Search

In light of the growing population of individuals with limb disorders, there is an increasing need to address the challenges they face in their daily lives. Existing rehabilitation technologies, often relying on single physiological signals and plagued by poor signal quality, have limitations in their effectiveness. To overcome these constraints, we present E2FNet, a multimodal physiological information fusion network designed for motor intent recognition in individuals with limb disorders. This study involved eight healthy participants who recorded electromyography (EMG) and electroencephalography (EEG) signals during various hand movements. E2FNet utilizes a multiscale convolutional neural network to extract features from EEG and EMG data, focusing on information fusion across different scales. We also introduce a cross-attention mechanism to capture cross-modal information interactions, enhancing EEG and EMG information fusion. Through extensive experiments, E2FNet achieved an impressive 92.08% classification accuracy, and the effectiveness of each module has been verified. Multiscale separable convolution and cross-attention significantly improved EEG and EMG signal fusion, enhancing accuracy and robustness in motion intent recognition. This research promises to enhance the quality of life and independence of individuals with movement disorders, while also advancing the field of rehabilitation robotics and assistive technology.

Blade icing detection is vital for wind turbines in cold climates, as it can prevent revenue loss and power degradation. Many machine learning models have been proposed to improve the detection of blade icing; however, earlier studies do not adequately address these issues due to the dynamics of sensor correlations and the imbalance of blade icing data, resulting in low precision and a high false alarm rate. In this study, we aim to address both of these challenges in order to identify blade icing more accurately. On this premise, we develop a spatial–temporal graph convolutional network (SGCN) that leverages the graph convolutional network for adaptively analyzing the dynamics of sensor correlations and a distance-based classifier to improve imbalanced learning. Experiments on the public UEA time series classification datasets and the real-world wind turbine datasets indicate that SGCN is capable of state-of-the-art accuracy, especially in the case of extremely imbalanced data.

Recently, the safety and reliability of unmanned aerial vehicles (UAVs) have gained increasing interest, and data-driven anomaly detection methods have been widely studied. However, it is still challenging to build an accurate and reliable detector due to the scarcity of labeled data and complex spatial-temporal characteristics of flight data with noises and disturbances. To this end, this article proposes an autoencoder-based unsupervised anomaly detection framework with multivariate flight data without labeled information. Specifically, we designed a new robust spatial-temporal autoencoder (RSTAE) model based on the temporal convolution network (TCN), the convolution neural network (CNN), and the attention mechanism to extract the complicated spatial-temporal correlations in multivariate flight data. Instead of using the traditional mean square error (mse) loss function, a modified loss function based on maximum correntropy criteria (MCC) is introduced to enhance the robustness of our RSTAE model during its training process. To further improve the anomaly detection performance, a dynamic threshold strategy is used. Experimental results on real flight data demonstrate the superior performance of the proposed method compared with several autoencoder-based methods in terms of six evaluation metrics.

Supervisory control and data acquisition (SCADA) data-based wind turbine blade icing detection has been widely studied due to its low cost and easy access. However, SCADA data often present severe class imbalance and thus challenge accurate icing detection. Moreover, since data distribution discrepancy exists in both spatio-temporal features of SCADA data from different wind turbines, the well-trained model has poor classification performance on new turbines. Building new models for different turbines is high-cost and time-consuming. Thus, model cross-turbine generalizability needs improvement. To solve these problems, a cross-turbine icing detection model is proposed based on the spatio-temporal alignment transfer learning method. Specifically, building an attention-based network to extract temporal and spatial features. Then, we apply maximum mean discrepancy (MMD) algorithms on shallow and deep networks to align spatio-temporal features of source and target domains. Besides, a self-adaptive weight (SAW) loss function is employed to address the class imbalance. Finally, we develop a loss weight assignment method based on analyzing the generated loss value variations with the number of training iterations for performance enhancement. The proposed method is evaluated on real SCADA datasets. Experiment results show our proposed transfer learning method significantly improves the model cross-turbine generalizability and classification performance.

