of adopting cross-media features.

Tumour heterogeneity describes the coexistence of divergent tumour cell clones within tumours, which is often caused by underlying epigenetic changes. DNA methylation is commonly regarded as a significant regulator that differs across cells and tissues. In this study, we comprehensively reviewed research progress on estimating of tumour heterogeneity. Bioinformatics-based analysis of DNA methylation has revealed the evolutionary relationships between breast cancer cell lines and tissues. Further analysis of the DNA methylation profiles in 33 breast cancer-related cell lines identified cell line-specific methylation patterns. Next, we reviewed the computational methods in inferring clonal evolution of tumours from different perspectives and then proposed a deconvolution strategy for modelling cell subclonal populations dynamics in breast cancer tissues based on DNA methylation. Further analysis of simulated cancer tissues and real cell lines revealed that this approach exhibits satisfactory performance and relative stability in estimating the composition and proportions of cellular subpopulations. The application of this strategy to breast cancer individuals of the Cancer Genome Atlas's identified different cellular subpopulations with distinct molecular phenotypes. Moreover, the current and potential future applications of this deconvolution strategy to clinical breast cancer research are discussed, and emphasis was placed on the DNA methylation-based recognition of intra-tumour heterogeneity. The wide use of these methods for estimating heterogeneity to further clinical cohorts will improve our understanding of neoplastic progression and the design of therapeutic interventions for treating breast cancer and other malignancies.

Situation awareness may be inferred from user context such as body posture transition and location data. Smartphones and smart homes incorporate sensors that can record this information without significant inconvenience to the user. Algorithms were developed to classify activity postures to infer current situations; and to measure user's physical location, in order to provide context that assists such interpretation. Location was detected using a subarea-mapping algorithm; activity classification was performed using a hierarchical algorithm with backward reasoning; and falls were detected using fused multiple contexts (current posture, posture transition, location, and heart rate) based on two models: “certain fall” and “possible fall.” The approaches were evaluated on nine volunteers using a smartphone, which provided accelerometer and orientation data, and a radio frequency identification network deployed at an indoor environment. Experimental results illustrated falls detection sensitivity of 94.7% and specificity of 85.7%. By providing appropriate context the robustness of situation recognition algorithms can be enhanced.

Three RFID reader based network deployment algorithms (grid-covering, diagonal and mixed) were evaluated in this paper. Experimental results show that the grid-covering method can be used to minimize hardware costs, but it leads to many indeterminate positions. The diagonal method can be used to solve the indeterminate problem, however increases the number of readers, especially in a large tracking field. The mixed algorithm can be used to avoid the indeterminate issue and also has the minimum reader number when deployed in a large space. However, it is not suitable for a small tracking field. An optimal deployment algorithm is selected from these three algorithms according to the environmental conditions and the localization requirement. In addition, an optimal RFID reader network deployment combined with a subarea-mapping algorithm can be used to minimize the hardware costs while improving the fine-grained indoor localization accuracy.

Extracting emotions from images has attracted much interest, in particular with the rapid development of social networks. The emotional impact is very important for understanding the intrinsic meanings of images. Despite many studies having been done, most existing methods focus on image content, but ignore the emotion of the user who published the image. One interesting question is: How does social effect correlate with the emotion expressed in an image? Specifically, can we leverage friends interactions (e.g., discussions) related to an image to help extract the emotions? In this paper, we formally formalize the problem and propose a novel emotion learning method by jointly modeling images posted by social users and comments added by their friends. One advantage of the model is that it can distinguish those comments that are closely related to the emotion expression for an image from the other irrelevant ones. Experiments on an open Flickr dataset show that the proposed model can significantly improve (+37.4% by F1) the accuracy for inferring user emotions. More interestingly, we found that half of the improvements are due to interactions between 1.0% of the closest friends.

In this paper we report on the operation and results of a pilot trial on the use of Science-Fiction Prototyping (SFP) as a means to motivate students to engage with STEM and language learning courses. In particular we describe two case studies. The first was conducted in Shijiazhuang University, China, and involved approximately 102 students following a course aimed at improving their English language abilities for computer science applications. The second case study concerned the use of micro science-fiction prototypes, in the form of Twitter-fiction, as a means of motivating pre-university students to take up STEM studies and careers. Finally, the paper concludes by describing future directions of this work.

Feature selection is a key step for activity classification applications. Feature selection selects the most relevant features and considers how to use each of the selected features in the most suitable format. This paper proposes an efficient feature selection method that organizes multiple subsets of features in a multilayer, rather than utilizing all selected features together as one large feature set. The proposed method was evaluated by 13 subjects (aged from 23 to 50) in a lab environment. The experimental results illustrate that the large number of features (3 vs. 7 features) are not associated with high classification accuracy using a single Support Vector Machine (SVM) model (61.3% vs. 44.7%). However, the accuracy was improved significantly (83.1% vs. 44.7%), when the selected 7 features were organized as 3 subsets and used to classify 10 postures (9 motionless with 1 motion) in 3 layers via a hierarchical algorithm, which combined a rule-based algorithm with 3 independent SVM models.

This paper presents details of a convenient and unobtrusive system for monitoring daily activities. A smart phone equipped with an embedded 3D-accelerometer was worn on the belt for the purposes of data recording. Once collected the data was processed to identify 6 activities offline (walking, posture transition, gentle motion, standing, sitting and lying). The processing technique adopted a novel hierarchical classification. In the first instance, rule-based reasoning is used to discriminate between motion and motionless activities. Following this the classification process utilizes two multiclass SVM (support vector machines) classifiers to classify the motion and motionless activities, respectively. The classifiers were trained on data from one subject and tested on 10 subjects. The experiments demonstrate that the hierarchical method can reduce misclassification between motion and motionless activities. The average accuracy was improved compared with using a single classifier by using this classification method (82.8% vs. 63.8%), and is important for providing appropriate feedback in free living applications.

The problem of approximate optimal rejection to sinusoidal disturbances with zero steady-state error (ZSSE) for nonlinear systems is considered. Based on the internal model principle, a disturbance compensator is constructed through which the plant model with external disturbances is transformed into an augmented nonlinear system without disturbances. Introducing a sensitivity parameter and expanding power series around it, the original optimal control problem is transformed into a series of linear two-point boundary value (TPBV) problems. The obtained optimal control law consists of a linear analytic term which is obtained by solving a Riccati matrix equation and a nonlinear compensatory term in form of series which is obtained by a recursive algorithm. By intercepting a finite sum of the compensation series, an approximate optimal control law is obtained. A numerical simulation shows that the algorithm is easily implemented and has a fast convergence rate. The designed controller has more robustness.