Automatic Organ Localization for Adaptive Radiation Therapy for Prostate Cancer

Abstract

The goal of this research is to develop automated methods to analyze daily CT scans taken during prostate therapy, in order to make adaptive radiation therapy (ART) more effective and readily implemented. The central technique is deformable image registration, which allows a CT image acquired for treatment planning to be deformed to match each daily image. The resulting deformations can then be applied to contours drawn at planning time, to generate segmentations of treatment images. Inverse deformations can also be applied to dose distributions (represented as images indicating the dose delivered to each voxel). In this way, the dose distributions can be deformed into the reference frame of the planning image, and summed over the course of treatment to determine the total delivered dose. Over the past year we have demonstrated both of these techniques on data from multiple subjects.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 2005
Accession Number
ADA442678

Entities

People

  • Brad Davis
  • Mark Foskey
  • Sarang C. Joshi

Organizations

  • University of North Carolina at Chapel Hill

Tags

DTIC Thesaurus Topics

  • Accuracy
  • Algorithms
  • Anatomy
  • Image Processing
  • Image Registration
  • Medical Personnel
  • Neoplasms
  • Oncology
  • Pattern Recognition
  • Prostate
  • Prostate Cancer
  • Radiation Oncology
  • Radiotherapy
  • Statistical Analysis
  • Three Dimensional
  • Tissues
  • X-Ray Computed Tomography

Fields of Study

  • Medicine
  • Physics

Readers

  • Computer Vision.
  • Medical Imaging.
  • Nuclear and Radiation Engineering.