Traditional fault diagnosis methods mainly rely on a single sensor signal, such as vibration or generator current signals, thus it often leads to limited diagnosis accuracy, primarily when multiple faults exist at the same time. Considering the electromechanical coupling characteristics of the electromechanical drive system, different sensors usually contain correlated and complementary information, which can improve the diagnosis performance. To this end, this article proposes a current-aided vibration fusion network (CAVFNet) to diagnose different faults in the electromechanical drive system. The raw vibration and current signals are decomposed via wavelet packet decomposition (WPD) into time-frequency matrices representing fault information in different frequency bands. Meanwhile, a current-aided fusion module (CAFM) is designed to achieve sufficient fusion of cross-modal information. Reweighting the fused features in spatial dimensions uses the excitation maps extracted from the current signals. Finally, an adaptive decision-level fusion strategy is developed to integrate information from different branches. Experimental results on both datasets demonstrate our proposed method has strong robustness and high diagnostic performance. The core code for this project is available at: https://github.com/LKLaii/project-CAVFNet.

Longitudinal questionnaire data were used to measure antidepressant treatment outcome.Machine learning models were used to predict outcome from electronic health records.The gradient boosting decision tree model achieved the best predictive results.Diagnostic codes and baseline severity were strong predictors of treatment outcome. Longitudinal questionnaire data were used to measure antidepressant treatment outcome. Machine learning models were used to predict outcome from electronic health records. The gradient boosting decision tree model achieved the best predictive results. Diagnostic codes and baseline severity were strong predictors of treatment outcome. To evaluate if a machine learning approach can accurately predict antidepressant treatment outcome using electronic health records (EHRs) from patients with depression. This study examined 808 patients with depression at a New York City‐based outpatient mental health clinic between June 13, 2016 and June 22, 2020. Antidepressant treatment outcome was defined based on trend in depression symptom severity over time and was categorized as either “Recovering” or “Worsening” (i.e., non‐Recovering), measured by the slope of individual‐level Patient Health Questionnaire‐9 (PHQ‐9) score trajectory spanning 6 months following treatment initiation. A patient was designated as “Recovering” if the slope is less than 0 and as “Worsening” if the slope was no less than 0. Multiple machine learning (ML) models including L2 norm regularized Logistic Regression, Naive Bayes, Random Forest, and Gradient Boosting Decision Tree (GBDT) were used to predict treatment outcome based on additional data from EHRs, including demographics and diagnoses. Shapley Additive Explanations were applied to identify the most important predictors. The GBDT achieved the best results of predicting “Recovering” (AUC: 0.7654 ± 0.0227; precision: 0.6002 ± 0.0215; recall: 0.5131 ± 0.0336). When excluding patients with low PHQ‐9 scores (<10) at baseline, the results of predicting “Recovering” (AUC: 0.7254 ± 0.0218; precision: 0.5392 ± 0.0437; recall: 0.4431 ± 0.0513) were obtained. Prior diagnosis of anxiety, psychotherapy, recurrent depression, and baseline depression symptom severity were strong predictors. The results demonstrate the potential utility of using ML in longitudinal EHRs to predict antidepressant treatment outcome. Our predictive tool holds the promise to accelerate personalized medical management in patients with psychiatric illnesses.

Intelligent fault diagnosis based on data-driven models has shown excellent effectiveness in recent years, significantly promoting diagnosis technology. However, data-driven models require large amounts of labeled data for supervised training, and the quantity of labeled data in industrial engineering is limited. Few-label learning is still a significant challenge for engineering applications in industry. To address this issue, this article proposes an impulse information-enhanced semi-supervised learning (IESSL) model to improve the performance of few-label fault diagnosis. We design a novel encoder module based on fault-related impulse knowledge to construct unsupervised features (UFs) from large amounts of data without labels. UFs are utilized to constrain the decision space of the backbone network. Then, a few labeled data are used to fine-tune the backbone networks to make the more accurate mapping between the fault types and decision space. The experimental results on two datasets demonstrate the effectiveness of our proposed method.

