Articles
Publication Date: 2025
Optics Express (10944087)33(10)pp. 21393-21412
We introduce what we believe to be a novel machine learning (ML)-based ResNet algorithm for predicting gas pressure from spectral imagery, eliminating the need for traditional peak fitting. Evaluated using simulated and experimental carbon monoxide (CO) spectra, the model accurately predicts pressures across a wide range (1 mbar - 2 bar), even with noisy data, outperforming conventional methods like PeakFit. The ResNet model demonstrates minimal discrepancies between predicted and actual pressures, achieving a mean absolute error (MAE) of 0.095 and mean squared error (MSE) of 0.009 in simulations, and maximum MAE of 1.2×10−2 and MSE of 1.46×10−4 experimentally below 94 mbar. This approach significantly enhances quantitative spectroscopy by focusing on line shape imagery, showing promising applications in atmospheric science, industrial monitoring, and environmental research. This work is a substantial improvement over our previous models. © 2025 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement.
Karimi, A.H.,
Das, I.J.,
Chegeni, N.,
Jabbari, I.,
Jafari, F.,
Geraily, G. Publication Date: 2024
Scientific Reports (20452322)14(1)
Grid therapy recently has been picking momentum due to favorable outcomes in bulky tumors. This is being termed as Spatially Fractionated Radiation Therapy (SFRT) and lattice therapy. SFRT can be performed with specially designed blocks made with brass or cerrobend with repeated holes or using multi-leaf collimators where dosimetry is uncertain. The dosimetric challenge in grid therapy is the mystery behind the lower percentage depth dose (PDD) in grid fields. The knowledge about the beam quality, indexed by TPR20/10 (Tissue Phantom Ratio), is also necessary for absolute dosimetry of grid fields. Since the grid may change the quality of the primary photons, a new kq,q should be evaluated for absolute dosimetry of grid fields. A Monte Carlo (MC) approach is provided to resolving the dosimetric issues. Using 6 MV beam from a linear accelerator, MC simulation was performed using MCNPX code. Additionally, a commercial grid therapy device was used to simulate the grid fields. Beam parameters were validated with MC model for output factor, depth of maximum dose, PDDs, dose profiles, and TPR20/10. The electron and photon spectra were also compared between open and grid fields. The dmax is the same for open and grid fields. The PDD with grid is lower (~ 10%) than the open field. The difference in TPR20/10 of open and grid fields is observable (~ 5%). Accordingly, TPR20/10 is still a good index for the beam quality in grid fields and consequently choose the correct kq,q in measurements. The output factors for grid fields are 0.2 lower compared to open fields. The lower depth dose with grid therapy is due to lower depth fluence with scatter radiation but it does not impact the dosimetry as the calibration parameters are insensitive to the effective beam energies. Thus, standard dosimetry in open beam based on international protocol could be used. © The Author(s) 2024.
Mahmoudi, F.,
Mohammadi, N.,
Haghighi, M.,
Alirezaei, Z.,
Jabbari, I.,
Chegeni, N.,
Elmtalab, S.,
Vega-carrillo, H.R.,
Kazemian, A.,
Geraily, G. Publication Date: 2023
PLoS ONE (19326203)18(1 January)
Neutron contamination in radiation therapy is of concern in treatment with high-energy photons (> 10 MV). With the development of new radiotherapy modalities such as spatially fractionated grid radiation therapy (SFGRT) or briefly grid radiotherapy, more studies are required to evaluate the risks associated with neutron contamination. In 15 MV SFGRT, high- Z materials such as lead and cerrobend are used as the block on the tray of linear accelerator (linac) which can probably increase the photoneutron production. On the other hand, the high-dose fractions (10-20 Gy) used in SFGRT can induce high neutron contamination. The current study was devoted to addressing these concerns via compression of neutron fluence (Φn) and ambient dose equivalent (H*n 10 ) at the patient table and inside the maze between SFGRT and conventional fractionated radiation therapy (CFRT). The main components of the 15 MV Siemens Primus equipped with different grids and located inside a typical radiotherapy bunker were simulated by the MCNPX® Monte Carlo code. Evidence showed that the material used for grid construction does not significantly increase neutron contamination inside the maze. However, at the end of the maze, neutron contamination in SFGRT is significantly higher than in CFRT. In this regard, a delay time of 15 minutes after SFGRT is recommended for all radiotherapy staff before entering the maze. It can be also concluded that H*n 10 at the patient table is at least 10 times more pronounced than inside the maze. Therefore, the patient is more at risk of neutrons compared to the staff. The H*n 10 at the isocenter in SFGRT with grids made of lead and cerrobend was nearly equal to CFRT. Nevertheless, it was dramatically lower than in CFRT by 30% if the brass grid is used. Accordingly, SFGRT with the brass grid is recommended, from radiation protection aspects. © 2023 Mahmoudi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
