With 6G-enabled Intelligent Internet of Vehicles (IIoV) generating massive amounts of sensory data, traditional deep learning models struggle to capture long-range relationships across different sensor types while preserving privacy. This paper proposes DT-Trans, a privacy-preserving federated learning framework that combines Digital Twin technology with Vision Transformers. Our framework first trains a global perception model on synthetic digital twin data, then fine-tunes it efficiently for real-world vehicles. By grouping vehicles with similar driving patterns and allowing them to collaboratively train personalized model components, DT-Trans achieves significant accuracy improvements while maintaining data privacy. The Twin-Enhanced Vision Transformer (TE-ViT) is introduced as the global perception backbone; it is pre-trained on massive synthetic DT data and then fine-tuned via parameter-efficient LoRA adapters to bridge the domain gap between virtual and physical worlds. The Cluster-Enhanced Decoupled PFL (CD-PFL-Trans) algorithm splits each TE-ViT into (i) a shared Transformer encoder (base layer) and (ii) client-specific Transformer decoder heads (personalized layer). Hierarchical clustering on decoder parameters groups clients with similar traffic patterns, enabling group-wise aggregation without exchanging raw sensory data. DT-Trans outperforms CNN-based FedAvg/FedPer by 9.3%-16.2% mAP on V&PKITTI perception tasks and up to 42.8% accuracy improvement on CINIC-10 classification under severe heterogeneity, while reducing on-device FLOPs by 34 % via Transformer sparsity techniques. Our work advances Transformer architectures for scalable, privacy-preserving perception in IIoV.

The proliferation of IoT devices has limited the efficiency of heterogeneous data communication in distributed environments and increased security risks. Balancing scalability, efficiency and data privacy in IoT transaction systems becomes critical, and intent-based networks enable optimal configuration with minimal intervention. To optimize the network management environment, we propose a three-stage execution scheme for blockchain cross-shard transactions, which combined with a timeout rollback mechanism ensures atomicity and reduces latency. In addition, we design a fragment-based consensus protocol utilizing a verifiable random function, which improves the consensus efficiency through the randomness of committee member selection. In order to enhance system security, we introduce a reputation evaluation mechanism and a malicious node detection method based on normalized entropy. The mechanism dynamically adjusts the reputation value of a node according to its performance in the consensus process, so that high-reputation nodes can play a greater role in the consensus and detect malicious nodes in the network accordingly. By embedding this mechanism into a network management framework based on users’ intention, it can accurately realize users’ expectations for network performance optimization, security enhancement and efficient operation. Experiments show that our scheme not only improves communication efficiency, but also enhances the security of sharded transactions, effectively matching users’ high-level intentions for network scalability, efficiency, and data privacy.

Cross-domain medical image segmentation has gained increasing interest for its potential to reduce annotation efforts and improve clinical generalization capabilities. Domain adaptation aims to tackle the domain shift that appears in different image modalities. In cross-domain segmentation, generative models often suffer from limited accuracy due to their lack of domain-specific representations. Besides, many transfer learning approaches rely on additional manual annotations for supervision, emerging paradigms such as Unsupervised Domain Adaptation (UDA) facilitate effective knowledge transfer even when labels in the target domain are entirely absent. In this study, we propose a novel Point-cloud Structure Constrained Unsupervised Domain Adaptation (PSC-UDA) framework based on a Contour-Aware Segmentation (CAS) model with a 3D contour point cloud to bridge the domain gaps appearing in cross-site and cross-domain medical images. The CAS model distills the domain-invariant kidney structure from image texture to distinguish the point cloud and characterize the kidney contour in a coarse-to-fine way. With point-to-voxel self-learning on 3D structure constraints, the proposed PSC-UDA framework addresses visual domain shift, adapting discriminative information of the kidney from the labeled source domain (CT) to the unlabeled target domain (CT/MRI), so that it realizes precise cross-domain kidney segmentation with limited labels. Experimental results prove that the proposed method outperforms the generative UDA methods and the source-free methods on three cross-domain kidney segmentation datasets, outperforming even without a target domain adaptation strategy. The source code is available at https://github.com/zzs95/PSC-UDA .

Pulmonary nodule (PN) is the typical radiological indicator of early lung cancer. Compared with CT and LDCT, PN detection based on Chest X-ray (CXR) images is more costeffective and involves lower radiation exposure. However, since the limited contrast of CXR images, PN screening always relies on manual detection by radiologists, which is time-consuming and labor-intensive. In this paper, we propose an automatical method for PN detection in CXR. As the small size of the PN region, we reformulate the PN detection as the point localization, which transfer the traditional coordinates regression in CNN to a heatmap regression by a dual-channel Gaussian modeling. For improving the detection accuracy of tiny PN, we construct a global localization network to realize coarse PN regression, which is cascaded by a local optimization network to further refine the results with high-resolution CXR image patches. The ablation and comparison experimental results indicate that our method can achieve superior and satisfying performance on spatial localization errors and detection metrics, respectively.

