Room: AAPM ePoster Library
Purpose: is not yet well understood how uncertainties associated with deformable registration impact dose calculations and the subsequent clinical outcome. The objective of this study was to investigate the correlation between the dose to the healthy tissue and clinical outcome using rigid vs. deformable registration.
Methods: lung cancer patients treated with three-dimensional conformal radiation therapy were analyzed for this study. Follow-up CT images as well as the corresponding radiation pneumonitis (RP) scores were available for patients at 3-6 months post-treatment. Planning CTs and associated dose maps were registered using rigid and deformable registrations to match follow-up CTs. Dosimetric parameters (MLD, V13, V20, V30) were used to compare rigid and deformable dose registrations in normal lung tissue. Patients were grouped into radiation pneumonitis CTCAE grades =1 and grades =2. Mann-Whitney test was performed between the two groups for MLD, V13, V20, and V30 obtained from both dose registration techniques.
Results: from the Mann-Whitney test showed no significant differences between the dosimetric parameters of rigid and deformable dose registrations. The correlation between the MLD of the deformed dose with RP was 47% higher than between the MLD of the rigid dose with RP. The two different registrations resulted in differences in MLD, V13, V20, and V30 of < 4.0+1.1%. Results showed that when the lung volume increases at follow-up, the dosimetric parameters were higher for deformable vs. rigid registration, when the lung volume decreased, the dosimetric parameters had lower values, and when the lung volume stayed the same, these values were similar for both registrations.
Conclusion: analysis of dose mapping indicates that there are no significant differences when comparing rigid and deformable dose. However, differences in dosimetric parameters are seen in individual case depending on the change in lung volume from the planning CT to the follow-up CT.