• 首页期刊简介编委会刊物订阅专栏专刊电子刊学术动态联系我们English
引用本文:顾智淳,潘忙忙,张弛,王佳,王昕,胡朦,严颐丹,沈珑,孔令璁,陆姣.基于人G共生的血栓性疾病管理:理论、应用与挑战[J].中国现代应用药学,2026,43(4):93-101.
guzhichun,panmangmang,zhangchi,wangjia,wangxin,humeng,yanyidan,shenlong,konglingcong,lujiao.Human-GenAI Symbiosis-Based Management of Thrombotic Diseases: Theoretical Framework, Clinical Applications, and Challenges[J].Chin J Mod Appl Pharm(中国现代应用药学),2026,43(4):93-101.
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 3次   下载 0 本文二维码信息
码上扫一扫!
基于人G共生的血栓性疾病管理:理论、应用与挑战
顾智淳,潘忙忙,张弛,王佳,王昕,胡朦,严颐丹,沈珑,孔令璁,陆姣
1.上海交通大学医学院附属仁济医院药剂科;2.上海交通大学医学院附属仁济医院心内科;3.西安交通大学公共政策与管理学院
摘要:
血栓性疾病已成为全球主要健康负担,传统模式在异质人群的个体化管理上存在不足。生成式人工智能(GenAI)的兴起催生“人G共生”范式:在医护团队—患者—GenAI三方的平等协作中,以多模态证据整合、可解释的个体化决策与闭环干预实现抗栓治疗的精准化。本综述提出“证据—决策—干预”三子系统框架:证据子系统将异构临床与组学数据转化为结构化知识;决策子系统以多准则分析与患者偏好建模生成可审计的治疗方案;干预子系统通过GenAI多角色与“助推—强推”协同策略提升依从性与持续管理效果。我们结合溶栓决策、围术期VTE预防、抗栓个体化用药与移动健康等场景的实践,展示共生范式在风险预测性能、透明共同决策与动态优化方面的增益,同时讨论幻觉、可解释性、责任归属与标准化等挑战。该范式为血栓性疾病药物治疗优化与以患者为中心的精准医疗提供了可复制的理论框架与落地路径。
关键词:  人G共生  生成式人工智能  血栓性疾病  精准医疗  抗栓治疗  临床决策支持
DOI:
分类号:R284.1;R917.101??????
基金项目:上海市科学技术委员会基金资助项目(23Y11902800);上海市浦东新区卫生健康委员会卫生科技基金资助项目(PW2023D-10);上海申康医院发展中心基金资助项目(SHDC2022CRS035);国家自然科学基金面上项目:以患者为中心的家庭医生团队社区慢性病“证据-决策”互动管理模式与驱动策略研究(72474174)
Human-GenAI Symbiosis-Based Management of Thrombotic Diseases: Theoretical Framework, Clinical Applications, and Challenges
guzhichun1, panmangmang, zhangchi, wangjia, wangxin, humeng, yanyidan, shenlong, konglingcong, lujiao2,3,4
1.Department of Pharmacy, Renji Hospital,Shanghai Jiao Tong University School of Midicine;2.School of Public Policy and Administration, Xi'3.'4.an Jiaotong University
Abstract:
Thrombotic diseases impose a substantial global burden, while traditional models fall short in tailoring care to heterogeneous patients. The rise of Generative AI (GenAI) enables a “Human–GenAI Symbiosis” paradigm in which clinicians, patients, and GenAI collaborate as equal partners to deliver precision antithrombotic therapy through multimodal evidence integration, explainable individualized decision-making, and closed-loop interventions. We propose a three-subsystem architecture—Evidence, Decision, and Intervention. The Evidence subsystem transforms heterogeneous clinical and omics data into structured disease knowledge; the Decision subsystem employs multi-criteria analysis and patient-preference modeling to generate auditable treatment plans; the Intervention subsystem leverages GenAI’s multi-role design and a coordinated “nudge–boost” strategy to strengthen adherence and long-term self-management. Across use cases—intravenous thrombolysis, perioperative VTE prevention, individualized antithrombotic selection, and mobile health—we illustrate gains in predictive performance, transparent shared decision-making, and dynamic optimization, while addressing challenges around hallucination, interpretability, liability, and standardization. This paradigm offers a reproducible theoretical and practical pathway toward patient-centered precision care in thrombotic disease management.
Key words:  human-GenAI symbiosis  generative AI  thrombotic diseases  precision medicine  antithrombotic therapy  clinical decision support
扫一扫关注本刊微信