Related Experiment Video
Updated: Feb 23, 2026
08:54
Treatment of Liver Metastases Using an Internal Target Volume Method for Stereotactic Body Radiotherapy
Published on: May 8, 2018
15.1K
Implementing Radiation Dose-Volume Liver Response in Biomechanical Deformable Image Registration
Daniel F Polan1, Mary Feng2, Theodore S Lawrence1
1Department of Radiation Oncology, University of Michigan, Ann Arbor, Michigan.
International Journal of Radiation Oncology, Biology, Physics
|September 3, 2017
Summary
This study improved liver radiation therapy by modeling anatomical changes using radiation dose. This enhances personalized medicine by enabling more accurate treatment planning and dose accumulation for patients requiring re-treatment.
Area of Science:
- Medical Physics
- Radiotherapy
- Medical Imaging
Background:
- Accurate modeling of liver anatomical and functional changes from radiation therapy is crucial for improving normal tissue complication probability models.
- Current deformable image registration (DIR) techniques do not account for dose-dependent volumetric liver changes, compromising image linking for follow-up and re-treatment.
- Advancing personalized medicine requires improved methods to predict and manage these anatomical alterations.
Purpose of the Study:
- To investigate the use of delivered radiation dose and patient factors to biomechanically model longitudinal liver anatomical changes.
- To improve deformable image registration (DIR) techniques by incorporating dose-volume response for more accurate modeling of liver anatomy.
- To enhance follow-up care and re-treatment planning in liver radiation therapy.
Main Methods:
- Developed population models of hepatic volume response to radiation dose using retrospective data from 33 patients.
- Enhanced a biomechanical DIR algorithm by adding boundary conditions based on dose-volume response.
- Evaluated the improved DIR technique on CT images from 7 patients, assessing volumetric change prediction, vessel bifurcation tracking, and target registration error.
Main Results:
- The proposed DIR technique accurately modeled volumetric liver lobe deformations within contour variability.
- Achieved an average target registration error of 7.3 mm, a significant improvement over the previous DIR algorithm (10.9 mm).
- The superior-inferior component of the error (5.1 mm) was within the average limiting slice thickness (6.0 mm).
Conclusions:
- Demonstrated the feasibility of incorporating dose-driven volumetric response into deformable registration for liver anatomy.
- The improved DIR accuracy facilitates better dose accumulation, especially for patients needing re-treatment.
- This approach holds potential for advancing personalized radiation medicine for liver cancer.