Abstract At present, the mainstream mode of machine learning algorithms is the data‐driven method, which mainly relies on the self‐learning ability of deep neural networks and continuously evolving models in data‐driven training. However, the pure data‐driven method has some critical problems, such as high data collection cost, poor interpretability and easy to be be disturbed by noise. Although the knowledge‐driven method has high stability, it lacks self‐learning and evolution ability in the face of comprehensive and complex problems. In recent years, the convergence of data and domain knowledge has combined the advantages of both learning paradigms. One typical way is to embed domain knowledge into the data‐driven model to improve the interpretability of the model, and then use the self‐learning ability of the data‐driven model to explore knowledge, and continuously iterate the domain knowledge to form a closed loop. The data‐knowledge dual‐driven methods have brought transformative innovations in machine learning. This review first introduced the advantages and necessity of the data‐knowledge dual‐driven model in the field of artificial intelligence. Then, the applications of the data‐knowledge dual‐driven model in the smart marine field were introduced. Finally, the challenges and trends of the data‐knowledge dual‐driven artificial intelligence are anticipated.

ABSTRACT Driven by the dual pressures of rapid global population growth and escalating climate change, there is a growing demand for real‐time monitoring of crop nitrogen levels to support precision agriculture. This necessity has catalysed the integration of crop modelling techniques with remote sensing technologies. Addressing challenges such as multi‐source remote sensing data heterogeneity and limited generalisation in nitrogen inversion models for cotton canopies, this paper designs a novel inversion framework based on the assimilation of diverse remote sensing sources and mechanistic crop models. Firstly, this paper employed spectral resampling techniques, fuzzy logic for uncertainty quantification, and Pearson correlation analysis to harmonise differences in spectral characteristics and spatial resolution between Sentinel‐2A and Landsat 8 imagery, ultimately identifying eight nitrogen‐sensitive features. Subsequently, a multi‐scale feature enhancement module was developed to improve representational richness. Additionally, the paper employed a satellite image fusion module, which effectively reduced data heterogeneity errors by 12.7% across sources. Building on this, a hybrid GNN‐EnKF model was proposed. GNN was used to establish spatial neighbourhood dependencies, while EnKF dynamically adjusted the parameters within the WOFOST crop model. This approach successfully fuses data‐driven learning with physically based modelling. Experimental evaluations revealed that the proposed architecture attained a mAP of 95.83%, outperforming baseline models such as ResNet18 (83.92%) and Transformer (92.84%), demonstrating robust adaptability in complex agricultural settings. In conclusion, the framework presented in this paper offers a high‐accuracy nitrogen monitoring solution tailored for precision farming, and provides strong data support for cotton nitrogen deficiency and additional fertilisation.

As a key economic crop, jujube’s external morphology directly affects quality grading and market value. However, traditional inspection methods relying on manual sampling or 2D image analysis suffer from inefficiency and limited feature characterization, particularly in quantifying complex geometric traits such as irregular wrinkles and localized depressions on jujube surfaces. Existing 3D reconstruction techniques have been applied in agricultural product inspection but face challenges in widespread adoption due to high costs and low resolution. This study proposes a single-view high-resolution RGB 3D reconstruction method for jujube based on generative artificial intelligence. Specifically, we designed a two-stage single-view 3D reconstruction framework and modified the cross-attention layers in the U-shaped network architecture of a stable video diffusion mode to meet the requirements of high-resolution and clear texture reconstruction for jujube. Additionally, we improved training and inference efficiency through parallel computing and achieved automation integration with Unmanned Agents. The proposed method successfully reconstructed 3D models from single-view 1,024 1,024 RGB images of jujube. Our model achieves a PSNR of 23.52 and an SSIM of 0.86 on the public dataset.This approach provides a low-cost, high-precision 3D digital solution for non-destructive phenotyping of jujube, offering practical value for advancing intelligent sorting and quality evaluation in agricultural production.

With the deepening integration of the Internet of Things (IoT) and Artificial Intelligence (AI), intelligent warehousing systems are increasingly confronted with the challenges of achieving high-precision navigation and task scheduling in dynamic environments. Although existing research has made notable strides in simultaneous localization and mapping (SLAM) and task scheduling, persistent issues–such as point cloud noise interference in dynamic scenes, inefficiencies in computational resource allocation, and insufficient multi-sensor collaboration–continue to constrain system performance. To address these challenges, this study proposes an AI-driven task scheduling SLAM framework named KORS designed to enhance navigational robustness and scheduling efficiency in dynamic warehousing environments. This study proposes the KCPoint model to achieve precise segmentation and elimination of dynamic point clouds. Building upon the PointNet++ architecture, KCPoint integrates K-Nearest Neighbors Enhanced Farthest Point Sampling (KFPS) and a Convolutional Block Attention Module (CBAM) to enhance feature extraction. In addition, a task scheduling mechanism is introduced to address the dynamic allocation of computational resources within vehicular networks. Relative to FAST-LIO2, the KORS system reduces absolute pose error (APE) by 31.32% and improves computational efficiency by 17.88% on the NCD and NCLT datasets. Furthermore, compared to conventional methods based on particle swarm optimization and genetic algorithms, the task scheduling algorithm achieves comparable decision-making benefits while reducing single-decision latency by over 42