Intelligent fault diagnosis of wind turbines (WTs) has gained considerable attention in recent years, and deep learning-based methods have shown excellent performance in modeling multivariate supervisory control and data acquisition (SCADA) data. However, these methods are performed in Euclidean space and cannot fully exploit the complex coupling relationship and potential fault propagation between different sensors of WTs. To address this issue, we transform SCADA data into graph structure data with sensors (as nodes) and topological connections (as edges) between sensors to represent complex interactive information. Specifically, we propose a multiview graph network with an interactive shared fusion model (MVGNet) for fault diagnosis of WTs. First, a multiview graph construction framework considering multiple measures is constructed to fully explore the fault-related features. In particular, we propose a directed causal graph for SCADA data modeling for the first time, which is beneficial for mining the causal relationships among multiple variables of WTs and providing complementary information with the correlation graph to enrich the feature representations. Then, a multihead graph attention network (GAT) with residual connection is designed to perform multiview graph representation learning. Furthermore, to take full advantage of the complementarity and consistency of multiview graphs, we propose a multigraph interactive shared fusion (MISF) module, which can adaptively perform shared learning among multiple views, provide more interactive operations, and effectively fuse multigraph features. Finally, the global graph embedding is generated for the fault diagnosis task. Experimental results on four real-world WTs datasets show that MVGNet has reliable and superior diagnosis performance compared with state-of-the-art methods.

Blades icing will seriously affect the performance of wind turbines with respect to power loss and dynamic load increase. Blades icing detection technique becomes necessary to advance de-icing maintenance. Extracting effective features from supervisory control and data acquisition (SCADA) data has become a challenging task during operating conditions under icing. Current research work lacks for the integration of physical information and insufficient analysis of icing dynamic evolution in the process of feature extraction. In order to eliminate these deficiencies, a multilevel feature extraction model combined with physical information and intelligent algorithm (MFT-PI) is proposed in this article. First, features that characterize the severity of icing are extracted by power characteristic curve technology and feature selection technology in the transient feature extraction level. These techniques are driven by the information of power loss and icing mass growth. Second, a fully connected neural network integrated with triplet loss function is designed for extracting fusion feature, which considers the similarity of data distribution between early-icing stage and non-icing stage. Finally, the proposed feature extraction model is evaluated in two icing wind turbines. Results show that fusion features extracted from the proposed model is more stable, reliable, and accurate by comparing the performance of other feature groups in different classifiers.

Current-based fault diagnosis has become a promising solution for electromechanical systems due to the low cost and easy access. However, most of the existing studies require additional signal processing or diagnostic expertise to preprocess the signals due to the effects of the fundamental component and electrical noise. It is challenging to extract effective fault-related features from the raw current signals. To this end, this article proposes a differential-augmented current feature learning network named CurrentNet for drive system fault diagnosis, which is an end-to-end model. First, a differential-augmented strategy based on the raw current signal is introduced to generate complementary current representations. Furthermore, a multi-information interaction module (MIIM) is designed to adaptively capture complementary shared information between the original and enhanced signals through a parallel mechanism. Also, efficient channel attention is used to reduce the complexity of the model. Our proposed method is experimentally evaluated on two datasets and presents obvious superiority over existing fault diagnosis methods.

The recent development of imaging and sequencing technologies enables systematic advances in the clinical study of lung cancer. Meanwhile, the human mind is limited in effectively handling and fully utilizing the accumulation of such enormous amounts of data. Machine learning-based approaches play a critical role in integrating and analyzing these large and complex datasets, which have extensively characterized lung cancer through the use of different perspectives from these accrued data. In this review, we provide an overview of machine learning-based approaches that strengthen the varying aspects of lung cancer diagnosis and therapy, including early detection, auxiliary diagnosis, prognosis prediction, and immunotherapypractice. Moreover, we highlight the challenges and opportunities for future applications of machine learning in lung cancer.

