Machine Learning Approaches for Range and Dose Verification in Proton Therapy Using Proton-induced Positron Emitters.

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 CT-based patient phantom. The 1D distributions of positron emitters and radiation dose were obtained. A feedforward neural network classification model comprising 2 hidden layers, was developed to estimate whether the range is within a preset threshold. A recurrent neural network (RNN) regression model comprising 3 layers and 10 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 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 signal-to-noise ratio (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. CONCLUSION 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.

[1]  D Sarrut,et al.  Machine learning-based patient specific prompt-gamma dose monitoring in proton therapy , 2013, Physics in medicine and biology.

[2]  P. Busca,et al.  Prompt gamma imaging with a slit camera for real-time range control in proton therapy , 2012, 2012 IEEE Nuclear Science Symposium and Medical Imaging Conference Record (NSS/MIC).

[3]  D Dauvergne,et al.  Design Guidelines for a Double Scattering Compton Camera for Prompt-$\gamma$ Imaging During Ion Beam Therapy: A Monte Carlo Simulation Study , 2011, IEEE Transactions on Nuclear Science.

[4]  Wolfgang Enghardt,et al.  On the effectiveness of ion range determination from in-beam PET data , 2010, Physics in medicine and biology.

[5]  Jinsong Ouyang,et al.  Monitoring proton radiation therapy with in-room PET imaging , 2011, Physics in medicine and biology.

[6]  Katia Parodi,et al.  Submillimeter ionoacoustic range determination for protons in water at a clinical synchrocyclotron , 2017, Physics in medicine and biology.

[7]  Keiichi Nakagawa,et al.  The development and clinical use of a beam ON-LINE PET system mounted on a rotating gantry port in proton therapy. , 2008, International journal of radiation oncology, biology, physics.

[8]  Wolfgang Enghardt,et al.  Application of a Yield Approach for the Prediction of Positron Emitter Distributions Produced During Therapeutic Carbon-Ion Beam Irradiation , 2016, IEEE Transactions on Nuclear Science.

[9]  J Fan,et al.  Absolute dose reconstruction in proton therapy using PET imaging modality: feasibility study , 2007, Physics in medicine and biology.

[10]  J. Hanley,et al.  The meaning and use of the area under a receiver operating characteristic (ROC) curve. , 1982, Radiology.

[11]  Yong Hyun Chung,et al.  A simulation study of a C-shaped in-beam PET system for dose verification in carbon ion therapy , 2013 .

[12]  Marco Durante,et al.  Speed and accuracy of a beam tracking system for treatment of moving targets with scanned ion beams , 2009, Physics in medicine and biology.

[13]  Katia Parodi,et al.  Patient study of in vivo verification of beam delivery and range, using positron emission tomography and computed tomography imaging after proton therapy. , 2007, International journal of radiation oncology, biology, physics.

[14]  Kevin C Jones,et al.  How proton pulse characteristics influence protoacoustic determination of proton-beam range: simulation studies , 2016, Physics in medicine and biology.

[15]  Christopher Kurz,et al.  Implementation and initial clinical experience of offline PET/CT-based verification of scanned carbon ion treatment. , 2013, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[16]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[17]  Phil Blunsom,et al.  Recurrent Continuous Translation Models , 2013, EMNLP.

[18]  Jatinder Palta,et al.  In vivo verification of proton beam path by using post-treatment PET/CT imaging. , 2009, Medical physics.

[19]  W Enghardt,et al.  On the feasibility of automatic detection of range deviations from in-beam PET data. , 2012, Physics in medicine and biology.

[20]  Jürgen Hesser,et al.  A deconvolution approach for PET-based dose reconstruction in proton radiotherapy , 2011, Physics in medicine and biology.

[21]  H Paganetti,et al.  The reliability of proton-nuclear interaction cross-section data to predict proton-induced PET images in proton therapy , 2011, Physics in medicine and biology.

[22]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[23]  Irène Buvat,et al.  PET-based dose delivery verification in proton therapy: a GATE based simulation study of five PET system designs in clinical conditions , 2013, Physics in medicine and biology.

