Scholarlink

<p>During video-assisted thoracoscopic surgery (VATS), surgical tools enter the patient’s body through a trocar placed at the incision position. Valid tool paths preserve the entry location, i.e., meet the Remote Center of Motion (RCM) constraint, and prevent collisions with sensitive organs. This paper proposes a surgical planning system for generating and visualizing good candidates for the incision points and path trajectories. In this task, the constraints regard both to points of the different parts of the articulated device and the joint angles. Because of the complex topology of the high-dimensional space of non-colliding tool poses, Monte Carlo sampling is applied. First, valid gripping poses are obtained by two complementary strategies. The target-to-incision method finds the gripping frame and samples the joint angles to determine the corresponding incision points. The incision-to-target approach, on the other hand, samples the incision point and searches for gripping frames and joint angles consistent with it. Having the gripping pose and the incision point, a start pose is set at the incision, and paths are built backwards from the end pose toward the start pose to ensure collision avoidance and smooth trajectories. Internal key frames of the path are initialized by interpolation and checked for validity. Invalid key frames are tried to be corrected by a set of collision resolution algorithms. Both good paths of valid key frames and bad paths containing also invalid key frames can be visualized on screen or by Augmented Reality (AR) headsets, making the results explainable. We also consider the multi-hole and multi-target scenarios. The proposed solution has been integrated into the <i>GAMMA Surgical Guidance</i><InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(^\textrm{TM}\)</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mrow /> <mrow /> <mtext>TM</mtext> </mmultiscripts> </math></EquationSource> </InlineEquation> software.</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

Virtual reality support for thoracoscopic surgery design

  • 期刊论文

  • 发表日期：

    2025年12月13日

查看全文
  • L. Szirmay-Kalos,
  • A. Fridvalszky,
  • L. Szécsi,
  • D. Varnyú,
  • J. I. Barabás,
  • L. Szilassy,
  • À. Ghimessy

摘要

During video-assisted thoracoscopic surgery (VATS), surgical tools enter the patient’s body through a trocar placed at the incision position. Valid tool paths preserve the entry location, i.e., meet the Remote Center of Motion (RCM) constraint, and prevent collisions with sensitive organs. This paper proposes a surgical planning system for generating and visualizing good candidates for the incision points and path trajectories. In this task, the constraints regard both to points of the different parts of the articulated device and the joint angles. Because of the complex topology of the high-dimensional space of non-colliding tool poses, Monte Carlo sampling is applied. First, valid gripping poses are obtained by two complementary strategies. The target-to-incision method finds the gripping frame and samples the joint angles to determine the corresponding incision points. The incision-to-target approach, on the other hand, samples the incision point and searches for gripping frames and joint angles consistent with it. Having the gripping pose and the incision point, a start pose is set at the incision, and paths are built backwards from the end pose toward the start pose to ensure collision avoidance and smooth trajectories. Internal key frames of the path are initialized by interpolation and checked for validity. Invalid key frames are tried to be corrected by a set of collision resolution algorithms. Both good paths of valid key frames and bad paths containing also invalid key frames can be visualized on screen or by Augmented Reality (AR) headsets, making the results explainable. We also consider the multi-hole and multi-target scenarios. The proposed solution has been integrated into the GAMMA Surgical Guidance \(^\textrm{TM}\) TM software.