This research proposes a solution integrating AIGC technology with optimized object detection networks to address challenges in brain tumor identification under the context of Medical Industry 5.0. First, to mitigate data scarcity, DCGAN is employed to augment the Br35H dataset by generating high-quality synthetic samples, enhancing model generalization. Second, a hierarchical feature-enhanced EfficientDet (EfficientDet-HFE) model is developed by combining FPN and EfficientDet architectures, fusing high-level semantic information with low-level spatial details to optimize feature transmission pathways. Additionally, the SimAM is introduced, integrated with a global-local feature optimization strategy to construct a recursive attention module. As BiFPN iterates progressively, the capacity for key region feature expression and the efficiency of multi-scale feature extraction are significantly enhanced. In response to the computational resource constraints of medical devices, LAMP techniques are applied to compress the network structure. With the assistance of fine-tuning strategies, the model parameters are reduced to 32.20 MB while preserving detection performance. Experimental results demonstrate that this method achieves 92.25% recall, 93.36% precision, 92.80% F1 Score, and 94.99% mAP on the augmented dataset, validating its efficacy in tumor detection. This research offers a lightweight, IoT-compatible solution for brain tumor detection, promoting the integration of AI-driven diagnostics into Healthcare Industry 5.0 ecosystems.

Many experiments have shown that circular RNA (circRNA) can act as biomarkers for complex diseases and play significant regulatory roles in multiple pathological processes. However, most circRNA-disease associations remain unknown, and discovering these associations through biological experimental approach is expensive and time-consuming. Taking into account the shortcomings of current methods, we introduce a new collaborative framework that utilizes multi-heterogeneous graphs, along with variational graph auto-encoders (VGAE) to predict associations between circRNA and diseases. First, we build multi-similarity networks using various biological attributes of circRNA and diseases, and integrate these similarity networks. Two subnetworks are constructed from association matrix and combined similarity network, which included a circRNA-based network and a disease-based network. We then employ random walk with restart and Singular Value Decomposition, for feature extraction from the similarity matrix. Finally, we use collaborative framework to predict the circRNAdisease association scores based on the two subnetworks. We integrate the two score matrices to obtain a final prediction scoring matrix. Using 5-fold cross-validation on the CircR2Disease dataset, our model achieved an AUC score of 0.9828 and an AUPR score of 0.9820. Additionally, among the top 30 highest-scoring circRNA-disease association pairs, 26 associations have already been validated. Our model shows strong performance and can accurately predict associations between circRNA and diseases, according to experimental results.

Circular RNA (circRNA) is an RNA molecule that plays an important role in both pathology and physiology. Accurate identification of associations between circRNAs and diseases is crucial for further physiology research. However, verifying the circRNA-disease associations (CDA) through biological experimental methods is time-consuming. Here, we propose a novel method combines Non-negative Matrix Factorization (NMF) and Adversarially Regularized Variational Graph Autoencoder (ARVGA) to accurately predict CDA. Our model first fuses multi-source information in order to build circRNA similarity, disease similarity and circRNA-disease association matrices. Thereby our model constructs graphs for circRNA and disease respectively and optimize them using a K-means clustering algorithm. We obtain linear features using NMF and non-linear features using ARVGA. Finally, an Extremely Randomized Trees classifier is employed to predict CDA. On the gold standard dataset CircR2Disease, our model achieved a prediction accuracy of 94.8% and an AUC of 0.984 under 5-fold cross-validation. In case study, 19 of top 20 predicted circRNAs associated with Hepatocellular Carcinoma were confirmed in relevant literature. Furthermore, ablation experiment, classifier experiment and independent datasets test fully demonstrate the effectiveness and robustness of our model.

Recent studies have indicated that circular RNAs (circRNAs) play a significant role in the diagnosis and treatment of disease. However, the prediction of associations between circRNAs and diseases using conventional biological methods is constrained by numerous factors. In this study, we proposed a novel computational model called EELMCDA that combines evolutionary ensemble learning (EEL) approach and matrix feature decomposition method to predict potential circRNA-disease associations. The model firstly integrates circRNA function information, disease semantic information, and circRNA and disease gaussian interaction profile kernel (GIPK) information into an integrated matrix and constructed the corresponding feature matrix, then uses the matrix feature decomposition algorithm to obtain its important feature, and finally adopted evolutionary ensemble learning module to predict circRNA-disease associations. The average accuracy of the EELMCDA model by 5-fold cross-validation on CircR2Disease, CircAtlasv2.0, Circ2Disease, and CircRNADisease datasets were 92.40%, 92.90%, 88.91%, and 90.74%, respectively. Moreover, in case studies, the 21 of the top 30 circRNA-disease pairs with the highest EELMCDA scores were validated in recent literatures. These results further demonstrate the effectiveness of EELMCDA in predicting circRNA-disease associations.

