Room: Exhibit Hall | Forum 5
Purpose: To introduce the implementation of image guidance through of the feedback based voluntary moderate breath hold (BH) proton pencil beam scanning treatment.
Methods: Three CT scannings to build the patientâ€™s anatomy model are performed during the patientâ€™s voluntary moderate BHs, where the BH plateau reaches 100%, 90% and 110%. All three CT scans are involved in the optimization during the treatment planning process, so that the final nominal plan is robust for a BH range. Both orthogonal kV imaging and pausable CBCT are implemented for patient alignment with BH prior to the treatment. BH CBCT on the patient is compared with planning CT to verify the alignment of tumor location and the surrounding patientâ€™s anatomy. KV imaging is used to verify the patient anatomy before beam on for each field. The CBCT images are analyzed for the BH reproducibility and virtual delivered dose (VDD) retrospectively. The rigid image registration between CBCT and planning CT (pCT) helps to find the BH reproducibility. To find the VDD, a series of deformable image registrations (DIR) are performed between CBCT and pCT. The dose calculations are performed on the deformed pCT images which own the patient geometry on the treatment day and calibrated HU values. The variation of lung density is not considered in the DIR. Therefore two additional dose calculations on the deformed CT images with intentional lung density change of +/- 15% are done and evaluated.
Results: The values of the mean, standard deviation, maximum and minimum of the tumor location difference between the CBCT and pCT are 1.9, 1.6, 4.7 and 0 mm respectively. The percentage difference in targets D95 of VDD is within 1.5% of nominal plan.
Conclusion: BH technique can be implemented in proton pencil beam scanning treatment safely and effectively with the implementation of image guidance.