Clinician-AI Collaboration in Endovascular Neurointerventions: In-Vitro User Study of Copilot Assistance Modes for Robotic Navigation Across the Autonomy Spectrum

📄 arXiv: 2611.03446v1 📥 PDF

作者: Nikola Fischer, Harry Robertshaw, S. M. Hadi Sadati, Xingyu Chen, Weiyuan Deng, Benjamin Jackson, Michael Adofo Kwakye, Sara Sciacca, Naga Kandasamy, Parthiban Balasundaram, Sahan Guruge, Charles Cummins, Joe Leyon, Nasr Abdelsalam, Matthew Benger, Andrew Falzon, Alejandro Granados, Christos Bergeles, Thomas Booth

分类: cs.RO

发布日期: 2026-11-02

备注: 13 pages, 10 figures, 4 tables


💡 一句话要点

提出Copilot辅助模式以提升机器人神经介入导航的临床协作

🎯 匹配领域: 支柱一:机器人控制 (Robot Control)

关键词: 机器人导航 神经介入 AI辅助 共享自主 临床协作 机械取栓术 用户研究

📋 核心要点

  1. 现有的机器人神经介入技术在自主导航方面存在性能不足,尤其是在真实环境中的表现较差。
  2. 本研究提出了Copilot辅助模式,旨在通过不同的自主级别实现临床医生与AI的有效协作。
  3. 实验结果表明,手动导航表现最佳,而共享自主模式的成功率显著高于完全自主导航,显示出协作的重要性。

📝 摘要(中文)

机械取栓术(MT)是急性缺血性中风的有效治疗方法,但由于神经介入专业知识的缺乏,获取仍然有限。远程机器人和AI辅助系统可能有助于解决这一挑战,但临床医生如何与不同自主水平的系统互动仍不明确。本研究提出了Copilot辅助模式(CAMs),为机器人内血管导航中的临床医生与AI协作提供了框架。通过对11名专家和2名学生进行的首次体外用户研究,评估了六种导航模式的表现。结果显示,手动导航的成功率最高(100%),而完全自主导航在体外的成功率仅为23%。相对而言,几种共享自主模式的成功率为92-97%,显示出临床医生与AI协作在减轻仿真与现实之间的差距方面的有效性。

🔬 方法详解

问题定义:本研究旨在解决机器人神经介入导航中临床医生与AI协作的有效性问题,现有方法在真实环境中的表现不佳,特别是完全自主导航的成功率低。

核心思路:提出Copilot辅助模式(CAMs),通过不同的自主级别实现临床医生与AI的协作,强调共享控制的重要性,而不仅仅是提高自主性。

技术框架:研究设计包括六种导航模式的评估,涵盖从手动操作到完全自主的不同级别。使用患者特定的血管模型进行体外实验,结合客观性能指标与主观评估。

关键创新:最重要的创新在于提出了共享自主模式,这些模式的成功率显著高于完全自主导航,表明临床医生的参与对提高导航性能至关重要。

关键设计:实验中评估了不同的任务委派和旋转共享控制策略,重点关注用户接受度与性能之间的平衡,具体参数设置和损失函数未详细披露。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果显示,手动导航的成功率为100%,而完全自主导航在体外的成功率仅为23%。共享自主模式的成功率为92-97%,显著高于完全自主导航,表明协作模式在实际应用中的优势。

🎯 应用场景

该研究的潜在应用领域包括急性缺血性中风的治疗和机器人手术导航。通过提高临床医生与AI的协作效率,可能会改善患者的治疗效果,并推动机器人技术在医疗领域的进一步应用与发展。

📄 摘要(原文)

Mechanical thrombectomy (MT) is an effective treatment for acute ischemic stroke, yet access remains limited by the availability of neurointerventional expertise. Telerobotic and AI-assisted systems may help address this challenge, but it remains unclear how clinicians interact with different levels of autonomy and how autonomous navigation systems perform in realistic environments. This work introduces Copilot Assistance Modes (CAMs), a framework for clinician-AI collaboration in robotic endovascular navigation spanning the autonomy spectrum from manual operation to full autonomy. In the first in-vitro user study comparing clinician-AI collaboration strategies, 11 experts and two students evaluated six navigation modes in a patient-specific vascular phantom. Objective performance metrics were combined with subjective assessments, including workload and usability. Autonomous performance was additionally evaluated both in silico and in vitro to investigate sim-to-real transfer. Hands-on manual navigation achieved the strongest overall performance (100% success). While fully autonomous navigation achieved 81% success in simulation, performance dropped to 23% in vitro. In contrast, several clinician-AI shared-autonomy modes achieved success rates of 92-97%, approximately four times higher than full autonomy and comparable to direct robotic control (95%), demonstrating mitigation of sim-to-real domain shift through clinician-AI collaboration. Among the evaluated AI-driven approaches, user-directed task delegation and rotational shared control demonstrated the most favorable balance between performance and clinician acceptance. These findings suggest that effective clinician-AI collaboration and allocation of control authority may currently be more important for reliable robotic neurointerventional navigation than increasing autonomy alone.