Feasibility of real-time lung tumor motion monitoring using intrafractional ultrasound and kV cone beam projection images

Med Phys. 2018 Oct;45(10):4619-4626. doi: 10.1002/mp.13104. Epub 2018 Aug 24.

Abstract

Purpose: The ability to monitor intrafractional tumor motion is essential for radiation therapy of thoracic and abdominal tumors. This study aims to develop a method to track lung tumor motion using intrafractional continuous ultrasound (US) and periodic cone-beam projection images (CBPI).

Methods: Time-sequenced b-mode US and CBPI data were extracted from the data acquired with the Clarity® and XVI platforms on an Elekta linac, respectively. The data were synchronized through a video capture card (VCE-PRO, IMPERX Inc.) which was triggered by the XVI system. In this way, a system was configured to allow real-time acquisition of the diaphragm position synchronized with periodic acquisition of the lung tumor position. Feasibility of the system was demonstrated by acquiring synchronized data on an in-house motion platform with embedded spheres of different materials and US images of the diaphragm on 5 volunteers of various body sizes. Finally, ultrasound b-mode images and CBPI were also acquired simultaneously from 3 lung cancer patients.

Results: Diaphragm motion monitoring under free breathing (FB) was successful with intracostal US imaging. We observed that diaphragm visualization decreased with the increase in the body size of the volunteer. The US system was able to track the motion as small as 2 mm in the phantom. The intrafractional CBPI acquired during VMAT delivery was successfully synchronized with US acquisition in a phantom study. Collected patient data showed a significant correlation between diaphragm motion, an internal surrogate monitored by US, and the tumor motion in superior-inferior (SI) direction monitored by XVI (P ˂ 0.0001).

Conclusions: The feasibility of real-time lung tumor motion tracking in SI direction with continuous ultrasound and periodic CBPI was demonstrated. The real-time estimation of the target position from the two streams for lung cancer patients would enable respiration gating or tracking during SBRT.

Keywords: lung tumor motion monitoring; ultrasound; x-ray imaging.

MeSH terms

  • Cone-Beam Computed Tomography
  • Dose Fractionation, Radiation*
  • Feasibility Studies
  • Humans
  • Lung Neoplasms / diagnostic imaging*
  • Lung Neoplasms / physiopathology
  • Lung Neoplasms / radiotherapy*
  • Movement*
  • Phantoms, Imaging
  • Radiotherapy, Image-Guided*
  • Respiration
  • Time Factors
  • Ultrasonography