Proton therapy requires accurate dose calculation for treatment planning to ensure the conformal doses are precisely delivered to the targets. The conversion of CT numbers to material properties is a significant source of uncertainty for dose calculation. The aim is to develop a physics-informed deep learning (PIDL) framework to derive accurate mass density and relative stopping power (RSP) maps from dual-energy computed tomography (DECT) images. The PIDL framework allows deep learning (DL) models to be trained with a physics loss function, which includes a physics model to constrain DL models. Five DL models were implemented including fully connected neural networks (FCNN), dual-FCNN (DFCNN), and three variants of ResNet: ResNet-v1 (RN-v1), ResNet-v2 (RN-v2), and dual-ResNet-v2 (DRN-v2). Two empirical DECT models were implemented to compare with the PIDL method. An artificial neural network (ANN) and the five DL models trained with and without physics loss were explored to evaluate the PIDL framework. DL training data were from CIRS electron density phantom 062M (Computerized Imaging Reference Systems, Inc., Norfolk, VA). The performance of DL models was tested by CIRS adult male, adult female, and 5-year-old child anthropomorphic phantoms. For density map inference, the physics-informed RN-v2 was 3.3%, 2.9% and 1.9% more accurate than ANN for the adult male, adult female, and child phantoms. The physics-informed DRN-v2 was 0.7%, 0.6%, and 0.8% more accurate than DRN-v2 without physics training for the three phantoms, respectfully. The results indicated that physics-informed training could reduce uncertainty when ANN/DL models without physics training were insufficient to capture data structures or derived significant errors. DL models could also achieve better image noise control compared to the empirical DECT parametric mapping methods. The proposed PIDL framework can potentially improve proton range uncertainty by offering accurate material properties conversion from DECT.
FLASH radiotherapy (RT) is a novel technique in which the ultrahigh dose rate (UHDR) (≥40 Gy/s) is delivered to the entire treatment volume. Recent outcomes of in vivo studies show that the UHDR RT has the potential to spare normal tissue without sacrificing tumor control. There is a growing interest in the application of FLASH RT, and the ultrahigh dose irradiation delivery has been achieved by a few experimental and modified linear accelerators. The underlying mechanism of FLASH effect is yet to be fully understood, but the oxygen depletion in normal tissue providing extra protection during FLASH irradiation is a hypothesis that attracts most attention currently. Monte Carlo simulation is playing an important role in FLASH, enabling the understanding of its dosimetry calculations and hardware design. More advanced Monte Carlo simulation tools are under development to fulfill the challenge of reproducing the radiolysis and radiobiology processes in FLASH irradiation. FLASH RT may become one of standard treatment modalities for tumor treatment in the future. This paper presents the history and status of FLASH RT studies with a focus on FLASH irradiation delivery modalities, underlying mechanism of FLASH effect, in vivo and vitro experiments, and simulation studies. Existing challenges and prospects of this novel technique are discussed in this manuscript.
Objective: Computed tomography (CT) to material property conversion dominates proton range uncertainty, impacting the quality of proton treatment planning. Physics-based and machine learning-based methods have been investigated to leverage dual-energy CT (DECT) to predict proton ranges. Recent development includes physics-informed deep learning (DL) for material property inference. This paper aims to develop a framework to validate Monte Carlo dose calculation (MCDC) using CT-based material characterization models. Approach: The proposed framework includes two experiments to validate in vivo dose and water equivalent thickness (WET) distributions using anthropomorphic and porcine phantoms. Phantoms were irradiated using anteroposterior proton beams, and the exit doses and residual ranges were measured by MatriXX PT and multi-layer strip ionization chamber. Two pre-trained conventional and physics-informed residual networks (RN/PRN) were used for mass density inference from DECT. Additional two heuristic material conversion models using single-energy CT (SECT) and DECT were implemented for comparisons. The gamma index was used for dose comparisons with criteria of 3%/3mm (10% dose threshold). Main results: The phantom study showed that MCDC with PRN achieved mean gamma passing rates of 95.9% and 97.8% for the anthropomorphic and porcine phantoms. The rates were 86.0% and 79.7% for MCDC with the empirical DECT model. WET analyses indicated that the mean WET variations between measurement and simulation were -1.66 mm, -2.48 mm, and -0.06 mm for MCDC using a Hounsfield look-up table with SECT and empirical and PRN models with DECT. Validation experiments indicated that MCDC with PRN achieved consistent dose and WET distributions with measurement. Significance: The proposed framework can be used to identify the optimal CT-based material characterization model for MCDC to improve proton range uncertainty. The framework can systematically verify the accuracy of proton treatment planning, and it can potentially be implemented in the treatment room to be instrumental in online adaptive treatment planning.
Rapid detection of illicit opium poppy plants using UAV (unmanned aerial vehicle) imagery has become an important means to prevent and combat crimes related to drug cultivation. However, current methods rely on time-consuming visual image interpretation. Here, the You Only Look Once version 3 (YOLOv3) network structure was used to assess the influence that different backbone networks have on the average precision and detection speed of an UAV-derived dataset of poppy imagery, with MobileNetv2 (MN) selected as the most suitable backbone network. A Spatial Pyramid Pooling (SPP) unit was introduced and Generalized Intersection over Union (GIoU) was used to calculate the coordinate loss. The resulting SPP-GIoU-YOLOv3-MN model improved the average precision by 1.62% (from 94.75% to 96.37%) without decreasing speed and achieved an average precision of 96.37%, with a detection speed of 29 FPS using an RTX 2080Ti platform. The sliding window method was used for detection in complete UAV images, which took approximately 2.2 sec/image, approximately 10× faster than visual interpretation. The proposed technique significantly improved the efficiency of poppy detection in UAV images while also maintaining a high detection accuracy. The proposed method is thus suitable for the rapid detection of illicit opium poppy cultivation in residential areas and farmland where UAVs with ordinary visible light cameras can be operated at low altitudes (relative height < 200 m).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.