Atlas-based semiautomatic target volume definition (CTV) for head-and-neck tumors

Gerd Strassmann1, Soulimane Abdellaoui, Detlef Richter

  • 1Department of Radiotherapy and Radiation Oncology, University of Marburg, Marburg, Germany. gstrassmann@t-online.de

Abstract

Insights

A new semiautomatic method significantly reduces head-and-neck cancer target delineation time by 51.8%. This atlas-based approach improves efficiency for radiation therapy planning.

Area of Science:

  • Medical Imaging
  • Radiation Oncology
  • Computational Anatomy

Background:

  • Accurate target delineation is crucial for effective radiation therapy in head-and-neck cancer.
  • Manual contouring is time-consuming and subject to inter-observer variability.
  • Existing automated methods may lack precision for complex anatomical regions.

Purpose of the Study:

  • To develop and evaluate a novel semiautomatic method for improving target volume delineation in head-and-neck cancer.
  • To reduce the time required for manual contouring in radiation therapy planning.
  • To assess the accuracy and efficiency of an atlas-based software with 3D correction.

Main Methods:

  • Implementation of an atlas-based software utilizing fourteen anatomic landmarks and CT slices for automatic delineation.
  • Application of affine transformation for image registration.
  • Evaluation through manual contour comparison and physician-led 3D correction on ten datasets.

Main Results:

  • Semiautomatic delineation achieved in a mean of 2.7 minutes, significantly faster than manual contouring (20.2 minutes).
  • Semiautomatic definition with rapid 3D correction took 9.7 minutes, a 51.8% time saving.
  • Mean similarity index of 77.2% for atlas adaptation, increasing to 85% after physician correction.

Conclusions:

  • A feasible semiautomatic contouring method using a target volume atlas and landmark model significantly reduces delineation time for head-and-neck cancer.
  • The developed software offers substantial time savings and maintains sufficient accuracy.
  • This method has broad applicability for tumors with stable target volumes, irrespective of respiration.