文献速递胃肠外科围术期管理与 ERAS

首例实时人工智能联合单孔机器人结直肠手术病例报告

The first reported case using real-time artificial intelligence with single-port robotics for colorectal surgery.
PubMed 文献 · Surg Endosc · 2026-03-16

本病例报告描述一例利用单孔机器人平台在造口部位对接,并将深度学习计算机视觉人工智能整合至机器人视觉输出系统,用于回肠造口还纳术。人工智能实时识别腹腔内组织并以颜色编码显示,辅助安全分离,单孔机器人提供良好活动范围并利用原造口部位避免额外切口。报告认为这是首例成功整合人工智能与单孔机器人平台完成结直肠手术,技术可行。

要点

  • 首例将实时计算机视觉人工智能与单孔机器人平台整合用于结直肠手术。
  • 人工智能颜色编码实时识别组织,辅助术中安全分离。
  • 单孔机器人经原造口部位对接,避免额外切口。

核心结论

  • 本报告首次成功整合计算机视觉与深度学习软件的人工智能与单孔机器人平台完成结直肠手术,展示联合技术可行性及在胃肠外科扩展应用的潜力。

研究设计与样本

  • 病例报告:一例选择单孔机器人平台经造口部位对接行回肠造口还纳术的患者,术中整合深度学习软件人工智能至机器人视觉输出系统。

主要发现

  • 人工智能准确精确识别术中确定需离断或保留的各个组织 structures。
  • 人工智能视觉显示以颜色编码,易于解读,有助于辅助安全分离。
  • 单孔机器人便于进入腹腔,各象限活动范围良好,成功游离必要肠管以安全完成吻合。
  • 利用原造口部位,避免额外切口。

临床怎么用

  • 本报告提示实时人工智能联合单孔机器人平台在结直肠手术中技术可行,可辅助组织识别与安全分离,并利用原造口部位减少额外切口。

局限性

  • 本报告为单例病例报告,未报告样本量、对照、随访或统计比较,技术可行性与临床获益需进一步研究验证。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: With the increased utilization of artificial intelligence (AI) tools in the operating room, we aimed to integrate real-time visual cues provided by AI (computer vision) with the Single-Port (SP) robotic platform for colorectal surgery. METHODS: A patient was selected for an end-ileostomy reversal using a Single-Port Robotic platform docked at the stoma site. Novel AI technology using deep-learning software was integrated into the robotic visual output system to identify different tissues within the abdominal cavity. Visual display of the information processed by the software was projected in real-time onto monitors and organized according to pre-selected color schemes. RESULTS: The AI technology accurately and precisely identified the individual tissue structures that had been determined intraoperatively appropriate to transect or preserve. The visual display of the AI software was color coded and easy to interpret, helping to aid safe dissection. The use of the SP robot allowed ease of access into the abdomen, providing an excellent range of motion in all quadrants, and successful mobilization of the necessary bowel to safely perform an anastomosis. This also took advantage of the prior stoma site, avoiding the need for additional incisions. CONCLUSION: This report demonstrates the first successful integration of artificial intelligence using computer vision and deep learning software with the Single-Port Robotic platform to perform colorectal surgery. The technical feasibilities of these combined modalities illustrate their potential for future use in an expanded role for gastrointestinal surgery.

来源:Surg Endosc,查看原文。版权归原作者所有。

引用

  1. Fortunato R, Wady H, Gebran S, et al. The first reported case using real-time artificial intelligence with single-port robotics for colorectal surgery. Surg Endosc. 2026. PMID: 41840080. doi:10.1007/s00464-026-12688-7

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。