The recent development of imaging and sequencing technologies enables systematic advances in the clinical study of lung cancer. Meanwhile, the human mind is limited in effectively handling and fully utilizing the accumulation of such enormous amounts of data. Machine learning-based approaches play a critical role in integrating and analyzing these large and complex datasets, which have extensively characterized lung cancer through the use of different perspectives from these accrued data. In this review, we provide an overview of machine learning-based approaches that strengthen the varying aspects of lung cancer diagnosis and therapy, including early detection, auxiliary diagnosis, prognosis prediction, and immunotherapypractice. Moreover, we highlight the challenges and opportunities for future applications of machine learning in lung cancer.

The rich semantic representation and sophisticated structure definition of the HL7 Fast Healthcare Interoperability Resources (FHIR) specification require relatively great efforts to understand and utilize. The objective of our study is to design, develop, and evaluate an open-source and user-friendly visualization interface for exploring the FHIR specification. We prototyped an interactive visualization tool for navigating and manipulating the FHIR core resources, profiles, and extensions. The utility of the tool was evaluated using evaluation metrics mainly focusing on its interactive mechanism and content expressiveness. We demonstrated that the visualization techniques are helpful for navigating the HL7 FHIR specification and aiding its profiling.

Identification of non-coding RNAs (ncRNAs) has been significantly improved over the past decade. On the other hand, semantic annotation of ncRNA data is facing critical challenges due to the lack of a comprehensive ontology to serve as common data elements and data exchange standards in the field. We developed the Non-Coding RNA Ontology (NCRO) to handle this situation. By providing a formally defined ncRNA controlled vocabulary, the NCRO aims to fill a specific and highly needed niche in semantic annotation of large amounts of ncRNA biological and clinical data.

In this paper, polymer nanocomposite films containing gold nanorods (AuNRs) and poly(2-vinyl pyridine) (P2VP) have been investigated for their structure–optical property relationship. Using transmission electron microscopy (TEM), the assembly of AuNRs (7.9 nm × 28.4 nm) grafted with a P2VP brush in P2VP films is examined as a function of the AuNR volume fraction ØAuNRsand film thickness h. For h∼ 40 nm, AuNRs are confined to align parallel to the film and uniformly dispersed at low ØAuNRs. Upon increasing ØAuNRs, nanorods form discrete aggregates containing mainly side-by-side arrays due to depletion–attraction forces. For ØAuNRs= 2.7%, AuNRs assemble into a 2D network where the discrete aggregates are connected by end-to-end linked nanorods. As ØAuNRsfurther increases, the polymer-rich regions of the network fill in with nanorods and rod overlap is observed. Monte Carlo simulations capture the experimentally observed morphologies. The effect of film thickness is investigated at ØAuNRs= 2.7%, where thicker films (40 and 70 nm) show a dense array of percolated nanorods and thinner films (20 nm) exhibit mainly isolated nanorods. Using Rutherford backscattering spectrometry (RBS), the AuNRs are observed to segregate near the substrate during spin-casting. Optically, the longitudinal surface plasmon resonance (LSPR) peaks are correlated with the local orientation of the AuNRs, where side-by-side and end-to-end alignments induce blue and red shifts, respectively. The LSPR undergoes a red shift up to 51 nm as ØAuNRsincreases from 1.6 to 2.7%. These studies indicate that the optical properties of polymer nanocomposite films containing gold nanorods can be fine-tuned by changing ØAuNRsand h. These results are broadly applicable and provide guidelines for dispersing other functional nanoparticles, such as quantum dots and carbon nanotubes.

