节点文献

数智赋能胃肠道肿瘤防治的临床应用进展与挑战

Progress and challenges of clinical applications of intelligence-driven strategies in the prevention and treatment of gastrointestinal cancers

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 周婷婷杨洋张涛李浩武文玉刘振宇苏晓兰田捷程京魏玮

【Author】 ZHOU Tingting;YANG Yang;ZHANG Tao;LI Hao;WU Wenyu;LIU Zhenyu;SU Xiaolan;TIAN Jie;CHENG Jing;WEI Wei;Department of Spleen and Stomach Diseases, Wangjing Hospital, China Academy of Chinese Medical Sciences;Key Laboratory of Molecular Imaging, Chinese Academy of Sciences/Institute of Automation, Chinese Academy of Sciences;Center for Systems Biology of Medicine, School of Medicine, Tsinghua University;National Engineering Research Center for Biochips;

【通讯作者】 魏玮;

【机构】 中国中医科学院望京医院脾胃科中国科学院分子影像重点实验室/中国科学院自动化研究所清华大学医学院医学系统生物学研究中心生物芯片北京国家工程研究中心

【摘要】 胃肠道肿瘤(gastrointestinal cancer, GIC)包括食管癌、胃癌和结直肠癌,是全球发病率高、致死率高的重大疾病。GIC起病隐匿,早期诊断与精准干预是改善患者预后的关键。近年来,深度学习、大型语言模型及机器学习等数智技术日益融入GIC防治全流程,在早期筛查、临床诊断、临床分期及预后评估等核心环节取得显著进展,有效推动诊疗模式向精准化、智能化转型,临床应用前景广阔。然而,其应用仍面临数据偏倚与异质、研究设计局限、可解释性不足及隐私安全等挑战。为此,研究者们正积极推进多中心数据共享机制、完善设计框架、优化算法性能与制度建设,加速其向真实世界临床场景的高效转化。该文系统综述了数智技术在GIC防治中的临床应用进展与现存不足,并对未来发展方向进行展望,以期为构建高质量、智能化的GIC防治体系提供参考。

【Abstract】 Gastrointestinal cancer(GIC),including esophageal,gastric,and colorectal cancers,represents a major global health burden with high incidence and mortality rates.The disease often develops insidiously,and early diagnosis coupled with precise intervention is critical for improving patient outcomes.In recent years,intelligent digital technologies such as deep learning,large language models,and machine learning have been increasingly integrated into the full spectrum of GIC prevention and treatment.These technologies have achieved remarkable progress in key domains,including early screening,clinical diagnosis,staging,and prognostic evaluation,thereby driving the transformation of clinical practice toward precision and intelligence-based paradigms with promising prospects for clinical implementation.However,their application still faces several challenges,such as data bias and heterogeneity,study design limitations,lack of interpretability,and data privacy concerns.To address these issues,researchers are actively promoting multi-center data sharing mechanisms,improving study frameworks,and enhancing both algorithmic performance and regulatory frameworks,in order to accelerate their translation into real-world clinical settings.This review systematically summarizes the clinical applications,advances,and limitations of intelligent digital technologies in GIC prevention and treatment,and discusses future directions,aiming to provide insights for establishing a high-quality and intelligent GIC prevention and management system.

【基金】 国家重点研发计划中医药现代化专项(编号:2023YFC3503601);首都卫生发展科研专项项目(编号:首发2024-1-4163);中国中医科学院科技创新工程项目(编号:CI2023C011YL)
  • 【文献出处】 现代肿瘤医学 ,Journal of Modern Oncology , 编辑部邮箱 ,2026年06期
  • 【分类号】R735
  • 【下载频次】37
节点文献中: 

本文链接的文献网络图示:

本文的引文网络