Scholarlink

<p>Complex surgical procedures may benefit from AI systems that integrate visual and textual data for holistic scene understanding. We present Surgical RARP Copilot, a vision-language model for robot-assisted radical prostatectomy (RARP) that enables open question answering during surgery. We adapted a large language model to RARP literature and used it to generate a dataset of RARP images paired with ~1 million Q&amp;A examples to train the model. Performance was evaluated for open-domain Q&amp;A, surgical phase recognition, and instrument detection, and the system was deployed and tested in real time during a live operation—the first surgical VLM implemented in live robotic surgery. On unseen RARP procedures, Copilot showed robust performance across tasks. This work demonstrates feasible real-time AI guidance and suggests benefits for training, team communication, and knowledge support; future work includes broadening procedures and measuring clinical impact of such a system.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Surgical RARP copilot: a vision language model for robot-assisted radical prostatectomy

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月1日

查看全文
  • Wouter Bogaert,
  • François Remy,
  • Javier Gamazo Tejero,
  • Sean Huver,
  • Edoardo Beatrici,
  • Frederiek D’Hondt,
  • Niki Rashidian,
  • Mahdi Azizian,
  • Tony Belpaeme,
  • Alexandre Mottrie,
  • Pieter De Backer

摘要

Complex surgical procedures may benefit from AI systems that integrate visual and textual data for holistic scene understanding. We present Surgical RARP Copilot, a vision-language model for robot-assisted radical prostatectomy (RARP) that enables open question answering during surgery. We adapted a large language model to RARP literature and used it to generate a dataset of RARP images paired with ~1 million Q&A examples to train the model. Performance was evaluated for open-domain Q&A, surgical phase recognition, and instrument detection, and the system was deployed and tested in real time during a live operation—the first surgical VLM implemented in live robotic surgery. On unseen RARP procedures, Copilot showed robust performance across tasks. This work demonstrates feasible real-time AI guidance and suggests benefits for training, team communication, and knowledge support; future work includes broadening procedures and measuring clinical impact of such a system.