Drug-induced hepatotoxicity (DIH), characterized by diverse phenotypes and complex mechanisms, remains a critical challenge in drug discovery. To systematically decode this diversity and complexity, we propose a multi-dimensional computational framework integrating molecular structure analysis with disease pathogenesis exploration, focusing on drug-induced intrahepatic cholestasis (DIIC) as a representative DIH subtype. First, a graph-based modularity maximization algorithm identified DIIC risk genes, forming a DIIC module and eight disease pathogenesis clusters. Network proximity values between drug targets and DIIC clusters were calculated to define drug-disease relationships. Subsequently, a random forest model combining Mordred molecular descriptors, structural alerts (SAs), and network proximity achieved robust DIIC prediction: Accuracy(ACC) = 0.740 ± 0.014 and area under the curve (AUC) = 0.828 ± 0.008 (ntraining = 342, nvalidation = 114, nexternal test = 295, randomly modeling 100 times). Notably, a K-nearest neighbors-graph convolutional network classified drugs into 8 clusters, with the Cluster 3 model demonstrating superior performance (ACC = 0.810 ± 0.024; AUC = 0.890 ± 0.014; ntraining = 186, nvalidation = 63, nexternal test = 172). Mechanistic analysis linked critical SAs to DIIC pathogenesis: (i) Furan (SA3) perturbed cytochrome P450-mediated metabolism and regulation of lipid metabolism by PPARα; (ii) Nitrogen-sulfur heteroatom chains (SA7) disrupted metabolism of steroids; (iii) Phenylthio groups (SA12) and their CYP450 metabolites induced cholestasis. This multi-dimensional framework bridges molecular features and disease mechanisms, offering a generalizable strategy for toxicity prediction and pathway-centric drug safety evaluation, especial for complex disease.

Dense low-power small cell base station (SBS)-based heterogeneous wireless cellular networks (HetNets) have attracted much attention to achieving high-traffic density and peak rate performance. However, the serious energy consumption problem is still a challenge for HetNets. The use of renewable energy (RE) has been considered as one promising solution for the above problem. This article proposes an energy trading scheme among base stations in RE-based HetNets. All SBSs in HetNets are considered as either the energy demander (SBS-ED) or the energy supplier (SBS-ES) based on their abilities in producing RE, and the macro base station (MBS) works as the energy trading manager to control the trading price. A dynamic evolutionary game-based energy trading model between SBS-ES and SBS-ED is established to achieve cooperative energy trading, and the evolutionary stable strategy (ESS) of the proposed model is analyzed. The pricing mechanism of MBS is also investigated, which can effectually affect the EES performance of the proposed model. It is concluded that the MBS’s strategy in the trading price can affect the energy trading strategies of the SBSs. An energy transmission model is proposed, and the minimum energy loss is considered as the goal to obtain the optimal solutions. Numerical results are given to prove the validity and correctness of our proposed method.

When degradation occurs in a dispersed computing network, mitigating the persistent effects of failures and improving the disposal efficiency is an open problem. However, the occurrence of failed nodes in a dispersed computing network is a random and low-probability incident. Further, the resilient resources for emergency allocation are from devices with complex spatial locations in practice. These factors pose challenges to the resource allocation problem using resilience mechanisms. In this article, we explore and analyze a resilience mechanism for dispersed computing networks as an optimization model. We investigate a resilience-aware dynamic resource allocation model to cope with a degraded dispersed computing network and obtain better emergency response at a lower cost. The uncertainties of node failures are uniquely explored to capture failure nodes more precisely and initiate the resilience mechanism for such nodes. In addition, we propose a novel approach to deal with the dynamic and complex coupling characteristics of decision variables in the model. This approach incorporates the induced artificial fish swarm algorithm with dynamic system simulation to generate and improve the scheme for the allocation of resilient resources. Finally, numerical simulation results verify the improved performance of our model and the effectiveness of the algorithm.

