Monte Carlo Simulation for the Radixact™ Tomotherapy Linac Using EGSnrc

Autor: Danial Seifi Makrani, Hassan Ali Nedaei, Ghazale Geraily, Alireza Khorami-Moghaddam, Nooshin Banaee, Hussam Jassim
Jazyk: angličtina
Rok vydání: 2024
Předmět:
Zdroj: Journal of Medical Physics, Vol 49, Iss 3, Pp 379-386 (2024)
Druh dokumentu: article
ISSN: 0971-6203
1998-3913
DOI: 10.4103/jmp.jmp_29_24
Popis: Purpose: When exact information regarding the treatment head and initial electron beam is available, the Monte Carlo (MC) approach can properly simulate any linear accelerator. However, manufacturers seldom offer information such as the incident electron beam’s energy, radial intensity (spot size), or angular spread. This research aims to forecast these features and verify an MC-simulated linear accelerator model using measurements. Materials and Methods: The BEAMnrc code simulated a 6 MV photon beam from a Radixact™ Tomotherapy Linac. Percentage depth dose and beam profile calculations were conducted using DOSYXZnrc by various electron energies and spot sizes and compared to measurements using a Gamma index with two distinct criterion sets. Furthermore, the fine-tuned electron energy and spot size profiles were created to minimize any disparities using distinct angle spreads. Finally, the output factors (OFs) for various field sizes were compared. Results: The MC model’s fine-tuned electron energy was determined to be 5.8 MeV, with 88.6% of the calculation points passing the 1%/1 mm γ test. A circular radial intensity of 1.4 mm best represented the 6 MV photon beam regarding spot size. Furthermore, a mean angular spread of 0.05 reduced the disparity in cross-field profile between computation and measurement. The most considerable disparities between the MC model OFs and observations were 1.5%. Conclusion: Using the BEAMnrc code, a reliable MC model of the Radixact™ Tomotherapy Linac can be created, as shown in this paper. This model can be used to compute dose distributions with confidence.
Databáze: Directory of Open Access Journals