Influence of deformable image registration on 4D dose simulation for extracranial SBRT: A multi-registration...

Area of Science:

• Medical Physics
• Radiotherapy
• Image Processing

Background:

• 4D dose simulation in SBRT relies on accurate motion modeling.
• Deformable Image Registration (DIR) is crucial for estimating internal organ motion from medical images.

Purpose of the Study:

• To assess the impact of various open-source DIR algorithms on 4D dose simulation accuracy in extracranial Stereotactic Body Radiation Therapy (SBRT).
• To evaluate how DIR accuracy influences correspondence model performance and subsequent dose simulation outcomes.

Main Methods:

• Six open-source DIR frameworks were evaluated for registration accuracy using 4D imaging data.
• Correspondence models were developed by correlating breathing signals with internal motion.
• DIR algorithms were applied to patient 4D CT data for motion estimation, model building, and 4D dose simulation (VMAT).
• Deviations between static and 4D-simulated dose distributions were analyzed.

Main Results:

• DIR accuracy varied across different algorithms, with lower accuracy leading to reduced correspondence model accuracy.
• For lung metastases, 4D dose simulation results showed good agreement regardless of DIR accuracy.
• Liver metastases 4D dose simulation results demonstrated significant variability depending on the DIR approach used.

Conclusions:

• The choice of DIR algorithm critically influences 4D dose simulation outcomes, especially in low-contrast regions like the liver.
• DIR-based 4D dose simulation reliability is questionable without careful consideration of the applied DIR method.