Room: Exhibit Hall | Forum 5
Purpose: To develop a robust algorithm that uses the displacement-vector-fields (DVF) from deformable image registration (DIR) on a voxel-by-voxel basis to perform 4D-dose optimization and calculation.
Methods: A thorax mobile phantom with three target inserts: small (10mm), medium (20mm) and large (40mm) was moved with cyclic motion patterns with varying amplitudes (0-20mm) and frequencies (0-0.5Hz. A simulation study was performed to map the CT-number values and dose distributions using the DVF that are obtained from DIR algorithms. The DVF represented voxel-by-voxel motion shifts of the tumor and surrounding organs-at-risk. A robust algorithm was developed that employed the DVF to modify the CT-number values in the CT-images and the fluence-maps of the dose distributions to perform 4D-optimization and dose calculation.
Results: The algorithm produced accurate volumes and CT-number values of the mobile targets. This algorithm provided corrections for anatomical changes and respiratory motion by remapping the CT-number values and 3D-optimized dose distributions achieving adaptive radiation therapy. The optimized dose distributions were shrunk by algorithm where they were expanded by phantom motion during dose delivery to achieve conformal 4D-dose distributions for mobile tumors. This technique required the following (a) acquisition of 4D-CT images during CT images, (b) extraction of DVF using deformable image registration during the treatment planning process (c) reproducibility of the patient breathing motion patterns between CT-simulations and dose delivery and (d) dose delivery over multiple respiratory cycles preferably intermediate or low dose rate.
Conclusion: Adaptive radiation therapy can be achieved by optimization using the voxel-by-voxel DVF induced by respiratory extracted from the 4D-CT images. This approach provides alternative approach for considering an internal target margin using 4D-CT images in the treatment planning process; and beam gating with external or internal makers during dose delivery. This robust dose optimization and calculation algorithm has potential clinical applications in adaptive radiation therapy.
Not Applicable / None Entered.