Background: The increasing adoption of artificial intelligence (AI) and augmented reality (AR) within vascular neurosurgery has become a prominent trend. The primary challenge before us is seamlessly integrating these advanced concepts and developing them further to improve patient outcomes. Methods: We combined peer-reviewed publications of our research group over the past 5 years with current research projects to form the basis of a narrative discussion, aiming to better understand drawbacks, challenges, and the developmental steps to be followed. Results: Four developmental phases were identified: (1) the integration of AI and AR to create adequate three-dimensional (3D) segmentations; (2) adding flow and pulsatility data to create 5D segmentations; (3) treatment planning in these models; and (4) treatment guidance using these models. The main drawback described was the limited added value in the microscopic phase of neurovascular surgery due to view obstructions and a lack of accuracy. The main challenge described was the current limitation in computational and graphical processing capabilities. Conclusion: Although drawbacks and challenges still exist, AI and AR are rapidly developing topics within vascular neurosurgery. The research in this field could lay the groundwork for fully automatized treatment strategies in the future.

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

Artificial Intelligence and Augmented Reality in Vascular Neurosurgery

  • Tristan van Doormaal,
  • Elisa Colombo,
  • Tim Fick,
  • Jesse A. M. van Doormaal,
  • Tessa M. Kos,
  • Mathijs de Boer,
  • Pierre Robe,
  • Eelco Hoving,
  • Lambertus W. Bartels,
  • Luca Regli

摘要

Background: The increasing adoption of artificial intelligence (AI) and augmented reality (AR) within vascular neurosurgery has become a prominent trend. The primary challenge before us is seamlessly integrating these advanced concepts and developing them further to improve patient outcomes. Methods: We combined peer-reviewed publications of our research group over the past 5 years with current research projects to form the basis of a narrative discussion, aiming to better understand drawbacks, challenges, and the developmental steps to be followed. Results: Four developmental phases were identified: (1) the integration of AI and AR to create adequate three-dimensional (3D) segmentations; (2) adding flow and pulsatility data to create 5D segmentations; (3) treatment planning in these models; and (4) treatment guidance using these models. The main drawback described was the limited added value in the microscopic phase of neurovascular surgery due to view obstructions and a lack of accuracy. The main challenge described was the current limitation in computational and graphical processing capabilities. Conclusion: Although drawbacks and challenges still exist, AI and AR are rapidly developing topics within vascular neurosurgery. The research in this field could lay the groundwork for fully automatized treatment strategies in the future.