人工智能在烧伤谵妄预警与管理中应用的研究进展
Research advances on the application of artificial intelligence in the warning and management of burn-related delirium
摘要
烧伤患者谵妄发生率高、识别困难且干预窗口狭窄,是重症监护中的突出难题。人工智能技术凭借其善于处理复杂高维数据、挖掘隐匿模式与实时连续监测的独特优势,为烧伤谵妄的预警与管理提供了新方向。该文基于烧伤谵妄的主要驱动机制与临床异质性,系统综述人工智能在烧伤谵妄预测模型建立、多模态数据融合与可解释输出、多源数据采集、智能处理与决策支持、临床应用与闭环反馈等方面的研究进展,分析当前面临的数据治理与算法公平性、临床场景的整合与协同、责任认定与伦理监管等挑战,并对未来研究方向进行展望,以期为人工智能在烧伤领域的应用提供借鉴与参考。
Delirium in burn patients presented a prominent challenge in intensive care, characterized by high incidence, difficult identification, and a narrow intervention window. Artificial intelligence (AI) technology, with its unique advantages of being adept at processing complex high-dimensional data, mining hidden patterns, and real-time continuous monitoring, provided new directions for the warning and management of burn-related delirium. Based on the main driving mechanisms and clinical heterogeneity of burn-related delirium, this article systematically reviewed the research advances of AI in establishment of predictive models for burn-related delirium, multimodal data fusion and interpretable outputs, multi-source data acquisition, intelligent processing and decision support, as well as clinical applications and closed-loop feedback. It analyzed current challenges such as data governance and algorithmic fairness, integration and collaboration in clinical scenarios, and liability attribution and ethical supervision, and prospected future research directions, aiming to provide references and insights for the application of AI in the field of burns.
年,卷(期):
2026, 42(10)
在线出版日期:
2026年10月20日