The International Congress on Cyberspace Intelligence and Health 2025
(The 4th Cyberspace Congress)
29-30, November, Online
Topic:
大模型驱动的远程运动康复系统构建及数字化迁移适应性评估
LLM-Driven Remote Exercise Rehabilitation System Construction and Digital Migration Adaptability Assessment
Abstract:
随着数字健康技术和远程医疗的快速发展,线上运动康复服务展现出巨大的潜力,但现有模式在提供实时、个性化和高质量的临床级服务方面仍面临专业依赖高、反馈滞后等核心障碍。本报告旨在探索以大型人工智能模型(LLMs)为核心驱动力的线上运动康复新范式,并提出一个指导康复项目数字化迁移的决策框架。首先,系统分析当前线上康复服务模式及其主要瓶颈;随后回顾重塑线上康复体验的关键核心技术。在此基础上,深入解析大模型技术如何赋能“感知-决策-执行-交互”四大关键环节,并通过系统评测揭示其在提升数据洞察、优化治疗方案、增强用户沉浸感和实现无缝医患沟通方面的潜力。最后,提出创新的双轴决策框架,以“临床线下依赖指数”和“线上技术赋能潜力指数”为核心,将运动康复项目科学归类,为资源配置、技术研发与服务转型提供理论与实践依据。本研究为构建下一代高效、个性化、可规模化的线上运动康复服务体系奠定基础。
As digital health and telemedicine advance, remote exercise rehabilitation is showing remarkable potential, yet current offerings struggle to deliver real-time, personalized, clinical-grade care due to heavy expert reliance and delayed feedback. This keynote explores a new paradigm for online rehabilitation powered by large language models (LLMs) and introduces a decision framework that guides the digital migration of rehabilitation programs. We begin with a systematic analysis of prevailing service models and bottlenecks, then review the core technologies reshaping online rehabilitation experiences. Building on this foundation, we examine how LLMs drive revolutionary upgrades across sensing, decision-making, execution, and interaction, and present a multi-dimensional evaluation that highlights their impact on data insight, treatment optimization, patient immersion, and frictionless clinician-patient communication. Finally, we propose an innovative dual-axis decision matrix based on the Offline Clinical Dependency Index and the Online Technological Enablement Potential Index to categorize rehabilitation projects, enabling precise resource allocation, R&D prioritization, and digital transformation strategies. The work lays the groundwork for the next generation of efficient, personalized, and scalable online exercise rehabilitation services.
Biography:
朱涛,男,博士,副教授,硕士生导师,IEEE高级会员,先后于中南大学和中国科学技术大学获得计算机科学与技术专业学士与博士学位,曾在微博与北京科技大学任职,研究方向聚焦智能感知、物联网和大语言模型在健康领域尤其是运动康复中的应用。
Tao Zhu is an Associate Professor, Master’s Supervisor, and Senior Member of IEEE. He earned his B.S. from Central South University and Ph.D. from the University of Science and Technology of China, both in Computer Science and Technology. He previously worked at Weibo and the University of Science and Technology Beijing. His research centers on intelligent perception, the Internet of Things, and large language models for healthcare, with an emphasis on exercise rehabilitation.