随着慢性疾病（简称慢病）患病率上升，需要机械通气的慢病患者显著增加，呼吸机依赖问题日益突出。慢病的状态是决定呼吸机依赖发生发展的关键因素。慢病人群呼吸机依赖涉及呼吸、循环、神经肌肉及代谢等多系统交互作用，常存在表型重叠与动态演变。构建以慢病为核心，整合慢病基线、实时生理参数、影像学特征、生物标志物及治疗反应等多维数据的人工智能（AI）预测体系有望降低呼吸机依赖发生率，推动重症医学向精准化诊疗发展。本文从慢病状态的核心作用、呼吸机依赖的表型多样及AI应用等方面进行论述，旨在为构建以慢病为核心的呼吸机依赖精准预测体系提供理论依据与实践思路。

The rising prevalence of chronic diseases has led to a significant increase in the number of patients with chronic conditions requiring mechanical ventilation, making ventilator dependence an increasingly prominent clinical issue. The status of underlying chronic disease serves as a pivotal determinant in the development and progression of ventilator dependence. In patientwith chronic diseases, ventilator dependence involves complex multisystem interactions, including respiratory, cardiovascular, neuromuscular, and metabolic systems, often presenting with phenotypic overlap and dynamic evolution. The development of an artificial intelligence-driven predictive system—centered on chronic disease profiles and integrating multidimensional data such as baseline chronic disease status, real-time physiological parameters, imaging features, biomarkers, laboratory indices, and therapeutic responses—holds significant promise for reducing the incidence of ventilator dependence and promoting the advancement of critical care medicine toward precision diagnosis and treatment. This article discusses the core role of chronic disease status, phenotypic diversity of ventilator dependence, and AI applications, aiming to provide theoretical basis and practical insights for establishing a precision prediction system for ventilator dependence centered on chronic diseases.