Connections between circular RNAs (circRNAs) and microRNAs (miRNAs) assume a pivotal position in the onset, evolution, diagnosis and treatment of diseases and tumors. Selecting the most potential circRNA-related miRNAs and taking advantage of them as the biological markers or drug targets could be conducive to dealing with complex human diseases through preventive strategies, diagnostic procedures and therapeutic approaches. Compared to traditional biological experiments, leveraging computational models to integrate diverse biological data in order to infer potential associations proves to be a more efficient and cost-effective approach. This paper developed a model of Convolutional Autoencoder for CircRNA-MiRNA Associations (CA-CMA) prediction. Initially, this model merged the natural language characteristics of the circRNA and miRNA sequence with the features of circRNA-miRNA interactions. Subsequently, it utilized all circRNA-miRNA pairs to construct a molecular association network, which was then fine-tuned by labeled samples to optimize the network parameters. Finally, the prediction outcome is obtained by utilizing the deep neural networks classifier. This model innovatively combines the likelihood objective that preserves the neighborhood through optimization, to learn the continuous feature representation of words and preserve the spatial information of two-dimensional signals. During the process of 5-fold cross-validation, CA-CMA exhibited exceptional performance compared to numerous prior computational approaches, as evidenced by its mean area under the receiver operating characteristic curve of 0.9138 and a minimal SD of 0.0024. Furthermore, recent literature has confirmed the accuracy of 25 out of the top 30 circRNA-miRNA pairs identified with the highest CA-CMA scores during case studies. The results of these experiments highlight the robustness and versatility of our model.

Emerging clinical evidence suggests that sophisticated associations with circular ribonucleic acids (RNAs) (circRNAs) and microRNAs (miRNAs) are a critical regulatory factor of various pathological processes and play a critical role in most intricate human diseases. Nonetheless, the above correlations via wet experiments are error-prone and labor-intensive, and the underlying novel circRNA-miRNA association (CMA) has been validated by numerous existing computational methods that rely only on single correlation data. Considering the inadequacy of existing machine learning models, we propose a new model named BGF-CMAP, which combines the gradient boosting decision tree with natural language processing and graph embedding methods to infer associations between circRNAs and miRNAs. Specifically, BGF-CMAP extracts sequence attribute features and interaction behavior features by Word2vec and two homogeneous graph embedding algorithms, large-scale information network embedding and graph factorization, respectively. Multitudinous comprehensive experimental analysis revealed that BGF-CMAP successfully predicted the complex relationship between circRNAs and miRNAs with an accuracy of 82.90% and an area under receiver operating characteristic of 0.9075. Furthermore, 23 of the top 30 miRNA-associated circRNAs of the studies on data were confirmed in relevant experiences, showing that the BGF-CMAP model is superior to others. BGF-CMAP can serve as a helpful model to provide a scientific theoretical basis for the study of CMA prediction.

With the continuous development of artificial intelligence, more and more human–computer interaction (HCI) applications have begun to appear in the field of education. This article investigates the relationship between the HCI and cognitive load and academic emotion by comparing the effect of classroom discussion in the smart classroom with that of traditional classroom discussion and classroom discussion that focuses too much on learning interest. By using classroom test questionnaire and interview questionnaire, the experiment counted the test scores and questionnaire satisfaction of participants in different classroom discussions, and obtained the learning effect and the emotional state and satisfaction degree of learners under different classroom discussions. It is concluded that the artificial intelligence in the HCI of smart classroom should always take the amount of cognitive load of learners as the core, improve the academic emotion of learners as the auxiliary, reduce the cognitive load of learners as much as possible, and improve the academic emotion of learners within a reasonable range, so as to improve the learning effect. It also provides methods and suggestions for how to reduce learners’ cognitive load and how to reasonably improve their academic emotions in smart classroom.

In recent years, the combination of hyperspectral imagery and deep learning has been widely used in agricultural Artificial Intelligence of Things (AIoT), such as agricultural product quality assessment and crop disease detection. However, this often comes at the cost of substantial computational power and energy consumption. In this paper, we focused on data-efficient green computing for low-carbon jujube moisture content detection. First, in order to compress the hyperspectral images capacity, a spectral selection algorithm based on swarm intelligence was proposed to screen necessary and sensitive spectral dimensions. Then, a spectral reconstruction model was established to realize the selected hyperspectral bands reconstruction from RGB image, aiming to reduce the high cost of hyperspectral imaging. Finally, a model based on the fusion of spectral data and reconstructed image was constructed to realize efficient jujube moisture content detection. The experimental results show that the 10 feature bands screened by the proposed selection method can adequately characterize the water information of jujube, and the proposed reconstruction model outperforms other works with the MRAE of 0.1635. The carbon emissions of our proposed reconstruction model are significantly lower than other methods. Further, the spectral-image fusion model achieves satisfactory detection result of jujube moisture content, with the RMSE of 0.0082. In summary, the proposed spectral selection, reconstruction, and detection methods can achieve high precision while reducing carbon emissions, which have important guidance for low-carbon and sustainable agricultural AIoT applications.

