人工智能在胰腺外科中的应用
作者:
通讯作者:
作者单位:

南京大学医学院附属鼓楼医院 胰腺与代谢外科,江苏 南京 210008

作者简介:

仇毓东负责文章终审及论文架构设计;毛谅负责文章修改及论文撰写指导;蔡正华负责文章初稿及文献调研。

基金项目:

江苏省南京市卫生科技发展专项基金资助项目(ZDXX25132)。


Application of artificial intelligence in pancreatic surgery
Author:
Affiliation:

Department of Pancreatic and Metabolic Surgery, Nanjing Drum Tower Hospital, Nanjing 210008, China

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 音频文件
  • |
  • 视频文件
    摘要:

    人工智能(AI)正快速融入胰腺外科诊疗全流程,推动这一高难度领域向精准化与智能化转型。术前,基于深度学习的影像分割与三维重建技术可实现胰腺肿瘤及周围血管的精准识别与空间关系可视化,为手术规划及可切除性评估提供客观依据;术中,增强现实导航及实时组织识别等技术有助于关键解剖结构定位与风险预警,从而提高手术的精准性与安全性;术后,AI通过整合多源临床数据构建并发症风险预测模型,实现早期预警与个体化管理。然而,数据异质性、模型可解释性不足及临床转化路径缺乏标准化,仍是制约其广泛应用的关键瓶颈。本文系统梳理AI在胰腺外科术前规划、术中导航与决策及术后并发症管理中的应用进展,分析当前面临的主要挑战,并提出相应发展策略。未来应聚焦高质量多中心标准化数据库建设、多模态融合与可解释性模型研发,以及“感知-决策-执行”闭环智能手术系统构建,以推动AI技术的规范化应用与临床落地。

    Abstract:

    Artificial intelligence (AI) is rapidly being integrated into the entire workflow of pancreatic surgery, driving the transition toward precision and intelligent surgery in this highly complex field. In the preoperative setting, deep learning-based image segmentation and three-dimensional reconstruction enable precise delineation of pancreatic tumors and surrounding vasculature, facilitating surgical planning and resectability assessment. Intraoperatively, technologies such as augmented reality navigation and real-time tissue recognition assist in anatomical localization and risk identification, thereby improving surgical accuracy and safety. Postoperatively, AI-driven models integrating multi-source clinical data allow early prediction of complications and support individualized management. However, several challenges remain, including data heterogeneity, limited model interpretability, and the lack of standardized pathways for clinical translation. This review systematically summarizes the current applications of AI in preoperative planning, intraoperative navigation and decision-making, and postoperative complication management in pancreatic surgery, and further discusses existing challenges and future directions. Future efforts should focus on the development of high-quality multicenter standardized databases, multimodal and interpretable AI models, and closed-loop intelligent surgical systems integrating perception, decision-making, and execution, to facilitate the standardized implementation and clinical translation of AI technologies.

    参考文献
    相似文献
    引证文献
引用本文

蔡正华,毛谅,仇毓东.人工智能在胰腺外科中的应用[J].中国普通外科杂志,2026,35(3):420-428.
DOI:10.7659/j. issn.1005-6947.260097

复制
文章指标
  • 点击次数:
  • 下载次数:
历史
  • 收稿日期:2026-02-24
  • 最后修改日期:2026-03-11
  • 录用日期:
  • 在线发布日期: 2026-05-11