Related Experiment Video
Updated: May 30, 2026
09:41
A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery
Published on: May 20, 2016
Detection of patient setup errors with a portal image - DRR registration software application
Kenneth Sutherland1, Masayori Ishikawa, Gerard Bengua
1Hokkaido University Graduate School of Medicine, Sapporo, Japan. kensuth@med.hokudai.ac.jp
Journal of Applied Clinical Medical Physics
|August 17, 2011
Summary
This study evaluated software for aligning portal images with digitally reconstructed radiographs (DRRs) to improve radiation therapy setup error estimation. The custom software and its auto-registration feature significantly improved accuracy and consistency for thoracic cases.
Area of Science:
- Medical Physics
- Radiotherapy Technology
- Image Registration
Background:
- Accurate patient setup is crucial in radiation therapy to ensure precise dose delivery.
- Image-guided radiation therapy (IGRT) relies on aligning treatment fields with patient anatomy using imaging techniques.
- Portal imaging and digitally reconstructed radiographs (DRRs) are key components in IGRT for setup verification.
Purpose of the Study:
- To evaluate a custom software application for registering portal images with DRRs.
- To assess the software's impact on improving the accuracy and consistency of radiation therapy setup error estimation.
- To compare the performance of manual registration, software-assisted registration, and an automatic registration algorithm.
Main Methods:
- An anthropomorphic thoracic and pelvis phantom study was conducted with known setup errors.
- Physicians estimated setup errors side-by-side without software, while technicians used the custom registration software.
- An automatic registration algorithm based on mutual information was also evaluated and compared against human estimations.
Main Results:
- The registration software significantly improved setup error estimation accuracy for the thoracic phantom (2.1 mm vs. 4.3 mm).
- The automatic algorithm further enhanced accuracy for the thoracic phantom (0.8 mm).
- Consistency of estimations improved with the software for both thoracic and pelvis phantoms, though accuracy differences were not statistically significant for the pelvis case.
Conclusions:
- The custom portal image-DRR registration software statistically improves setup error estimation accuracy and consistency, particularly for thoracic cases.
- The automatic registration algorithm demonstrates high accuracy but requires further development for challenging cases like pelvis imaging due to contrast limitations.
- The developed software and its unique algorithms offer potential advancements in IGRT accuracy and efficiency.
