人工智能在增龄相关衰弱患者康复中的应用专家共识OA

With the acceleration of population aging,age-related frailty,a clinical syndrome characterized by the progressive decline of physiological functional reserves across multiple organ systems with advancing age,has become a key public health con-cern affecting health outcomes and quality of life in older adults.This condition manifests across four dimensions:physical,cogni-tive,psychological,and social.Traditional assessment methods demonstrate significant limitations in dynamic monitoring and per-sonalized interventions.Artificial intelligence(AI)and intelligent rehabilitation technologies,which feature strengths including cross-modal data fusion,dynamic monitoring,and individualized predictive modeling,provide innovative approaches for the precise management of age-related multidimensional frailty.This consensus document was developed by an interdisciplinary expert panel following evidence-based methodological principles.Based on systematic literature review and the grading criteria of the Oxford Centre for Evidence-Based Medicine(OCEBM),this study systematically reviews current evidence and puts forward standardized recommendations for AI applications in the assessment and intervention of age-related multidimensional frailty.The content covers the technical framework of AI across perception,data processing,model integration,and intervention guidance.It proposes typical AI-assisted assessment and intervention pathways across the physical,cognitive,psychological,and social dimensions of frailty,and puts forward clinical application recommendations for various technologies,including exoskeleton robots,virtual reality(VR)com-bined with brain-computer interfaces(BCI),digital psychological intervention platforms,and social companion robots,with corre-sponding evidence-based recommendation grades.Furthermore,the consensus clarifies the indications and contraindications for AI applications in the rehabilitation of age-related multidimensional frailty,emphasizing the importance of multicenter clinical valida-tion,ethical compliance,data security,and interdisciplinary collaboration.In the future,AI technology will evolve towards digital twin systems,personalized rehabilitation algorithms,and multimodal deep learning platforms,promoting home-based,lightweight,and integrated development of rehabilitation devices,and constructing a new ecosystem of full-cycle,multi-dimensional human-ma-chine collaborative rehabilitation.This consensus aims to provide practical guidance for clinicians,rehabilitation specialists,and other relevant practitioners,so as to promote the scientific,standardized,and sustainable application of AI technology in the rehabilitation of patients with age-related frailty.