With the rapid development of information technology, social network has become an indispensable part of daily life. People have been able to get news from all over the world through social networks for a long time. People spend more time online than they do in real life. However, the information we get in the world of social network is not purely benign. Due to the development of artificial intelligence technology, more and more tampered media information appears in social networks, some for entertainment, while others become the dark side of social networks, of which the most harmful is to people in the media tamper. For fake news and misinformation caused by media tampering, we need to trace the source and clearly distinguish the truth from the manipulated. This article proposes an image media forgery classification method of multiorder attribute nodes. First, we use different methods to extract the edge, texture, grayscale, and color attributes of the image. Second, according to the characteristics of different attributes, we calculate the first-order entropy of edge attributes, the second-order entropy of texture attributes, local entropy of grayscale, and color properties. Finally, we represent each image with some nodes and build a graph convolutional network (GCN) to classify real and fake images. Experimental results on mainstream media manipulation datasets show that our method is the state-of-the-art compared with similar methods.

Health disparities among marginalized populations with lower socioeconomic status significantly impact the fairness and effectiveness of healthcare delivery. The increasing integration of artificial intelligence (AI) into healthcare presents an opportunity to address these inequalities, provided that AI models are free from bias. This paper aims to address the bias challenges by population disparities within healthcare systems, existing in the presentation of and development of algorithms, leading to inequitable medical implementation for conditions such as pulmonary embolism (PE) prognosis. In this study, we explore the diverse bias in healthcare systems, which highlights the demand for a holistic framework to reducing bias by complementary aggregation. By leveraging de-biasing deep survival prediction models, we propose a framework that disentangles identifiable information from images, text reports, and clinical variables to mitigate potential biases within multimodal datasets. Our study offers several advantages over traditional clinical-based survival prediction methods, including richer survival-related characteristics and bias-complementary predicted results. By improving the robustness of survival analysis through this framework, we aim to benefit patients, clinicians, and researchers by enhancing fairness and accuracy in healthcare AI systems.

The usual active learning is to sample unlabeled set by designing efficient sample information evaluation algorithms. However, information redundancy between candidate sets is often overlooked. This can cause similar data to be labeled repeatedly, producing ineffective gains for the model. In this paper, we proposed an Unsupervised Redundant Feature Elimination Active Learning module (URFEAL), which utilizes the information feature coincidence of the unlabeled set to eliminate information redundant data, thus guaranteeing the validity of each candidate data. URFEAL consists of feature clusterer and eliminator. The feature clusterer computes class boundaries based on feature densities to discretize each class of the candidate set, and the eliminator judges data similarity by overlapping degree to eliminate redundant data features. Furthermore, we propose an anti-noise sampling strategy Outlier Feature Elimination (OFE) in URFEAL to filter mislabeled sets for relabeling in the data sampling stage. We extensively evaluate our method by image classification and perform experimental validation on CIFAR-10, CIFAR-100 and CALTECH-101. The experimental results show that the improvements we make are especially significant for most existing active learning algorithms in the low data stage, which demonstrates the effectiveness and generality of URFEAL.

Generative adversarial networks have shown impressive results in the modeling of movies and games, but what if such powerful image generation capability is used to harm the Multimedia? The face replacement methods represented by Deepfakes are becoming a threat to everyone, so the development of image authenticity detection methods has become a top priority. For achieving accurate detection resistant to compression effects, we propose a weighted complementary dual-stream detection method. First, to alleviate the influence of image compression on manipulation detection, we propose the concept of pixel-wise saliency invariance. We map fake images onto saliency maps via Quaternary Fourier Transform, which discovers the invariant properties of image phase spectra on different compressions. Meanwhile, to capture boundary traces more easily, we propose the concept of pixel-wise detail enhancement. We apply Bilateral Filtering to preserve the texture edges of fake images and amplify the fake boundaries. Finally, to take full advantage of the two proposed concepts, a weighted complementary dual-stream network is designed as a classifier to fuse features and identify real and fake. On different benchmarks like FaceForensics++ (FF++), Celeb-DF, and DFDC, the experimental results show that the proposed method has the average best detection accuracy compared to existing methods.

In this paper, we propose a wavelet-aware transformer network (WATNet) for multi-contrast knee MRI super-resolution. Unlike conventional image domain-based super-resolution methods that can not explicitly model the lost high-frequency information, our WATNet endeavors to adaptively fuse the complementary frequency information of the multi-contrast image in the wavelet domain and further refine it in the image domain. The proposed WATNet consists of the multi-scale wavelet transformation (MSWT) module, wavelet-aware transformer (WAT) module, and reconstruction (Rec) module. Specifically, the MSWT module learns to transform the MR image to multi-scale wavelet domain features by the wavelet transformation. The WAT module can adaptively search and transfer similar wavelet domain reference information to the low-resolution one. The Rec module can restore high-quality images in the image domain. To further capture more high-frequency details, we also design the wavelet-based high-frequency loss. The qualitative and quantitative experimental results indicate that our proposed WATNet outperforms most state-of-the-art methods.