The formation of a nanostructured interpenetrating polymer network (IPN) via electropolymerization is described. The electro‐copolymerization of alternate layer‐by‐layer (LbL) self‐assembled polyelectrolytes with thiophene and carbazole pendant monomers was demonstrated facilitating IPN formation of π‐conjugated polymers or conjugated polymer network (CPN) films. UV–Vis spectroscopy, QCM, and ellipsometry confirmed linear nanostructured LbL film growth. Electrochemical crosslinking by cyclic voltammetry (CV) manifested highly regular peak current increases with successive cycles. A quantitative correlation of the LbL layer number with the cathodic charge and scan rate was observed. Electrochemical impedance analysis confirmed CPN film formation and the change in capacitance behavior.

The self‐assembled monolayer formation of a series of thiophene dendron thiols with different generations and alkyl chain lengths onto gold surfaces is studied. Average film thicknesses and dielectric constants were calculated from SPR using a two‐subphase differential method. The in situ adsorption of the thiophene dendron thiols was monitored by SPR kinetics measurements and fitted using an empirical three‐step model. Electrochemical redox molecular probing experiments provided additional insight into surface coverage, which showed different abilities to block electron transfer at the monolayer/electrode interfaces. The surface coverage of the films was also estimated by QCM measurements and supported by static WCA measurements.

We report a detailed and quantitative study on the supramolecular complexation of amine-functionalized polyamidoamine (PAMAM) dendrimer G 4-NH 2with carboxylic acid terminal dendrons containing peripheral electroactive carbazole groups of different generations (G 0COOH, G 1COOH, and G 2COOH). While the focus is on a detailed understanding and mechanism of complex formation, subsequent electrochemical oxidation of the dendron surfmers resulted in the formation of nanoringstructures electrodeposited on the conducting substrate. Complexation was confirmed by NMR, UV−vis, and IR measurements. Critical micelle concentration (CMC), atomic force microscopy (AFM), and X-ray photoelectron spectroscopy (XPS) studies revealed that the ringlike structures were formed during the equilibrium−decomplexation stage and that the electrochemical process did not destroy the complex but rather stabilized it. The different generations of the dendrons provided various structures and complex formation efficacy. This type of template polymerization combined with electrochemically anodic oxidation has not been previously reported.

Investigations on optical and dielectric properties of hybrid ultrathin layer-by-layer (LbL) films of gold nanoparticles (AuNPs) and polyelectrolytes under different pH conditions resulted in surface plasmon resonance (SPR) signal enhancement under attenuated total reflection (ATR) spectroscopy conditions. Theoretical considerations on the basis of the Maxwell−Garnett theory were made to compare with experimental results. LbL films with different layer architectures were fabricated from AuNPs and two polyelectrolytes:  poly(allylamine hydrochloride) (PAH) and poly(styrene sulfonate) (PSS) on Au and Ag thin film substrates for ATR. UV−vis spectroscopy and SPR spectroscopy were applied to study LbL film growth and monitor SPR shift upon pH switching. The (PAH/AuNPs)xmultilayers showed an interesting dual-responsive SPR change as a function of pH and distance between AuNP layers and metal film. The addition of (PAH/PSS)ylayers was found to act as an effective cushion to enhance this SPR response due to the significant swelling/shrinking of the film. In the case of (PAH/PSS)y(PAH/AuNPs)x, both theoretical calculations and experimental results showed that the SPR response can either (a) move toward lower incident angle with a sharp peak shape upon swelling in pH 2 or (b) shift to a higher angle with a broadened peak shape after contraction in pH 10. This effect was found to be opposite to what is expected from LbL films made up of the polyelectrolytes alone. Moreover, increasing the distance between AuNPs and the metal films also decouples this enhancement effect. A two-wavelength experiment (red and green lasers) was used to quantitatively demonstrate this SPR response to pH switching. Finally, SPR imaging was employed to monitor the SPR change with pH switching between 2 and 10 on the (PAH/PSS)4(PAH/AuNPs)2film. Thus, on the basis of the SPR spectroscopic and SPR imaging pH response, a reproducible and stable sensing system can be successfully fabricated with these films.