Ponente
Descripción
Background and aims: In-beam PET offers rapid treatment feedback, yet faces challenges with high event rates. Clinical implementation requires on-the-fly integration of a fast dose reconstruction algorithm. In this work, we present on-the-fly dose reconstruction from clinical in-beam PET data, using a novel In-beam Dose Estimation tool (IDE-PET), capable of obtaining on-line dose and of detecting range deviations.
Methods: The specific PET setup consisted of 6 phoswich detector blocks with 338 pixels each, with 1.55 x 1.55 x LYSO (7mm)+GSO (8mm) detectors. The system was coupled to a fast data acquisition system able to sustain rates up to 10 Msingles/sec.
Several cylindrical (50-mm diameter and 50-mm height) homogeneous PMMA phantoms were irradiated with a monoenergetic proton beam of 70 MeV oriented along the longitudinal axis of the scanner. Additionally, 5 PMMA range shifter foils of varying thickness (from 1 to 5 mm) were also placed at the proximal surface to investigate range shift prediction accuracy.
For real-time dose estimation, we have developed the IDE-PET tool, which combines a GPU-based 3D reconstruction algorithm [2] with a dictionary-based software capable of estimating deposited doses from the 3D PET activity images [3].
Results: The dose estimation algorithm requires from 0.25 to 1.0 seconds to calculate and display the deposited dose. For a 2 Gy dose fraction, the method was able to spot range variations as small as 1 mm. The average range estimation has a statistical error of 0.1 mm (1σ). Assessment of system sensitivity to proton number changes showed satisfactory results for doses as low as 0.2 Gy.
Conclusions: We can reconstruct dose maps from PET activation on-line, at clinically relevant dose levels and during the beam-on period, with an accuracy better than one millimeter, in the BP fall-off region. This validates the feasibility of the proposed experimental setup to be used for in-beam on-the-fly reconstruction of the 3D dose in a clinical scenario.
[1]Parodi, K., et al (2007). Patient study of in vivo verification of beam delivery and range, using positron emission tomography and computed tomography imaging after proton therapy. International Journal of Radiation Oncology Biology Physics, 68(3), 920-934.
[2] Galve, P. et al (2020). GPU based fast and flexible iterative reconstructions of arbitrary and complex PET scanners: application to next generation dedicated brain scanners. In 2020 IEEE (NSS/MIC).
[3] Onecha, V. V., et al, (2022). Dictionary-based software for proton dose reconstruction and submilimetric range verification. Phys. Med. Biol., 67(4), 045002.