[24]  Alfred R. Smith,et al.  Vision 20/20: proton therapy. , 2009, Medical physics.

[25]  W. Enghardt,et al.  Direct time-of-flight for quantitative, real-time in-beam PET: a concept and feasibility study , 2007, Physics in medicine and biology.

[26]  I Buvat,et al.  Distributions of secondary particles in proton and carbon-ion therapy: a comparison between GATE/Geant4 and FLUKA Monte Carlo codes , 2013, Physics in medicine and biology.

[27]  Lei Xing,et al.  Range and dose verification in proton therapy using proton-induced positron emitters and recurrent neural networks (RNNs) , 2019, Physics in medicine and biology.

[28]  C A Tobias,et al.  High energy beams of radioactive nuclei and their biomedical applications. , 1981, International journal of radiation oncology, biology, physics.

[29]  F Verhaegen,et al.  Monte Carlo calculations of positron emitter yields in proton radiotherapy , 2012, Physics in medicine and biology.

[30]  H. Paganetti Range uncertainties in proton therapy and the role of Monte Carlo simulations , 2012, Physics in medicine and biology.

[31]  Katia Parodi,et al.  In-beam PET measurements of β+ radioactivity induced by proton beams , 2002 .

[32]  Wolfgang Enghardt,et al.  Comparison of PHITS, GEANT4, and HIBRAC simulations of depth-dependent yields of β+-emitting nuclei during therapeutic particle irradiation to measured data , 2013, Physics in medicine and biology.

[33]  Katia Parodi,et al.  In vivo range verification in particle therapy. , 2018, Medical physics.

[34]  K Parodi,et al.  TH-C-144-06: TPSPET - A TPS-Based Approach for In-Vivo Dose Verification with PET in Proton Therapy. , 2013, Medical physics.

[35]  Hiroshi Uchida,et al.  Dose-volume delivery guided proton therapy using beam on-line PET system. , 2006, Medical physics.

[36]  A P Gibson,et al.  The physics of Cerenkov light production during proton therapy , 2014, Physics in medicine and biology.

[37]  Freek Beekman,et al.  Real-time prompt gamma monitoring in spot-scanning proton therapy using imaging through a knife-edge-shaped slit , 2012, Physics in medicine and biology.

[38]  Hermann Ney,et al.  LSTM Neural Networks for Language Modeling , 2012, INTERSPEECH.

[39]  T Bortfeld,et al.  TPSPET—A TPS-based approach for in vivo dose verification with PET in proton therapy , 2014, Physics in medicine and biology.

[40]  H Paganetti,et al.  Extension and validation of an analytical model for in vivo PET verification of proton therapy—a phantom and clinical study , 2011, Physics in medicine and biology.

[41]  Katia Parodi,et al.  A filtering approach based on Gaussian–powerlaw convolutions for local PET verification of proton radiotherapy , 2006, Physics in medicine and biology.

[42]  K Parodi,et al.  Dose quantification from in-beam positron emission tomography. , 2004, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[43]  Hideyuki Mizuno,et al.  Positron camera for range verification of heavy-ion radiotherapy , 2003 .

[44]  W Enghardt,et al.  From prompt gamma distribution to dose: a novel approach combining an evolutionary algorithm and filtering based on Gaussian-powerlaw convolutions. , 2016, Physics in medicine and biology.

[45]  H Paganetti,et al.  Monte Carlo patient study on the comparison of prompt gamma and PET imaging for range verification in proton therapy , 2011, Physics in medicine and biology.

[46]  Masataka Komori,et al.  High resolution Cerenkov light imaging of induced positron distribution in proton therapy. , 2014, Medical physics.

[47]  S Stute,et al.  GATE V6: a major enhancement of the GATE simulation platform enabling modelling of CT and radiotherapy , 2011, Physics in medicine and biology.

[48]  Antony Lomax,et al.  In vivo proton range verification: a review , 2013, Physics in medicine and biology.

[49]  Hsiao-Ming Lu,et al.  A potential method for in vivo range verification in proton therapy treatment , 2008, Physics in medicine and biology.