Search

Background: Positron emission tomography (PET) has been investigated for its ability to reconstruct proton‐induced positron activity distributions in proton therapy. This technique holds potential for range verification in clinical practice. Recently, deep learning‐based dose estimation from positron activity distributions shows promise for in vivo proton dose monitoring and guided proton therapy. Purpose: This study evaluates the effectiveness of three classical neural network models, recurrent neural network (RNN), U‐Net, and Transformer, for proton dose estimating. It also investigates the characteristics of these models, providing valuable insights for selecting the appropriate model in clinical practice. Methods: Proton dose calculations for spot beams were simulated using Geant4. Computed tomography (CT) images from four head cases were utilized, with three for training neural networks and the remaining one for testing. The neural networks were trained with one‐dimensional (1D) positron activity distributions as inputs and generated 1D dose distributions as outputs. The impact of the number of training samples on the networks was examined, and their dose prediction performance in both homogeneous brain and heterogeneous nasopharynx sites was evaluated. Additionally, the effect of positron activity distribution uncertainty on dose prediction performance was investigated. To quantitatively evaluate the models, mean relative error (MRE) and absolute range error (ARE) were used as evaluation metrics. Results: The U‐Net exhibited a notable advantage in range verification with a smaller number of training samples, achieving approximately 75% of AREs below 0.5 mm using only 500 training samples. The networks performed better in the homogeneous brain site compared to the heterogeneous nasopharyngeal site. In the homogeneous brain site, all networks exhibited small AREs, with approximately 90% of the AREs below 0.5 mm. The Transformer exhibited the best overall dose distribution prediction, with approximately 92% of MREs below 3%. In the heterogeneous nasopharyngeal site, all networks demonstrated acceptable AREs, with approximately 88% of AREs below 3 mm. The Transformer maintained the best overall dose distribution prediction, with approximately 85% of MREs below 5%. The performance of all three networks in dose prediction declined as the uncertainty of positron activity distribution increased, and the Transformer consistently outperformed the other networks in all cases. Conclusions: Both the U‐Net and the Transformer have certain advantages in the proton dose estimation task. The U‐Net proves well suited for range verification with a small training sample size, while the Transformer outperforms others at dose‐guided proton therapy. [ABSTRACT FROM AUTHOR]

Diatomaceous earth has been used as a nonorganic carrier to prepare a sodium humate modified diatomaceous space using the sol‐gel method. X‐ray diffraction, field emission scanning electron microscopy, transmission electron microscope, the Brunauer–Emmett–Teller (BET) method, and other methods were used to characterize its physio‐chemical properties. The effects of sodium humate/diatomite on the pH of tetracycline in water, the dosage of adsorbent, and pollutant temperature on the adsorption effect were studied. The results show that after alkali modification, the porous structure of diatomite remains undamaged, and the specific surface area of the modified sample is increased to 12.712 m2/g, and the average pore diameter is reduced to 146.19 nm. Adding 20‐mg sodium humate/diatomite into 100‐mL tetracycline wastewater with an initial concentration of 20 mg/L, and pH of 5 at a temperature of 55°C, the adsorption capacity reaches 75.1 mg/L, and the removal rate reaches 78.3%. The quasi‐second‐order kinetic model was fitted by regression and found to have R2 > 0.99, indicating that the adsorption process was one of surface adsorption. It is a process that involves initially rapid adsorption and then slowly reaching equilibrium. As an adsorption thermodynamics model, the Temkin, Redlich‐Peterson and Koble‐Corrigan models can explain humic acid sodium/diatomite behavior in the adsorption of tetracycline in water (each with R2 ≥ 0.99), and the effects of surface behavior in multilayer adsorption and electrostatic attraction on uneven heterogeneous adsorption. [ABSTRACT FROM AUTHOR]

Objective . In-beam positron emission tomography (PET) has important development prospects in real-time monitoring of proton therapy. However, in the beam-on operation, the high bursts of radiation events pose challenges to the performance of the PET system. Approach . In this study, we developed a dual-head in-beam PET system for proton therapy monitoring and evaluated its performance. The system has two PET detection heads, each with6×3Plug&Imaging (PnI) detection units. Each PnI unit consists of6×6lutetium-yttrium oxyorthosilicate crystal arrays. The size of each crystal strip is3.95×3.95×20 mm 3 , which is one-to-one coupled with a silicon photomultiplier. The overall size of the head is15.3×7.65 cm 2 . Main results . The in-beam PET system achieved a single count rate of 48 Mcps at the activity of 144.9 MBq, an absolute sensitivity of 2.717%, and a spatial resolution of approximately 2.6 mm (full width at half maximum) at the center of the field-of-view. When imaging a Derenzo phantom, the system could resolve rods with a diameter of 2.0 mm. Time-dynamic [ 18 F]-Fluorodeoxyglucose mouse imaging was performed, demonstrating the metabolic processes in the mouse. This shows that the in-beam PET system has the potential for biology-guided proton therapy. The in-beam PET system was used to monitor the range of a 130 MeV proton beam irradiating a polymethyl methacrylate (PMMA) phantom, with a beam intensity of6.0×109p s -1 and an irradiation duration of one minute. PET data were acquired only during the one-minute irradiation. We simulated the range shift by moving the PMMA and adding an air gap, showing that the error between the actual and the measured range is less than 1 mm. Significance . The results demonstrate that the system has a high count rate and the capability of range monitoring in beam-on operation, which is beneficial for achieving real-time range verification of proton beams in the future., (© 2025 Institute of Physics and Engineering in Medicine. All rights, including for text and data mining, AI training, and similar technologies, are reserved.)