Efficient task scheduling is crucial in edge computing to improve computational efficiency. However, due to dynamic workload variations and exclusive resource allocation, traditional scheduling mechanisms often fail to effectively reuse scheduling optimization results, generating wasteful scheduling overhead. To reduce scheduling overhead while improving system performance, we propose an Ahead-of-Time (AoT) scheduling mechanism for workflow applications. Through decoupling requests and isolating resources, our mechanism enables sustainable reuse of the pre-cached scheduling plan, thereby improving computational efficiency in edge computing environments. Specifically, our mechanism decouples incoming requests into two categories: those that match the pre-cached scheduling plan and those that do not. The matching requests are scheduled to plan-managed resources, while the rest requests are scheduled to the remaining resources using conventional online scheduling algorithms. To achieve sustainable reuse of the scheduling plan, we isolate the plan-managed resources and remaining resources to prevent interference, and the AoT scheduling algorithm dynamically adjusts the plan’s execution progress to accommodate workload variations. Experimental results in an intelligent transportation scenario demonstrate that, compared to heuristic-based and reinforcement learning-based baselines, the proposed AoT scheduling reduces scheduling overhead by 63.57

Directed acyclic graph (DAG)-based blockchain is a promising paradigm of a blockchain system. During DAG-based consensus processes, nodes generate blocks in parallel, as well as voting YES on the previous blocks. It makes the DAG-based blockchain a performance advantage in terms of confirmation delay and transaction throughput. However, Byzantine faults affect DAG-based blockchain performance by not voting on blocks. The related analysis has not been explored. To this end, based on the most typical DAG-based consensus mechanism with Byzantine Fault Tolerance, hashgraph, we investigate the resilience of the DAG-based blockchain when Byzantine faults do not vote on blocks. First, we propose differential equations to model the running processes of the DAG-based blockchain in both high-load and low-load networks. It reveals the impact of Byzantine nonvoting behaviors on blockchain performance in a mathematical manner. Second, Byzantine nonvoting adversaries can leave the target node orphaned in an opportunistic network of low-load regimes. We propose a ranger-assisted DAG-based blockchain to alleviate the problem. It employs a group of rangers to collectively commit blocks and introduces the reputation, selection probabilities, and shuffling of rangers to supervise node behaviors. The performance of the proposed blockchain is also quantitatively analyzed. Third, we develop a blockchain simulator. The numerical results indicate the validity of the proposed analysis and the efficiency of the proposed blockchain.

With the widespread adoption of blockchain applications, smart contract security has emerged as a critical challenge. Although recent deep learning approaches incorporate pre-trained models, they still face two primary hurdles: (1) the semantic gap between general-purpose pre-trained representations and specific vulnerability detection tasks, which hinders the capture of fine-grained vulnerability patterns; and (2) the lack of effective alignment when fusing heterogeneous syntactic and semantic features within dual-branch architectures.To address these issues, we propose CHASE, a vulnerability detection framework based on Contrastive Heterogeneous Alignment and Semantic Enhancement. Specifically, we design a Vulnerability Semantic Prototype Encoder (VSPE) that utilizes a semantic prototype vector to guide the pre-trained model to focus on vulnerability semantics, thereby enhancing feature representation. Furthermore, we introduce a Semantic-Structural Weighted Contrastive Loss (S2W -Loss) to align heterogeneous features, and employ an adaptive gating network for dynamic fusion. Extensive experiments demonstrate that CHASE consistently outperforms existing methods, achieving F1-scores of 93.68%, 94.37%, and 94.66% on Reentrancy, Timestamp Dependence, and Integer Overflow, respectively.

