(Sunday, 7/14/2019)
Room: ePoster Forums
Purpose: To evaluate the dosimetric compacts when using three different ways of rigid registration between planning and reviewed CT in proton therapy for breast cancer
Methods: Nine patients previously treated at out center were enrolled in this study. The prescription dose to CTV was 50 GyE in 25 fractions, using pencil beam scanning technique. Three ways of rigid image registration were performed between planning CT and reviewed CT, which were grey level-based auto registration(GLAR) by Syngo (V13B, Siemens）treatment planning system, correspondences for the patient's sternum and ribs(CPSR) and correspondences for the metal fiducial points (CMFP) located at the body surface of the upper, lower and lateral border of the patients' breast. Doses of patient's CTV, the affected lung, inner breast on the contralateral side and left anterior descending(LAD) were assessed or dosimetric difference evaluation.
Results: The V95% of CTV was 97.84±1.69% in GLAR, 98.2±2.23% in CPSR and 98.31±1.69% in CMFP. In GLAR, the mean doses (V20GyE) of the patients' ipsilateral lung were 7.65±4.39 GyE (16.24±10.91%), 7.81±4.60 GyE (16.74±11.37%) in CPSR and 7.72±4.32 GyE (16.64±10.74%) in CMFP. The mean doses of the patients' LAD were 6.46±6.57GyE in GLAR, 6.20±5.68GyE in CPSR and 6.10±5.26GyE in CMFP. The mean doses of the contralateral inner breast were 10.06±5.68GyE, 10.01±4.90GyE, 9.81±4.74GyE in GLAR, CPSR and CMFP respectively.
Conclusion: All three ways of rigid registration could satisfy the dose coverage to CTV (at least 95% of the target volume covered by 95% prescription dose). Since CMFP can be a clear marker in 2D orthogonal portal vision images compared to sternum and rib, the combination of CMFP and CPSR may improve the accuracy of patient's alignment further without introducing higher doses to the OARs in proton breast therapy. However, verification is needed based on a larger dataset.
Keywords
Taxonomy
IM/TH- Image registration : CT
Contact Email
