Feasibility of predicting tumor motion using online data acquired during treatment and a generalized neural network optimized with offline patient tumor trajectories.

Journal: Medical physics
Published Date:

Abstract

PURPOSE: The accurate prediction of intrafraction lung tumor motion is required to compensate for system latency in image-guided adaptive radiotherapy systems. The goal of this study was to identify an optimal prediction model that has a short learning period so that prediction and adaptation can commence soon after treatment begins, and requires minimal reoptimization for individual patients. Specifically, the feasibility of predicting tumor position using a combination of a generalized (i.e., averaged) neural network, optimized using historical patient data (i.e., tumor trajectories) obtained offline, coupled with the use of real-time online tumor positions (obtained during treatment delivery) was examined.

Authors

  • Troy P Teo
    CancerCare Manitoba, 675 McDermot Avenue, Winnipeg, Manitoba, R3E 0V9, Canada.
  • Syed Bilal Ahmed
    CancerCare Manitoba, 675 McDermot Avenue, Winnipeg, Manitoba, R3E 0V9, Canada.
  • Philip Kawalec
    CancerCare Manitoba, 675 McDermot Avenue, Winnipeg, Manitoba, R3E 0V9, Canada.
  • Nadia Alayoubi
    CancerCare Manitoba, 675 McDermot Avenue, Winnipeg, Manitoba, R3E 0V9, Canada.
  • Neil Bruce
    Department of Computer Science, E2-445 EITC, University of Manitoba, Winnipeg, Manitoba, R3T 2N2, Canada.
  • Ethan Lyn
    Medical Board, Individual Customer, Great-West Life Assurance Company, 60 Osborne Street North, Winnipeg, Manitoba, R3C 1V3, Canada.
  • Stephen Pistorius
    CancerCare Manitoba, 675 McDermot Avenue, Winnipeg, Manitoba, R3E 0V9, Canada.