The integration of Internet of Vehicles (IoV) and blockchain technology ensures the security of IoV data. As an asynchronous Byzantine fault-tolerant directed acyclic graph (DAG) consensus, Hashgraph consensus effectively addresses the performance bottleneck of a single block-producing node in IoV. However, there are still limitations, such as malicious behavior and node passivity in IoV. The process of Hashgraph consensus relies on the interactions between individual nodes. However, malicious behavior and negative interactions of nodes can reduce the performance and efficiency of consensus. To tackle the issues, this paper proposes a reputation-enhanced Hashgraph consensus. A reputation model is developed to calculate the reputation scores of nodes. The virtual voting weights are dynamically adjusted based on these reputation scores during the consensus process. The proposed consensus mechanism reduces the influence of both the malicious nodes and the passive nodes in the consensus voting process, and strengthens the voting weight of high-reputation nodes. It improves the performance and efficiency of the consensus process. Simulation experiments show that, compared to Hashgraph consensus, the proposed consensus mechanism shows improved throughput performance in IoV scenarios and exhibits strong resistance in scenarios with a majority of malicious nodes.

Lesion morphology and quantity evaluation in computer tomography (CT) images are critical for precise disease diagnosis. Most existing methods employ machine learning-based methods to separately evaluate the morphology and quantity of individual lesion, neglecting the synergy between morphological structure and quantitative distribution. This limitation presents challenges when handling multiple complex lesions. This paper proposes an unsupervised correlation learning-based clustering model for evaluating lesion morphology and quantity in scenarios involving multiple complex lesions without predefined specific-logic. Specifically, the model utilizes clinical knowledge and changes in the in- or out-degree of lesion regions to learn their interdependencies, automatically recognizing domain-specific morphological features. These morphological features serve as key representations for morphology estimation and provide essential contextual information for quantity analysis. Furthermore, the model perceives quantity evaluation as a density-based clustering process. By interacting with domain-specific morphological features, the model dynamically adjusts the search objects, followed by designing morphology-specific parameter search strategies to autonomously learn spatial relationships between lesion regions. This approach facilitates the exploration of optimal parameters for accurate lesion evaluation without manual intervention. Experiments conducted on the kidney stone dataset including 53 samples and the kidney tumor dataset comprising 300 samples, indicate that the proposed model has achieved 92.45% and 95.33% accuracy in morphology analysis, respectively. For quantity analysis, the proposed model has achieved 79.25% and 94.33% accuracy, outperforming the well-performing AR-DBSCAN method by +30.19% and DRL-DBSCAN method by +6%. The proposed model is demonstrated to be effective in handling morphology and quantity estimation for multiple complex lesions.

Smart contract vulnerabilities represent a significant threat to the security and reliability of blockchain ecosystems, highlighting the need for efficient and accurate detection methods. Existing methods struggle to capture the rich semantic and structural information in smart contracts. To address these issues, this work proposes SeqGraphFusion (SGF), a framework that leverages the complementary strengths of token sequence and graph-based representations extracted from smart contract source code. SGF’s key insight lies in its heterogeneous fusion of these sequence and graph features, enabling a more comprehensive understanding of the contract. To enhance the modeling of long-range dependencies, which have been constrained in traditional Graph Neural Networks due to the nature of message passing, SGF incorporates a Graph Transformer-based architecture, facilitating more effective information propagation. The experimental results demonstrate that our method significantly outperforms existing approaches, achieving a notable improvement in the F1 score, with an average increase of 2.82

Lung cancer remains a leading cause of mortality, making accurate subtype identification crucial for effective lung cancer diagnosis and treatment. Recent advancements in medical image classification, particularly through Convolutional Neural Networks (CNNs) and Transformers, have significantly improved the analysis of CT and histopathological images. However, CNNs are limited in capturing global features due to their restricted receptive fields. Alternatively, Transformers, while effective at modeling long-range dependencies, lack local inductive bias. To address these limitations, this paper introduces a novel multi-scale feature fusion network that integrates CNN and Transformer architectures. At its core is the Adaptive Feature Fusion (AFF) Block, which adaptively fuses hierarchical features from both networks, capitalizing on their strengths to improve classification performance. This module forms the feature extraction backbone and is complemented by a novel classification head. The classification head utilizes multi-head attention to perform cross-attention between class embeddings and image features. This process involves calculating cosine similarity for classification. The objective is to mitigate overfitting and address inter-class similarity issues. A contrastive loss function further maximizes the separation between class embeddings, enhancing classification accuracy. The proposed model was evaluated through multiple experiments on a clinical lung tumor surgical lesion slice image dataset. FusionClassNet achieved a classification accuracy of 88.68% and an F1-score of 83.28%. The experimental results demonstrate the effectiveness of the model in handling nonstandard medical images of lung tumors.