Internet of Things (IoT) has extended its coverage to various spatial domains and has established interconnection to serve widespread applications of a larger spatial scale. Such IoT is called the space–air–ground–ocean-integrated network (SAGOI-Net), which consists of multiple battery-powered heterogeneous devices. Hence, energy efficiency is the key point of SAGOI-Net to be stably operated for a long time without manual maintenance. This article proposes a novel scheme of energy-efficient autonomous and decentralized SAGOI-Net establishment using an intelligent autonomous underwater glider (AUG) to serve marine applications. The proposed SAGOI-Net is energy efficient because the energy consumption is minimized by: 1) employing nonpropeller-driven AUG; 2) navigating AUG under water without acoustic sensor or extra energy-consuming vision sensors; and 3) equipping the self-navigation (SN) system based on lightweight neural network model to save the energy consumption of onboard computing resource. Moreover, assuming the AUG navigation problem as time-series regression, the proposed scheme designs SAGOI-Net to be autonomous and decentralized with the aid of lightweight long short-term memory (LSTM) network-based SN (SN-LSTM) system of AUG. The lightweight SN-LSTM model is trained end-to-end on dynamically modeled AUG motion information along with numerically modeled ocean environment data to quantitatively analyze the impact of the ocean environment on AUG. The simulation results demonstrate a superior performance of the AUG SN along with energy efficiency of the proposed SAGOI-Net.

As one of the commonly used vehicles for underwater detection, underwater robots are facing a series of problems. The real underwater environment is large-scale, complex, real-time and dynamic, and many unknown obstacles may exist in the underwater environment. Under such complex conditions and lack of prior knowledge, the existing path planning methods are difficult to plan, therefore they cannot effectively meet the actual demands. In response to these problems, a three-dimensional marine environment including multiple obstacles is established with the real ocean current data in this paper, which is consistent with the actual application scenarios. Then, we propose an N-step Priority Double DQN (NPDDQN) path planning algorithm, which potently realizes obstacle avoidance in the complex environment. In addition, this study proposes an experience screening mechanism, which screens the explored positive experience and improves its reuse rate, thus efficiently improving the algorithm stability in the dynamic environment. This paper verifies the better performance of reinforcement learning compared with a variety of traditional methods in three-dimensional underwater path planning. Underwater robots based on the proposed method have good autonomy and stability, which provides a new method for path planning of underwater robots.Note to Practitioners—The goal of this study is to provide a new solution for obstacle avoidance in path planning of underwater robots, which is consistent with the dynamic and real-time demands of the real environment. Existing underwater path planning researches lack a consistent environment with the actual application, and therefore we firstly construct a three-dimensional ocean environment with real ocean current data to provide support for the algorithms. Additionally, most of the algorithms are pre-planning methods or require long-time calculation, and there is little research on obstacle avoidance. In the face of obstacle changes, underwater robots with poor adaptability will cause performance decline and even economic losses. The proposed algorithm learns through interaction with the environment, and therefore it does not require any prior experience, and has good adaptability as well as fast inference speed. Especially, in the dynamic environment, algorithm performance is difficult to guarantee due to less positive experience in exploration. The proposed experience screening mechanism improves the stability of the algorithm, so that the underwater robot maintains stable performance in different dynamic environments.

Considering the efficiency and security of healthcare data processing, indiscriminate data collection, annotation, and transmission are unwise. In this article, we propose the normalized double entropy (NDE) method to assess image data quality in the form of metatask. In specific, the probability entropy and distance entropy are both adopted and normalized to evaluate the data quality. The experimental results show the stable ability of the NDE to distinguish good and bad data in terms of information contribution. Furthermore, the model's diagnostic performances driven by selected good and bad data are compared, and a clear gap exists between them under the premise of the same amount of data. Screening 70% of the dataset can achieve almost the same accuracy as that based on all data. This article focuses on healthcare data quality and data redundancy and provides a practical evaluation tool to facilitate the identification and collection of valuable data, which is beneficial to improve efficiency and protect cybersecurity in healthcare systems.

Segmentation of liver from CT scans is essential in computer-aided liver disease diagnosis and treatment. However, the 2DCNN ignores the 3D context, and the 3DCNN suffers from numerous learnable parameters and high computational cost. In order to overcome this limitation, we propose an Attentive Context-Enhanced Network (AC-E Network) consisting of 1) an attentive context encoding module (ACEM) that can be integrated into the 2D backbone to extract 3D context without a sharp increase in the number of learnable parameters; 2) a dual segmentation branch including complemental loss making the network attend to both the liver region and boundary so that getting the segmented liver surface with high accuracy. Extensive experiments on the LiTS and the 3D-IRCADb datasets demonstrate that our method outperforms existing approaches and is competitive to the state-of-the-art 2D-3D hybrid method on the equilibrium of the segmentation precision and the number of model parameters.

