Deep learning-based reconstruction of interventional tools and devices from four X-ray projections for tomographic interventional guidance.

Journal: Medical physics
Published Date:

Abstract

PURPOSE: Image guidance for minimally invasive interventions is usually performed by acquiring fluoroscopic images using a monoplanar or a biplanar C-arm system. However, the projective data provide only limited information about the spatial structure and position of interventional tools and devices such as stents, guide wires, or coils. In this work, we propose a deep learning-based pipeline for real-time tomographic (four-dimensional [4D]) interventional guidance at conventional dose levels.

Authors

  • Elias Eulig
    German Cancer Research Center (DKFZ), Heidelberg, Germany.
  • Joscha Maier
    German Cancer Research Center (DKFZ), Heidelberg, Germany.
  • Michael Knaup
    Division of X-Ray Imaging and Computed Tomography, German Cancer Research Center (DKFZ), Heidelberg, Germany.
  • N Robert Bennett
    Department of Radiology, Stanford University, Stanford, California, USA.
  • Klaus Hörndler
    Ziehm Imaging GmbH, Nürnberg, Germany.
  • Adam S Wang
    Department of Radiology, Stanford University, Stanford, California, USA.
  • Marc Kachelrieß
    German Cancer Research Center, Heidelberg, 69120, Germany.