Search

Background Type 2 diabetes is a major public health problem that jeopardizes the health of the population in China. The prediabetic state population is a huge reserve force for diabetes, for which appropriate intervention can prevent or delay diabetes, but community health management is ineffective at this stage. Objective To investigate the problems and influencing factors in community health management practice in prediabetic state, and propose systematic and operable strategies and recommendations on health management measures. Methods A systematic search was conducted from March to April 2023 to retrieve literature related to community health management of prediabetes in CNKI, Wanfang Data, CQVIP, PubMed, and Web of Science. The key informant interviews based on multiple perspectives were conducted from April to May 2023 in Shanghai communities, hospitals, and CDCs, which included 20 community health service center staff, health administrators, clinical endocrinologists, CDC health management staff, diabetes patients or prediabetes and their family members, and people with diabetes risk factors. The interviews content covered the current status of community health management of prediabetes, related attitudes and perceptions of prediabetes, and acceptance of relevant services. Fishbone diagram analysis were used for data analysis to sort out the hierarchy between the problems related to community health management of prediabetes. Results 14 relevant literature were eventually included to summarize 22 existing problems in community health management of prediabetic state based on literature and interviews, derive the problems in 4 dimension: patients, intervention scope, service ability and information system, and propose 8 influencing factors: disease risk perception level, self-management skill level, funding budget, work experience, workload, service accessibility, electronic health record construction level and information sharing scope level. Conclusion The prediabetic state population is an important target of community health management, and the process of policy change reflects the strengthening of attached to the them, but multiple sources of evidence prove that it is still a weak link at present. In view of the problems in the current practice, it is necessary to improve the knowledge, skills and other management capabilities of community health service personnel, optimize the information system platform based on the medical treatment combination, and further develop a more feasible and continuous health management model for prediabetic state integrating screening, management and intervention.

With the development of digital technology， an increasing number of artificial intelligence （AI） technologies are being applied in the field of public health， significantly improving the efficiency of healthcare systems. However， such technological advancement also introduces a series of ethical risks. In this article， we conducted a systematic review by searching nine domestic and international databases and analyzing the ethical issues related to AI in public health， ultimately including 158 articles. Based on the analysis of the included literature， ethical risks were categorized into four aspects： data， algorithms， rights and responsibilities， and social impact. A total of 15 key issues were identified， among which privacy and confidentiality， informed consent， data security， and fairness， justice and inclusion emerged as the most prominent issues. The ethical challenges posed by AI in the field of public health cannot be ignored， and it is necessary to formulate ethical guidelines and practical recommendations for AI in this field， establish sound regulatory and review mechanisms， thereby ensuring the healthy development of AI research in public health.

Behavior Trees are widely used in many games, however, they are normally designed in a static way which is hard to deal with dynamic environments. One of the factors that restrict the performance of BTs is the optimization of resources used among subtasks, which has not been paid much attention. In this paper, we concentrate on the problem of dynamic allocation of agents for BTs, i.e., an important resource in games. We design a novel hierarchical learning framework, including the allocation learning module and the action learning module, to find strategies for agent allocation and action selection. Each agent is associated with a network for encoding observations, which is then assigned to a subtask by the allocation learning network. Each subtask uses a network to compute action for the corresponding agent. We design how the losses are computed for the two modules respectively and present the whole training procedure.