Circular RNA (circRNA), a novel endogenous noncoding RNA molecule with a closed-loop structure, can be used as a biomarker for many complex human diseases. Determining the relationship between circRNAs and diseases helps us to understand the diagnosis, treatment, and pathogenesis of complex diseases, which plays a critical role in clinical research. Nevertheless, the discovery of new circRNA-disease associations by wet-lab methods is not only time-consuming and costly but also randomized and blinded, which is also limited to small-scale studies. Thus, there is an urgent need to establish efficient and reliable computational methods to infer potential circRNA-disease associations on a large scale to effectively reduce costs and save time, and avoid high false-positive rates. In this paper, we propose a novel computational method for predicting circRNA-disease association based on the Similarity Assessing Graph Convolution Network (SAGCN) algorithm, which combines the multi-source similarity network constructed by circRNA and disease. Firstly, we fuse the multi-source similarity information of circRNAs and diseases and construct the multi-source similarity network respectively. Then we use the SAGCN algorithm to extract the hidden feature representations of circRNAs and diseases efficiently and objectively in the way of measuring the similarity between different nodes in the network. Finally, the obtained high-level features of circRNAs and diseases are fed to the multilayer perceptron (MLP) classifier for accurate prediction. Using the 5-fold cross-validation method, the AUC scores of the four SAGCN algorithms, on the benchmark circR2Disease dataset are 93.30%, 92.98%, 92.22% and 91.94%, respectively. Furthermore, case studies further validated that the proposed model was supported by biological experiments, and 25 of the top 30 circRNA-disease associations with the highest scores were confirmed by recent literature. Based on these reliable results, it can be anticipated that the proposed model can be used as an effective computational tool to predict circRNA-disease associations and can provide the most promising candidates for biological experiments.

Accurate ovarian tumor differentiation is a challenging task where the benign and malignant tumors share similar T1C and T2WI MRI appearances. Therefore, it is necessary to leverage additional multi-modal data, e.g., the age, CA125level, and other clinical information, which are helpful but rarely exploited. In this paper, we propose a dynamic fusion network that can adaptively make full use of multi-modal data, including MRI and clinical information, to realize precise ovarian tumor differentiation. Specifically, we design a dynamic nonlinear module (D-Non-L module) on the top of the image representation. The D-Non-L module is formulated as an iterative nonlinear projection parameterized by the learned features of the patient-wise clinical information. With the help of this module, the interaction between clinical features and image features could be achieved to adaptively improve the discrimination of visual representations. Moreover, we construct a dual-path-based architecture to fully exploit the complementary information from T1C and T2WI MRIs. Extensive experimental results on the locally organized ovarian tumor dataset demonstrate that our methods are superior to the single-modal and single-path-based methods. And the proposed dynamic non-linear module obtains the best performance compared with other multi-modal fusion strategies.

Emerging evidence suggests that circular RNA (circRNA) is an important regulator of a variety of pathological processes and serves as a promising biomarker for many complex human diseases. Nevertheless, there are relatively few known circRNA-disease associations, and uncovering new circRNA-disease associations by wet-lab methods is time consuming and costly. Considering the limitations of existing computational methods, we propose a novel approach named MNMDCDA, which combines high-order graph convolutional networks (high-order GCNs) and deep neural networks to infer associations between circRNAs and diseases. Firstly, we computed different biological attribute information of circRNA and disease separately and used them to construct multiple multi-source similarity networks. Then, we used the high-order GCN algorithm to learn feature embedding representations with high-order mixed neighborhood information of circRNA and disease from the constructed multi-source similarity networks, respectively. Finally, the deep neural network classifier was implemented to predict associations of circRNAs with diseases. The MNMDCDA model obtained AUC scores of 95.16%, 94.53%, 89.80% and 91.83% on four benchmark datasets, i.e., CircR2Disease, CircAtlas v2.0, Circ2Disease and CircRNADisease, respectively, using the 5-fold cross-validation approach. Furthermore, 25 of the top 30 circRNA-disease pairs with the best scores of MNMDCDA in the case study were validated by recent literature. Numerous experimental results indicate that MNMDCDA can be used as an effective computational tool to predict circRNA-disease associations and can provide the most promising candidates for biological experiments.

BACKGROUND: Self-interacting proteins (SIPs), two or more copies of the protein that can interact with each other expressed by one gene, play a central role in the regulation of most living cells and cellular functions. Although numerous SIPs data can be provided by using high-throughput experimental techniques, there are still several shortcomings such as in time-consuming, costly, inefficient, and inherently high in false-positive rates, for the experimental identification of SIPs even nowadays. Therefore, it is more and more significant how to develop efficient and accurate automatic approaches as a supplement of experimental methods for assisting and accelerating the study of predicting SIPs from protein sequence information. RESULTS: In this paper, we present a novel framework, termed GLCM-WSRC (gray level co-occurrence matrix-weighted sparse representation based classification), for predicting SIPs automatically based on protein evolutionary information from protein primary sequences. More specifically, we firstly convert the protein sequence into Position Specific Scoring Matrix (PSSM) containing protein sequence evolutionary information, exploiting the Position Specific Iterated BLAST (PSI-BLAST) tool. Secondly, using an efficient feature extraction approach, i.e., GLCM, we extract abstract salient and invariant feature vectors from the PSSM, and then perform a pre-processing operation, the adaptive synthetic (ADASYN) technique, to balance the SIPs dataset to generate new feature vectors for classification. Finally, we employ an efficient and reliable WSRC model to identify SIPs according to the known information of self-interacting and non-interacting proteins. CONCLUSIONS: Extensive experimental results show that the proposed approach exhibits high prediction performance with 98.10% accuracy on the yeast dataset, and 91.51% accuracy on the human dataset, which further reveals that the proposed model could be a useful tool for large-scale self-interacting protein prediction and other bioinformatics tasks detection in the future.