Mobile Edge Computing (MEC) offers computational services near data sources to meet numerous real-time data processing demands of end devices. Scheduling dependent tasks in resource-constrained environments is a key research focus in MEC, aimed at enhancing the completion rates of applications within their deadlines. However, most existing works overlook the resource competition among requests arriving at different times, leading to decreased application completion rates. In this paper, we propose a dependent task online scheduling approach for multiple applications to optimize application completion rates. For multiple applications with deadline constraints, where tasks within each application may have dependencies, we propose a multi-priority task sequencing algorithm to determine the execution order of tasks. To accommodate scenarios where requests arrive at different times, we introduce a priority-based queue to dynamically adjust task execution order based on urgency and resource demands. Finally, by comparing with baseline approaches, experimental results demonstrate that our approach can improve the application completion rate by approximately 41.02%, demonstrating its effectiveness.

Randomized controlled trials(RCTs)provide optimal evidence of the effectiveness and safety of a new drug,a new medical device,or a new therapeutic strategy with the necessary scientific design 1].

Deep learning is widely used in medical image analysis, but obtaining large-scale annotation data is challenging due to the need for expert medical professionals to perform the annotations. This paper proposes a semi-supervised active learning framework to classify medical images with limited annotated data, helping clinicians assess lung tumor risk levels. This study designs an active learning strategy that estimates the uncertainty of samples based on their training dynamics, allowing for the selection of the most informative samples for manual labeling. To address the cold start problem of active learning, the framework constructs an initial labeled dataset with a uniform distribution using unsupervised training and clustering techniques. The semi-supervised learning framework strategically combines limited annotated samples with abundant unlabeled datasets during model training. This study conducts supervised training on the labeled data, while consistency regularization and pseudo-labeling techniques are applied to the unlabeled data. The framework effectively combines the advantages of active learning and semi-supervised learning. Compared to traditional semi-supervised learning frameworks, this framework effectively prioritizes the selection of samples on the decision boundary for labeling. Moreover, in each iteration of the active learning training process, the framework integrates the latest semi-supervised trained model, further improving the selection performance of active learning. Extensive evaluations on a clinical lung tumor surgical lesion slice image dataset show that the proposed framework achieves 88.79

Accurate identification of lung cancer subtypes is critical for the diagnosis and treatment of pulmonary lesions in clinical practice. Deep learning-based diagnostic systems for surgical lesion section images of lung tumors can assist clinicians in efficiently and accurately recognizing lung cancer subtypes, enabling targeted therapies. However, existing methods often fail to fully leverage prior knowledge from medical images, making them susceptible to interference from high-intensity light regions caused by external illumination. Moreover, many existing methods suffer from high computational complexity and excessive parameters, which hinder their practical application. To address these challenges, this paper proposes MTPKDistillNet, a novel framework designed for subtype recognition of lung cancer from surgical lesion section images. The framework incorporates a prior knowledge module to extract edge and texture features from images. It then performs multi-level information distillation from multiple teacher networks to enhance the feature extraction capability of the student network. Finally, the framework employs decoupled and category relation distillation, allowing the student network to effectively learn both the predicted outputs and the class relevance information from the teacher network. This enhances the diagnostic capability of the model. Experimental results on a clinical lung cancer surgical slide dataset demonstrate that MTPKDistillNet achieves a classification accuracy of 87.55

