SU-E-J-202: Treatment Dose Assessment with the Consideration of Radiation Dose Inducible Organ Shrinkage/Deformation.

PURPOSE For asymmetric organ shrinkage caused by heterogeneity dose distribution, tissue elasticity based deformable organ registration cannot be directly applied to the treatment image to construct the treatment dose. A novel approach is proposed to include radiation dose inducible organ shrinkage/deformation in the consideration of treatment dose assessment. METHODS A model and algorithm were developed to include dose inducible organ shrinkage in the deformable image registration. In the model, we assume that the logarithm of tissue-element volume shrinkage ratio is proportional to the LQ survival function, and the tissue-element displacement field is the gradient of a scalar potential function. The derived Poisson equation of the potential was solved using a finite element method. Two steps registration was implemented to determine the tissue-element volume and displacement. The first is the conventional deformable image registration to determine the organ surface. The second is to map the element volume/position in the shrinking organ with using the dose inducible shrinkage model. Treatment dose was constructed by applying both the conventional deformable registration and the shrinkage model on CBCT images obtained from h&n treatment. The dose distribution in the patient parotid was evaluated. RESULTS Treatment dose-volume, V30, constructed including the dose inducible parotid shrinkage was 11% higher than the one from the conventional deformable image registration and dose construction. Tissue elements in the high dose region shrink more than those in a low dose region which results in extra tissue elements in the organ move into the high dose area during the treatment course and an unfavorable dose-volume relationship in the organ. CONCLUSIONS Without including dose inducible organ shrinkage, treatment dose in a shrinking organ could be underestimated. On the other hand, including this effect in the treatment evaluation and adaptive planning optimization will minimize the potential detriment of the effect.