The path planning of the autonomous underwater vehicle (AUV) has shown great potential in various Internet of Underwater Things (IoUT) applications. Although considerable efforts had been made, prior studies are confronted with some limitations. For one thing, existing work only uses the ocean current simulation model without introducing real ocean information, having not been supported by real data. For another, traditional path planning algorithms have strong environment dependence and lack flexibility: once the environment changes, they need to be remodeled and replanned. To overcome these challenges, this article proposes comprehensive ocean information D3QN (COID), an AUV path planning scheme exploiting comprehensive ocean information and reinforcement learning (RL), which consists of three steps. First, we introduce the comprehensive real ocean data, including weather, temperature, thermohaline, current, etc., and apply them into the regional ocean modeling system to generated reliable ocean current. Next, through well-designed state transition function and reward function, we build a 3-D grid model of ocean environment for RL. Furthermore, based on the framework of the double dueling deep$Q$network (D3QN), COID integrates local ocean current and position features to provide state input and uses priority sampling to accelerate network convergence. The performance of COID has been evaluated and proved by numerical results, which demonstrate efficient path planning and high flexibility for expansion into different ocean environments.

Lots of Deepfake videos are circulating on the Internet, which not only damages the personal rights of the forged individual, but also pollutes the web environment. What’s worse, it may trigger public opinion and endanger national security. Therefore, it is urgent to fight deep forgery. Most of the current forgery detection algorithms are based on convolutional neural networks to learn the feature differences between forged and real frames from big data. In this paper, from the perspective of image generation, we simulate the forgery process based on image generation and explore possible trace of forgery. We propose a multi-scale self-texture attention Generative Network(MSTA-Net) to track the potential texture trace in image generation process and eliminate the interference of deep forgery post-processing. Firstly, a generator with encoder-decoder is to disassemble images and performed trace generation, then we merge the generated trace image and the original map, which is input into the classifier with Resnet as the backbone. Secondly, the self-texture attention mechanism(STA) is proposed as the skip connection between the encoder and the decoder, which significantly enhances the texture characteristics in the image disassembly process and assists the generation of texture trace. Finally, we propose a loss function called Prob-tuple loss restricted by classification probability to amend the generation of forgery trace directly. To verify the performance of the MSTA-Net, we design different experiments to verify the feasibility and advancement of the method. Experimental results show that the proposed method performs well on deep forged databases represented by FaceForensics++, Celeb-DF, Deeperforensics and DFDC, and some results are reaching the state-of-the-art.

Android is a growing target for malicious software (malware) because of its popularity and functionality. Malware poses a serious threat to users’ privacy, money, equipment and file integrity. A series of data-driven malware detection methods were proposed. However, there exist two key challenges for these methods: (1) how to learn effective feature representation from raw data; (2) how to reduce the dependence on the prior knowledge or human labors in feature learning. Inspired by the success of deep learning methods in the feature representation learning community, we propose a malware detection framework which starts with learning rich-features by a novel unsupervised feature learning algorithm Merged Sparse Auto-Encoder (MSAE). In order to extract more compact and discriminative feature from the rich-features to further boost the malware detection capability, a hybrid deep network learning algorithm Stacked Hybrid Learning MSAE and SDAE (SHLMD) is established by further incorporating a classical deep learning method Stacked Denoising Auto-encoders (SDAE). After that, we feed the feature learned by MSAE and SHLMD respectively to classification algorithms, e.g., Support Vector Machine (SVM) or K-NearestNeighbor (KNN), to train a malware detection model. Evaluation results on two real-world datasets demonstrate that SHLMD achieves 94.46 and 90.57 percent accuracy respectively, which outperforms the classical unsupervised feature representation learning Sparse Auto-encoder (SAE). MSAE performs similarly to SAE. SHLMD can further improve the performance of MSAE and the supervised fine-tuned method SDAE. Besides, we compare the performance of our methods with that of state-of-the-art detection approaches, including classical deep-learning-based methods. Extensive experiments show that our proposed methods are effective enough to detect Android malware.

Identification of drug–target interactions (DTIs) is critical for discovering potential target protein candidates for new drugs. However, traditional experimental methods have limitations in discovering DTIs. They are time-consuming, tedious, and expensive, and often suffer from high false-positive rates and false-negative rates. Therefore, using computational methods to predict DTIs has received extensive attention from many researchers in recent years. To address this issue, in this paper, an effective prediction model is presented which is based on the information of drug molecular structure data and protein sequence data. It performs prediction with the following procedures. First, we transform the sequences of each target into a position-specific scoring matrix (PSSM), such that the features can retain biological evolutionary information. We then use a feature vector of molecular substructure fingerprints to describe the chemical structure information of the drug compounds. Second, the Legendre moments algorithm is used to extract new features from the PSSM. Finally, a classification algorithm called rotation forest is used to perform prediction, we tested its prediction performance on four golden standard data sets: enzymes, G-protein-coupled receptors, ion channels, and nuclear receptors. As a result, the proposed method achieves average accuracies of 0.9026, 0.8260, 0.8703, and 0.7444 on these four data sets using five-fold cross-validation. We also compare the proposed method with the support vector machine and other existing approaches. The proposed model is proved to be superior to comparative methods, showing that it is feasible, effective, and robust for predicting potential DTI.