From prompt gamma distribution to dose: a novel approach combining an evolutionary algorithm and filtering based on Gaussian-powerlaw convolutions

Purpose/objective(s) Online proton range/dose verification based on measurements of proton‐induced positron emitters is a promising strategy for quality assurance in proton therapy. Because of the nonlinear correlation between the dose distribution and the activity distribution of positron emitters in addition to the presence of noise, machine learning approaches were proposed to establish their relationship. Materials/methods Simulations were carried out with a spot‐scanning proton system using GATE‐8.0 and Geant4‐10.3 toolkit with a computed tomography (CT)‐based patient phantom. The one‐dimensional (1D) distributions of positron emitters and radiation dose were obtained. A feedforward neural network classification model comprising two hidden layers, was developed to estimate whether the range is within a preset threshold. A recurrent neural network (RNN) regression model comprising three layers and ten neurons in each hidden layer was developed to estimate dose distribution. The performance was quantitatively studied in terms of mean squared error (MSE) and mean absolute error (MAE) under different signal‐to‐noise ratio (SNR) values. Results The feasibility of proton range and dose verification using the proposed neural network framework was demonstrated. The feedforward NN model achieves high classification accuracy close to 100% for individual classes without bias. The RNN model is able to accurately predict the 1D dose distribution for different energies and irradiation positions. When the SNR of the input activity profiles is above 4, the framework is able to predict with an MAE of ~0.60 mm and an MSE of ~0.066. Moreover, the model demonstrates a good capability of generalization. Conclusions The RNN model is found to be effective in identifying the relationship between the distributions of dose and positron emitters. The machine learning‐based framework and RNN models may be a useful tool to allow for accurate online range and dose verification based on proton‐induced positron emitters.

Section: Discussionmentioning

confidence: 99%

Section: B Comparison To Kernel-based Modelsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Technical Note: Machine learning approaches for range and dose verification in proton therapy using proton‐induced positron emitters

Wang

et al. 2019

“…Generally speaking, a verification is performed based on the comparison between a reference image and an image acquired during actual treatment. The reference can be either obtained from previous treatment fractions, 23 or expected PET distribution (using Monte Carlo simulation [26][27][28][29] or analytical modeling [30][31][32][33][34][35] ). The reverse routine of the abovementioned pipeline (i.e., from measured PET to predict/compare with a reference dose) has also been studied, and was found to suffer from limitations such as extensive computational workload, [19][20] being sensitive to counting statistics and image artifacts.…”

Section: Introductionmentioning

confidence: 99%

Feasibility study of patient‐specific dose verification in proton therapy utilizing positron emission tomography (PET) and generative adversarial network (GAN)

et al. 2020

Purpose: Online dose verification based on proton-induced positron emitters is a promising strategy for quality assurance in proton therapy. Because of the nonlinear correlation between dose and the activity distributions, a machine learning-based approach was developed to establish their relationship. Methods: Simulations were carried out using a pencil beam scanning system and a computed tomography (CT) image-based phantom. A DiscoGAN model was developed to perform dose verification for both central and off-center lines. Besides the activity as input, HU information from CT images and stopping power (SP) prior were incorporated as auxiliary features for the model. The performance was quantitatively studied in terms of mean absolute error (MAE) and mean relative error (MRE), under different signal-to-noise ratios (SNRs). In addition to a dataset comprising monoenergetic beams, two additional datasets were generated to evaluate the model's generalization capability: five reconstructed PET images based on an in-beam PET system and a dataset comprising spread-out Bragg peaks (SOBPs). Results: The feasibility of dose verification was successfully demonstrated for all three datasets. For the monoenergetic case (i.e., raw activity of positron emitters), the MRE is found to be <1% for the central lines and 5% for the off-center lines, respectively. The range uncertainty is found to be less than 1 mm. The prediction based on five PET images, which take into account the detection of 511-keV photons and image reconstruction, yields slightly inferior performance. For the SOBP case, the MRE of the center lines is found to be <3% and the range uncertainty is <1 mm. The inclusion of anatomical information (HU and SP) improves both accuracy and generalization of the DiscoGAN model. Conclusion: The combination of proton-induced positron emitters, in-beam PET, and machine learning may become a useful tool allowing for patient-specific online dose verification in proton therapy.

Proton spot dose estimation based on positron activity distributions with neural network

Zhang,

Mu,

et al. 2024

BackgroundPositron 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.PurposeThis 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.MethodsProton 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.ResultsThe 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.ConclusionsBoth 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.