Mobile crowdsensing (MCS) is an effective perception paradigm for large-scale tasks, driven by the proliferation of mobile devices with more powerful sensing and computing capabilities. An effective incentive mechanism is critical to the operation of an MCS system in promoting public engagement. However, the great majority of works discuss fixed task pricing, while the inherent inequality of the supply–demand relationship of the tasks exists. Therefore, it is essential to study the dynamic task pricing problem in the peer-to-peer data sharing MCS system. In this article, we formulate the interactions between the requester and the sensors as a two-stage Stackelberg differential game model, while considering the average behavior of sensors to solve the dynamic task pricing problem. Specifically, in the game model, the requester is the leader who first announces the issued task rate and provides decisive state-changing task pricing dynamics to the sensors. Then, the sensors are the followers who decide the rate of tasks completed noncooperatively based on requesters’ observed strategy, using the level of effort as the state dynamics. The requester and the sensors interact through a mean-field term included in the dynamic state functions, which catches the average behavior of all users. By solving the model, the optimal strategies for the users and the optimal tasks pricing trends in the dynamic environment are obtained. Furthermore, the effectiveness and feasibility of the scheme are verified by a series of numerical simulation experiments.

The ubiquitous mobile portable devices have accelerated the rise of mobile crowdsensing (MCS), a distributed perception paradigm. In MCS, multiple requesters issue their sensing tasks on a server platform, and then the platform distributes the tasks to multiple workers. In this process, requesters typically specify the tasks and requirements; meanwhile, the needed task pricing to workers is also specified, which is used to offsetting worker’s efforts by completing the tasks. However, for different application scenarios, the task requirements, the time periods, and the resource consumptions for completing the tasks are varied, which results in a challenge to raise an appropriate task pricing to diverse workers. Therefore, we study the dynamic task pricing problem in the MCS network system with diverse factors (e.g., multiple requester queueing competitions, dynamic task requirements, and distinct waiting time costs). To solve the problem, we resort to the theory of Age of Information (AoI). Specifically, we leverage the AoI timeliness metric in modeling the requester’s waiting time costs, and then we use queueing game theory to build the dynamic task pricing model. The model analysis has shown the existence of the optimal pricing strategies under the first-come–first-served (FCFS) queueing rule and the last-come–first-served with preemptive in waiting (LCFSW) queueing rule. Finally, numerical simulations are conducted to validate the existence of the optimal task pricing.

Video super-resolution (VSR) and video frame interpolation (VFI) are inter-dependent for enhancing videos of low resolution and low frame rate. However, most studies treat VSR and temporal VFI as independent tasks. In this work, we design a spatial-temporal super-resolution network based on exploring the interaction between VSR and VFI. The main idea is to improve the middle frame of VFI by the super-resolution (SR) frames and feature maps from VSR. In the meantime, VFI also provides extra information for VSR and thus, through interacting, the SR of consecutive frames of the original video can also be improved by the feedback from the generated middle frame. Drawing on this, our approach leverages a simple interaction of VSR and VFI and achieves state-of-the-art performance on various datasets. Due to such a simple strategy, our approach is universally applicable to any existing VSR or VFI networks for effectively improving their video enhancement performance.

The last few years have seen the great potential of artificial intelligence (AI) technology to efficiently and effectively deal with an incredible deluge of data generated by the Internet of Things (IoT) devices. If all the massive data is transferred to the cloud for intelligent processing, it not only brings considerable challenges to the network bandwidth but also cannot meet the needs of AI applications that require fast and real-time response. Therefore, to achieve this requirement, mobile or multiaccess edge computing (MEC) is receiving a substantial amount of interest, and its importance is gradually becoming more prominent. However, with the emerging of edge intelligence, AI also suffers from several tremendous security threats in AI model training, AI model inference, and private data. This paper provides three novel defense strategies to tackle malicious attacks in three aspects. First of all, we introduce a cloud-edge collaborative antiattack scheme to realize a reliable incremental updating of AI by ensuring the data security generated in the training phase. Furthermore, we propose an edge-enhanced defense strategy based on adaptive traceability and punishment mechanism to effectively and radically solve the security problem in the inference stage of the AI model. Finally, we establish a system model based on chaotic encryption with the three-layer architecture of MEC to effectively guarantee the security and privacy of the data during the construction of AI models. The experimental results of these three countermeasures verify the correctness of the conclusion and the feasibility of the methods.