Mobile Edge Computing (MEC) effectively addresses the real-time data processing needs of devices by providing computing resources closer to data-generating devices. Efficient scheduling of workflows composed of multiple tasks with data dependencies within MEC environments is crucial for improving Quality of Service (QoS). However, workflow scheduling in dynamic MEC environments faces significant challenges due to unpredictable resource fluctuations. Existing studies often rely on static assumptions or precise predictions. Fluctuations cause task completion times to deviate from expectations, resulting in suboptimal scheduling performance and degraded QoS. To address these challenges, this paper introduces an Online Workflow Application Scheduling method based on Deep Reinforcement Learning (OWAS-DRL) to minimize average workflow completion time in dynamic MEC. Our method utilizes Proximal Policy Optimization (PPO) to learn optimal scheduling policies under uncertain environments. Specifically, OWAS-DRL employs Frame Stacking to capture the temporal dynamics of the system and introduces a Long and Short-term Reward Function (LSRF) to balance immediate and long-term optimization goals effectively. Compared to existing heuristic and reinforcement learning algorithms, extensive experimental results demonstrate that OWAS-DRL achieves about 43% and 20% reductions in average workflow completion times, respectively.

Smart contracts typically store crucial application state data. Methods like SPV proofs are commonly used to verify the authenticity of cross-chain transactions. However, the heterogeneity between blockchains makes common cross-chain data validation methods impractical for smart contract state data. To address this challenge, this paper proposes MSCV to achieve cross-chain verification of smart contract state data. MSCV utilizes MTC to provide a universal verification method for state data and preserves the chronological order of state data. Furthermore, MSCV introduces zero-knowledge proof technology to reduce the additional verification overhead of MTC. Finally, this paper implements smart contracts containing MTC and a batch verification zero-knowledge proof circuit and evaluates their performance and effectiveness.

Nowadays, the integration of blockchain and MANET improves both MANET security and blockchain scalability. Hashgraph can better support the MANET-based blockchain by its inherent features, i.e., gossip about gossip, Byzantine fault tolerance, etc. In order to understand the limitations and performance of hashgraph in MANETs, this work investigate the theoretical model with two metrics. The metrics include the degree of node isolation and witness confirmation delay. By using the differential equations and markov chain, this work quantitatively analyze the general hashgraph with an optimal contact probability and the improved hashgraph resisting eclipse attacks, as well as identifying the expressions of these performance metrics. Extensive experiments support the analysis and reveal the impact of moving nodes and Byzantine nodes on the hashgraph performance in MANETs. To the best of our knowledge, our work is the first to apply DAG (Directed Acyclic Graph) blockchain for the MANETs from a methodological perspective.

Traffic matrix (TM) prediction aims to forecast future traffic data for networks using historical traffic matrices. TM prediction plays a critical role in solving network engineering tasks such as routing management, capacity planning, and network security. Previous research assumes that the TM sequences fed into the prediction models are complete and precise. However, considering the unavoidable failure of network systems and their large monitoring costs, it is impractical to collect complete TMs from large-scale networks. In this paper, we study the TM prediction problem with randomly missing values. To perform TM completion, we introduce a masked matrix modeling method based on self-supervised learning that can learn better matrix representations. As the matrix completion and prediction tasks are highly correlated, we propose an end-to-end framework that performs the two tasks simultaneously through joint learning. Specifically, we design a 3D-UNet architecture that is able to exploit multi-scale spatio–temporal correlations in a TM sequence as the completion module. An LSTM2D architecture is employed as the prediction module to take advantage of spatio–temporal dependencies. Extensive experiments are conducted on publicly available datasets, and the results show that our model significantly outperforms previous state-of-the-art methods. Source code is available at https://github.com/FreeeBird/TM_prediction_random_missing.

