Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

3D Path Planning from a Single 2D Fluoroscopic Image for Robot Assisted Fenestrated Endovascular Aortic Repair

Jian-Qing Zheng, Xiao-Yun Zhou, Celia Riga, Guang-Zhong Yang

TL;DR

This work addresses the lack of 3D context in intraoperative Fenestrated Endovascular Aneurysm Repair (FEVAR) navigation from 2D fluoroscopy. It introduces a real-time pipeline that links preoperative 3D AAA skeletons to intraoperative 2D skeletons through graph matching and non-rigid registration using TPS with length and smoothness regularization, aided by deep learning-based AAA segmentation for automation. Validation on simulation, phantom, and patient data shows improved accuracy, robustness, and speed over ICP-TPS, including sub-second processing per fluoroscopic frame. The approach reduces radiation exposure and supports real-time navigation, and the authors provide open-source code.

Abstract

The current standard of intra-operative navigation during Fenestrated Endovascular Aortic Repair (FEVAR) calls for need of 3D alignments between inserted devices and aortic branches. The navigation commonly via 2D fluoroscopic images, lacks anatomical information, resulting in longer operation hours and radiation exposure. In this paper, a framework for real-time 3D robotic path planning from a single 2D fluoroscopic image of Abdominal Aortic Aneurysm (AAA) is introduced. A graph matching method is proposed to establish the correspondence between the 3D preoperative and 2D intra-operative AAA skeletons, and then the two skeletons are registered by skeleton deformation and regularization in respect to skeleton length and smoothness. Furthermore, deep learning was used to segment 3D pre-operative AAA from Computed Tomography (CT) scans to facilitate the framework automation. Simulation, phantom and patient AAA data sets have been used to validate the proposed framework. 3D distance error of 2mm was achieved in the phantom setup. Performance advantages were also achieved in terms of accuracy, robustness and time-efficiency. All the code will be open source.

3D Path Planning from a Single 2D Fluoroscopic Image for Robot Assisted Fenestrated Endovascular Aortic Repair

TL;DR

This work addresses the lack of 3D context in intraoperative Fenestrated Endovascular Aneurysm Repair (FEVAR) navigation from 2D fluoroscopy. It introduces a real-time pipeline that links preoperative 3D AAA skeletons to intraoperative 2D skeletons through graph matching and non-rigid registration using TPS with length and smoothness regularization, aided by deep learning-based AAA segmentation for automation. Validation on simulation, phantom, and patient data shows improved accuracy, robustness, and speed over ICP-TPS, including sub-second processing per fluoroscopic frame. The approach reduces radiation exposure and supports real-time navigation, and the authors provide open-source code.

Abstract

The current standard of intra-operative navigation during Fenestrated Endovascular Aortic Repair (FEVAR) calls for need of 3D alignments between inserted devices and aortic branches. The navigation commonly via 2D fluoroscopic images, lacks anatomical information, resulting in longer operation hours and radiation exposure. In this paper, a framework for real-time 3D robotic path planning from a single 2D fluoroscopic image of Abdominal Aortic Aneurysm (AAA) is introduced. A graph matching method is proposed to establish the correspondence between the 3D preoperative and 2D intra-operative AAA skeletons, and then the two skeletons are registered by skeleton deformation and regularization in respect to skeleton length and smoothness. Furthermore, deep learning was used to segment 3D pre-operative AAA from Computed Tomography (CT) scans to facilitate the framework automation. Simulation, phantom and patient AAA data sets have been used to validate the proposed framework. 3D distance error of 2mm was achieved in the phantom setup. Performance advantages were also achieved in terms of accuracy, robustness and time-efficiency. All the code will be open source.

Paper Structure

This paper contains 18 sections, 9 equations, 9 figures, 2 tables, 2 algorithms.

Table of Contents

  1. INTRODUCTION
  2. Methodology
  3. Pre-operative 3D and Intra-operative 2D Skeleton Extraction
  4. Graph Matching
  5. Skeleton Graph Construction
  6. Node Classification
  7. Branch and Trunk Matching
  8. Branch Node and Trunk Node Assigning
  9. Skeleton Deformation
  10. Iterative Optimization
  11. Post-deformation
  12. Experiment and Validation
  13. Results
  14. Simulation Data Experiment
  15. Phantom Data Experiment
  16. ...and 3 more sections

Figures (9)

  • Figure 1: An AAA example (a) where the inlarged part is near renal arteries and a fenestrated stent graft example (b) with fenestrations and scallops.
  • Figure 2: Pipeline for the proposed real-time 3D robotic path planning.
  • Figure 3: (a) Experiments on a phantom, with a pumper simulating the blood circulation; (b) Deformation of a phantom, using a string
  • Figure 4: The comparison of the performance in 20 simulation cases between the proposed method, ICP-TPS and the shape variation (SV), with mean$\pm$std of 2D(a)/3D(b) distance errors in 2D(a)/3D(b) environment.
  • Figure 5: The comparison of the performance in one simulation data with additional translations (a) and rotations (b) between the proposed method, ICP-TPS and the shape variation (SV), with mean$\pm$std of 2D/3D distance errors in 2D(left)/3D(right) environment.
  • ...and 4 more figures

Theorems & Definitions (4)

  • Definition 1
  • Definition 2
  • Definition 3
  • Definition 4