Program Information
Activity-Equivalent Path Length Approach for the 3D PET-Based Dose Reconstruction in Proton Therapy
A Attili1*, A Kraan2,3 , F Dalmasso1,4 , A Vignati1 , S Giordanengo1 , G Battistoni5 , (1) Istituto Nazionale di Fisica Nucleare, Sez. Torino, Torino, Italy , (2) Istituto Nazionale di Fisica Nucleare, Sez. Pisa, Pisa, Italy , (3) Universita degli Studi di Pisa, Pisa, Italy , (4) Universita degli Studi di Torino, Torino, Italy , (5) Istituto Nazionale di Fisica Nucleare, Sez. Milano, Milano, Italy
Presentations
SU-E-J-141 (Sunday, July 12, 2015) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose: Ion beam therapy is sensitive to uncertainties from treatment planning and dose delivery. PET imaging of induced positron emitter distributions is a practical approach for in vivo, in situ verification of ion beam treatments. Treatment verification is usually done by comparing measured activity distributions with reference distributions, evaluated in nominal conditions. Although such comparisons give valuable information on treatment quality, a proper clinical evaluation of the treatment ultimately relies on the knowledge of the actual delivered dose. Analytical deconvolution methods relating activity and dose have been studied in this context, but were not clinically applied. In this work we present a feasibility study of an alternative approach for dose reconstruction from activity data, which is based on relating variations in accumulated activity to tissue density variations.
Methods: First, reference distributions of dose and activity were calculated from the treatment plan and CT data. Then, the actual measured activity data were cumulatively matched with the reference activity distributions to obtain a set of activity-equivalent path lengths (AEPLs) along the rays of the pencil beams. Finally, these AEPLs were used to deform the original dose distribution, yielding the actual delivered dose. The method was tested by simulating a proton therapy treatment plan delivering 2 Gy on a homogeneous water phantom (the reference), which was compared with the same plan delivered on a phantom containing inhomogeneities. Activity and dose distributions were were calculated by means of the FLUKA Monte Carlo toolkit.
Results: The main features of the observed dose distribution in the inhomogeneous situation were reproduced using the AEPL approach. Variations in particle range were reproduced and the positions, where these deviations originated, were properly identified.
Conclusions: For a simple inhomogeneous phantom the 3D dose reconstruction from PET-activity induced by proton beams was shown to be feasible.
Contact Email: