Premium
TH‐C‐144‐06: TPS PET ‐ A TPS‐Based Approach for In‐Vivo Dose Verification with PET in Proton Therapy
Author(s) -
Frey K,
Bauer J,
Unholtz D,
Kurz C,
Kraemer M,
Bortfeld T,
Parodi K
Publication year - 2013
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4815800
Subject(s) - imaging phantom , radiation treatment planning , proton therapy , positron emission tomography , monte carlo method , nuclear medicine , positron , dosimetry , computer science , proton , physics , radiation therapy , mathematics , medicine , nuclear physics , statistics , electron
Purpose: PET (positron‐emission tomography)‐based treatment verification uses the irradiation‐induced tissue activation inside the patient. The comparison between measured PET data and corresponding predictions, obtained under the assumption of a correct treatment delivery, is clinically used to validate the proton beam range and the dose application. In contrast to the typically performed time‐consuming Monte Carlo (MC) simulations, TPS PET is introduced as a new approach using the algorithms of the clinical treatment planning system (TPS). The calculation yields β + ‐emitter distributions being fully consistent with the treatment planning dose. Methods: Positron‐emitter basic data are obtained from the convolution of TPS reference depth‐dose profiles in water with reaction‐channel dependent filter functions. This database and fast post‐processing enables the calculation of three‐dimensional β + ‐emitter distributions based on the same algorithms as used for the planning of the treatment dose. Validation of this TPS PET concept is performed in phantom and patient studies against MC simulations and one‐dimensional filtering of three‐dimensional dose distributions. Results: TPS and MC are based on intrinsically different calculation models, which affect dose and positron‐emitter density predictions. Especially in the case of small fields and tissue inhomogeneities, pronounced limitations of the one‐dimensional filtering of three‐dimensional dose distributions are observed, which can be overcome by TPS PET . In the case of extended patient treatment, good agreement is achieved between TPS PET and MC. Conclusion: In contrast to previously introduced methods, TPS PET provides a faster implementation and is not limited by sensitivity to lateral field extensions. The calculations are performed without modifying the TPS pencil beam algorithm, resulting in positron‐emitter predictions, which are consistent to the planned treatment dose. All these aspects confirm TPS PET as a promising alternative to full‐