Preliminary results in using Deep Learning to emulate BLOB, a nuclear interaction model.

PURPOSE A reliable model to simulate nuclear interactions is fundamental for Ion-therapy. We already showed how BLOB ("Boltzmann-Langevin One Body"), a model developed to simulate heavy ion interactions up to few hundreds of MeV/u, could simulate also 12C reactions in the same energy domain. However, its computation time is too long for any medical application. For this reason we present the possibility of emulating it with a Deep Learning algorithm. METHODS The BLOB final state is a Probability Density Function (PDF) of finding a nucleon in a position of the phase space. We discretised this PDF and trained a Variational Auto-Encoder (VAE) to reproduce such a discrete PDF. As a proof of concept, we developed and trained a VAE to emulate BLOB in simulating the interactions of 12C with 12C at 62 MeV/u. To have more control on the generation, we forced the VAE latent space to be organised with respect to the impact parameter (b) training a classifier of b jointly with the VAE. RESULTS The distributions obtained from the VAE are similar to the input ones and the computation time needed to use the VAE as a generator is negligible. CONCLUSIONS We show that it is possible to use a Deep Learning approach to emulate a model developed to simulate nuclear reactions in the energy range of interest for Ion-therapy. We foresee the implementation of the generation part in C++ and to interface it with the most used Monte Carlo toolkit: Geant4.

[1]  Katia Parodi,et al.  The FLUKA code and its use in hadron therapy , 2008 .

[2]  K Parodi,et al.  Dosimetric accuracy assessment of a treatment plan verification system for scanned proton beam radiotherapy: one-year experimental results and Monte Carlo analysis of the involved uncertainties , 2013, Physics in medicine and biology.

[3]  Joseph Cugnon,et al.  Comparisons of hadrontherapy-relevant data to nuclear interaction codes in the Geant4 toolkit , 2013 .

[4]  K. Perez.  Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment , 2014 .

[5]  D. Rogers.  Fifty years of Monte Carlo simulations for medical physics , 2006, Physics in medicine and biology.

[6]  Tsuyoshi Murata,et al.  {m , 1934, ACML.

[7]  Mario Canadas,et al.  GAMOS: A Geant4-based easy and flexible framework for nuclear medicine applications , 2008, 2008 IEEE Nuclear Science Symposium Conference Record.

[8]  J. Johns,et al.  Fourier Transform Spectroscopy of the B2Σ-X2Σ Transition of BaH , 1985 .

[9]  Ugo Amaldi,et al.  Radiotherapy with beams of carbon ions , 2005 .

[10]  Robert A. Weller,et al.  An algorithm for computing screened Coulomb scattering in Geant4 , 2005 .

[11]  D. Durand.  An event generator for the study of nuclear collisions in the Fermi energy domain (I). Formalism and first applications , 1992 .

[12]  R. Paramatti,et al.  Secondary radiation measurements for particle therapy applications: Charged secondaries produced by 4He and 12C ion beams in a PMMA target at large angle , 2016, 1608.04624.

[13]  Huaiyu Zhu.  On Information and Sufficiency , 1997 .

[14]  Klaus-Robert Müller,et al.  Efficient BackProp , 2012, Neural Networks: Tricks of the Trade.

[15]  A Mairani,et al.  Benchmarking nuclear models of FLUKA and GEANT4 for carbon ion therapy , 2010, Physics in medicine and biology.

[16]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[17]  Maria Grazia Pia,et al.  GEANT4 low energy electromagnetic models for electrons and photons , 1999 .

[18]  V. Ivanchenko,et al.  Recent Developments in Pre-Equilibrium and De-Excitation Models in Geant4 (Selected Papers of the Joint International Conference of Supercomputing in Nuclear Applications and Monte Carlo : SNA + MC 2010) , 2011 .

[19]  Alán Aspuru-Guzik,et al.  Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules , 2016, ACS central science.

[20]  A. Dell'Acqua,et al.  Geant4 - A simulation toolkit , 2003 .

[21]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[22]  Samy Bengio,et al.  Generating Sentences from a Continuous Space , 2015, CoNLL.

[23]  E. Viezzer,et al.  Up to two billion times acceleration of scientific simulations with deep neural architecture search , 2020, ArXiv.

[24]  C. Mancini-Terracciano,et al.  Cluster formation in nuclear reactions from mean-field inhomogeneities , 2018, 1801.07623.

[25]  V Patera,et al.  Carbon fragmentation measurements and validation of the Geant4 nuclear reaction models for hadrontherapy , 2012, Physics in medicine and biology.

[26]  Paul Babyn,et al.  Unsupervised and semi-supervised learning with Categorical Generative Adversarial Networks assisted by Wasserstein distance for dermoscopy image Classification , 2018, ArXiv.

[27]  T. Koi.  New native QMD code in Geant4 , 2010 .

[28]  T. Koi,et al.  Geometry and physics of the Geant4 toolkit for high and medium energy applications , 2009 .

[29]  M. Asai,et al.  Preliminary results coupling "Stochastic Mean Field" and "Boltzmann-Langevin One Body" models with Geant4. , 2019, Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics.

[30]  D. Durand,et al.  Benchmarking geant4 nuclear models for hadron therapy with 95 MeV/nucleon carbon ions , 2013, 1309.1544.

[31]  M Senzacqua,et al.  Fred: a GPU-accelerated fast-Monte Carlo code for rapid treatment plan recalculation in ion beam therapy , 2017, Physics in medicine and biology.

[32]  A. Kraan,et al.  Range Verification Methods in Particle Therapy: Underlying Physics and Monte Carlo Modeling , 2015, Front. Oncol..

[33]  Riccardo Paramatti,et al.  Design of a new tracking device for on-line beam range monitor in carbon therapy. , 2017, Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics.

[34]  K Parodi,et al.  Monte Carlo simulations to support start-up and treatment planning of scanned proton and carbon ion therapy at a synchrotron-based facility , 2012, Physics in medicine and biology.

[35]  George Loudos,et al.  A review of the use and potential of the GATE Monte Carlo simulation code for radiation therapy and dosimetry applications. , 2014, Medical physics.

[36]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[37]  H Paganetti,et al.  TOPAS: an innovative proton Monte Carlo platform for research and clinical applications. , 2012, Medical physics.

[38]  R. Paramatti,et al.  Secondary radiation measurements for particle therapy applications: prompt photons produced by 4He, 12C and 16O ion beams in a PMMA target , 2016, Physics in medicine and biology.

[39]  A. Dotti,et al.  Validation of Geant4 Nuclear Reaction Models for Hadron Therapy and Preliminary Results with BLOB , 2018, IFMBE Proceedings.

[40]  R. Paramatti,et al.  Secondary radiation measurements for particle therapy applications: nuclear fragmentation produced by 4He ion beams in a PMMA target , 2016, Physics in medicine and biology.

[41]  M. Colonna,et al.  Bifurcations in Boltzmann–Langevin one body dynamics for fermionic systems , 2013, 1302.0239.