Smart contracts, as a critical application of blockchain technology, play a pivotal role in automating contract rules and enhancing the transparency of transactions. They are programmed to define rules and automate protocols on the blockchain, ensuring that transactions are decentralized, efficient, and immutable. Once deployed, smart contracts cannot be altered, and any existing vulnerabilities can be maliciously exploited, leading to potential financial losses or data breaches. Consequently, the security of smart contracts has become a critical focus in blockchain security. This paper presents the XLNET-HyBA model for Ethereum smart contract vulnerability detection based on the fusion of heterogeneous data. Initially, the source code is transformed into Abstract Syntax Tree (AST) serialized information by the SmartConvert preprocessing algorithm, which effectively integrates the scattered control flow elements. Subsequently, a novel information fusion embedding technique is proposed to optimize the contribution weights of source code and AST serialized code, aiming for a more accurate localization of potential vulnerabilities. Finally, a hybrid loss function strategy based on metric learning is constructed, enhancing the model’s sensitivity to subtle differences. The experimental results show that the accuracy of all four vulnerabilities is over 96

The accurate identification of low-risk and high-risk lung tumors is essential for clinicians to develop lung cancer treatment strategies during surgery. Despite the achievements in deep learning, most medical image analysis applications are still hampered by the difficulty of obtaining large amounts of labeled data. In this paper, a semi-supervised deep learning framework, DS-FixMatch, is proposed for identifying lung tumors intra-operatively while alleviating the issue of sparse annotations. DS-FixMatch combines selective labeling and semi-supervised training: (1) For the acquired unlabeled images, a subset that best represents the distribution of the entire dataset is selected by an unsupervised algorithm. Compared to the traditional random labeling strategy, this method can avoid introducing samples that are highly influenced by the intraoperative environment for labeling. This subset is then sent to a human expert for labeling. (2) Supervised training is performed using labeled images. For the remaining unlabeled samples, DS-FixMatch utilizes model predictions to generate pseudo-labels for consistency regularization, further enhancing the model’s generalization ability. A dataset consisting of 2221 natural images, each capturing the Region of Interest (ROI) in lung tumors, is constructed to evaluate the effectiveness of the designed framework. Experiments show that DS-FixMatch leads in performance for the task of lung tumor recognition compared to other baselines.

Blockchain is a peer-to-peer network system that stores chains of transaction data. Blockchain is decentralized and immutable. Once data is added to the chain, it cannot be modified or deleted. Smart contracts are intelligent protocols running at the application layer of blockchain, enabling trustworthy transactions without third parties. However, the openness and immutability of the architecture introduce various security risks. Integer bugs arise from improper handling of numerical variables. They pose significant dangers in smart contract because, in contracts, integer variables are often used to represent critical financial information, such as account balances, asset prices, and transaction amounts. If miscalculations occur, they can lead to substantial financial losses. This paper describes SolSecure, a framework based on abstract interpretation theory [1, 2], designed to find integer bugs in Ethereum smart contracts. Experimental results show that SolSecure can effectively detect integer vulnerabilities in Solidity smart contracts.

Mobile crowdsourcing (MCS) takes advantage of widely distributed mobile devices to complete some temporal-spatial tasks. Edge computing is integrated into MSC to reduce the service delay from a remote cloud, and enhance the quality of services (QoS) through preprocessing data. However, the profit-driven edge nodes (workers) may provide fake or low-quality answers for lowering their data processing cost, which results in serious QoS challenges. We propose a reputation-based accountability mechanism, in which workers are accountable for their answer provision through reputation values, and edge nodes take on different responsibilities of blockchain management based on reputation values and some key factors. Specifically, the combination of the data-centric method and entity-centric method is utilized for precisely evaluating the quality of answers and managing the worker’s reputation. Storage nodes, mining candidate nodes, relay nodes, and verification nodes are mainly responsible for transaction retrieval, block generation, block forwarding and block verification, respectively. It reduces the latency of blockchain maintenance in a large-scale network. Finally, the experimental results show that our scheme achieves high-level reliability and reasonable efficiency for large-scale services in MCS.