Related Experiment Video
Updated: May 1, 2026

10:35
Stereotactic Radiosurgery for Gynecologic Cancer
Published on: April 17, 2012
22.0K
Automating RTOG-defined target volumes for postmastectomy radiation therapy.
Hiral P Fontanilla1, Wendy A Woodward1, Mary E Lindberg2
1Department of Radiation Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas.
Practical Radiation Oncology
|March 29, 2014
Summary
Deformable image registration-based breast segmentation (DEF-SEG) aids in contouring for postmastectomy radiation therapy. This automated technique shows promise for improving consistency in defining target structures.
Area of Science:
- Radiation Oncology
- Medical Imaging
- Computational Anatomy
Background:
- Accurate contouring of target structures in radiation therapy (RT) is crucial for conformal treatment, plan evaluation, and multi-institutional trial quality assurance.
- The Radiation Therapy Oncology Group (RTOG) has established consensus guidelines for contouring targets in postmastectomy RT.
- Automated contouring techniques can potentially improve consistency and efficiency.
Purpose of the Study:
- To evaluate the utility of deformable image registration-based breast segmentation (DEF-SEG) for contouring structures in postmastectomy RT according to RTOG guidelines.
- To assess the accuracy of DEF-SEG by comparing its auto-generated contours with manually modified contours based on RTOG definitions.
- To determine the dosimetric impact of differences between auto-segmented and manually adjusted contours.
Main Methods:
- RTOG definitions were used to contour the chest wall (CW), axillary nodes (Ax1-3), supraclavicular (SCV) nodes, internal mammary (IM) nodes, and heart.
- DEF-SEG was employed to generate auto-segmented contours from templates applied to CT scans of 20 postmastectomy cases (10 left, 10 right).
- Manual modifications were made to auto-segmented contours to match RTOG definitions, followed by comparison of overlap and evaluation of dosimetric impact using dose-volume histograms.
Main Results:
- Mean volume-overlap ratios between DEF-SEG and manual contours were high for the chest wall (0.91) and heart (0.93), but lower for lymph node regions (0.32-0.68).
- Mean differences in the volume receiving 45 Gy (V45) were generally small, with the largest observed for internal mammary nodes (10.1%).
- Mean differences in the volume receiving 10 Gy (V10) for the heart were minimal (0.4-0.5%), indicating minimal dosimetric impact.
Conclusions:
- DEF-SEG demonstrates utility in delineating structures for postmastectomy RT, particularly the chest wall and heart, aligning with RTOG consensus guidelines.
- No clinically significant dosimetric differences were identified between DEF-SEG auto-contours and manually adjusted contours.
- DEF-SEG holds potential as a valuable tool for evaluating treatment plans in multi-institutional postmastectomy RT trials, enhancing consistency and quality assurance.

