Cardiovascular health is vital to human well-being, and cardiac magnetic resonance (CMR) imaging is considered the clinical reference standard for diagnosing cardiovascular disease. However, its adoption is hindered by long scan times, complex contrasts, and inconsistent quality. While deep learning methods perform well on specific CMR imaging sequences, they often fail to generalize across modalities and sampling schemes. The lack of benchmarks for high-quality, fast CMR image reconstruction further limits technology comparison and adoption. The CMRxRecon2024 challenge, attracting over 200 teams from 18 countries, addressed these issues with two tasks: generalization to unseen modalities and robustness to diverse undersampling patterns. We introduced the largest public multi-modality CMR raw dataset, an open benchmarking platform, and shared code. Analysis of the best-performing solutions revealed that prompt-based adaptation and enhanced physics-driven consistency enabled strong cross-scenario performance. These findings establish principles for generalizable reconstruction models and advance clinically translatable AI in cardiovascular imaging.

Edge node placement is a key topic to edge cloud systems for that it affects their service performances significantly. Previous solutions based on the existing information are not suitable for the mobile environment due to the mobility and random Internet access of end users. In this article, we propose a dynamic virtual edge node placement scheme, in which the edge node placement strategy is generated based on the prediction information. Our placement scheme applies the pay-as-you-go and Spot Instance model of cloud computing, which may allocate the service resources with low cost conveniently and flexibly. What’s more, Long Short-Term Memory (LSTM) is implemented to predict the information of end users’ requests and the resources’ prices, endowing the generated placement strategy with the adaptability to the change of end users. At last, a set of hierarchical-clustering-based placement algorithms are proposed, which not only locate virtual edge nodes and allocate their corresponding service resources actively, but also guarantee the service quality of end users with low time complexity. The simulation with trace data shows that compared with K-means-clustering-based placement schemes, our virtual edge node placement scheme can provide users with high-quality service in terms of network delay with relatively low placement cost time-efficiently.

Floating Car Data (FCD) is a kind of emerging data in the field of traffic engineering. There are three problems in its application of the traffic state discrimination: First, the existing traffic state discrimination model is only for road segment detection; Second, the road-segment based discrimination model is not conducive to the spatial-temporal evolution analysis of traffic state. Third, the existing road segment traffic state discrimination model directly adopts the prototypical Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm, and the detection result is limited in accuracy. We proposed a multi-dimensional traffic state discrimination method. Firstly, the traffic state of the road segment is determined based on the improved DBSCAN algorithm. The dynamic segmentation technology is used to realize the visualization of the road traffic state. Then, the traffic incident point discrimination model is constructed according to the spatial-temporal evolution pattern of the road conditions under traffic incidents. The visualization results show that the proposed method can achieve relatively fine multidimensional traffic state discrimination.

According to the fact that the function of existing map templates is difficult to extend and the control is not flexible enough, a dynamic map template based on combination of cartographic rules was designed. First of all, according to the characteristics of cartography, the cartographic rules contained in each cartographic link were summarized, and then the dynamic map template applied to the whole cartographic process was designed by using cartographic rules as the basic unit, and finally through converting the user's cartographic operations into the cartographic rules and the organic combination of cartographic rules, a dynamic map template was constructed. Compared with the traditional map template, the dynamic map template has strong hierarchical and scalability, which helps to simplify the cartographic process and ensure the scientificity of the user's actions.

Public safety has been a great concern in recent years as terrorism occurs everywhere. When a public event is held in an urban environment like Olympic games or soccer games, it is important to keep public safe and at the same time, to have a specific plan to control and rescue the public in the case of a terrorist attack. In order to better position public safety in communities against potential threats, it is of utmost importance to identify existing gaps, define priorities and focus on developing approaches to address those. In this paper, we present a system which aims at providing a decision support, threats response planning and risk assessment. Threats can be in the form of Chemical, Biological, Radiological, Nuclear and Explosive (CBRNE). In order to assess and manage possible risks of such attacks, we have developed a computational framework of simulating terrorist attacks, crowd behaviors, and police or safety guards' rescue missions. The characteristics of crowd behaviors are modeled based on social science research findings and our own virtual environment experiments with real human participants. Based on gender and age, a person has a different behavioral characteristic. Our framework is based on swarm intelligence and agent-based modeling, which allows us to create a large number of people with specific behavioral characteristics. Different test scenarios can be created by importing or creating 3D urban environments and putting certain terrorist attacks (such as bombs or toxic gas) on specific locations and time-lines.